DE102021005196B4 - Restructuring of digital images by adapting to point-based network structures - Google Patents

Abstract

Computer-implemented method for generating and restructuring a digital image, characterized in that - a point-based, adjustable, clearly reconstructable network structure is generated, - the individual fields of which gradually grow or shrink outwards from a central area of the network structure in such a way that the entire area is covered by those fields without there being any overlaps or areas not taken into account - the network structure in question is then placed over a digital image (hereinafter also referred to as the “original image material”), with the center point the network structure can be located somewhere on the same level, inside or outside the image, - then the respective color of a field of this network structure by calculation, for example based on the Gaussian filter model, or selection from the color information of the original image section immediately below, is determined and- the resulting image is saved in the form of a vector-based graphic,- the underlying original image material is created by taking several photographs in a row manually or using a program,- while the distance to the object or the The optical zoom factor is changed - for the purpose of using the method, either the photographs with the higher optical zoom factor are superimposed/hidden on the photographs with the lower zoom factor, or all the color information contained in the photographs created in this way is evaluated .

Description

Technical area

The invention is a development in the field of computer science, i.e. it is a computer-implemented invention.

Within this technical field, the invention can be specified in the area of image and video editing.

Prior art known to the applicant

• - (a) Das Weichzeichnenim Sinne einer bewussten Veränderung eines Fotos, wodurch Zeichnung oder Kontrast eines Bildes reduziert werden, wobei ein bestimmter Bereich oder bspw. ein Objekt innerhalb eines Bildes zur Bearbeitung ausgewählt wird. Ich nehme hier auch Bezug auf den „Gaußschen Weichzeichner“, welcher eine mathematische Funktion auf ein Bild anwendet, welche sich darstellt, als würde man ein wenig lichtdurchlässiges Material über ein Foto halten. Der „Gaußsche Weichzeichner“ ist geläufig und kann überall recherchiert werden.
• - (b) Tiefenkarten, durch welche die bereits natürlich vorhandene Tiefenschärfe bzw. Tiefenunschärfe einer Fotografie stufenlos nachjustiert werden kann
• - (c) Vektorisieren von Rasterbildgrafiken im Sinne eines automatisierten Prozesses der Konvertierung von Rastergrafiken in Vektorgrafiken, nach bestimmten Algorithmen, welche die Bildpunkte einer Rastergrafik abtasten und ggf. Linien oder Formen erstellen, wodurch das Bild frei skalierbar wird, ohne unscharf/verpixelt zu erscheinen
• - (d) Verlustfreies Konvertieren einer Rastergrafik in eine Vektorgrafik, indem die Lage, Ausdehnung, sowie der Farbwert jedes einzelnen Bildpunktes (Feldes) in Form eines Blockes festgehalten wird, sodass das Ergebnis einer Konvertierung dem Ursprungsbild rein optisch zu 100% gleicht
• - (e) Konvertieren einer Rastergrafik in eine Vektorgrafik nach der unter (d) beschriebenen Weise, dadurch gekennzeichnet, dass
  • - die Auflösung (im Sinne von Anzahl der Pixel) einzelner Bereiche eines Bildes reduziert bzw. erhöht wird, möglich in den meisten geläufigen Bildbearbeitungsprogrammen
  • - wiederholen dieser Aktion, bis ein Bild in einem gewissen Bereich besonders gut aufgelöst wird, während die Auflösung zu den Rändern hin langsam abnimmt.
• - (f) Lokale Verfeinerung von zweidimensionalen und dreidimensionalen Objekten.
• - (g) Finite-Elemente Generatoren, welche im zwei- und dreidimensionalen Bereich Volumenkörperelemente, Schalenelemente oder Linienelemente erzeugen und auf diese Weise eine Darstellung bzw. ein Objekt in eine endliche Anzahl von kleineren Elementen aufteilen.
• - (h) Verfahren zum Kodieren von Bilddateien, die eine geografische Linie repräsentieren DE 10 2020 122 052 B3 offenbart ein Verfahren zum Kodieren von Bilddaten. Dazu wird vorgeschlagen, eine Bildhierarchie mit unterschiedlichen Auflösungen zu bilden, um damit eine komprimierte Darstellung der Bildinhalte zu erreichen, s. Abs. [0021]-[0025]. Dazu wird eine Struktur mit Feldern gebildet, die das gesamte Bild partitioniert, s. 4. Diese Struktur wird über das gesamte Bild gelegt. Die Struktur selbst wird dabei abhängig von erkannten Bildinhalten wie Straßen über dem Bild zentriert, wobei die Auflösung mit dem Abstand von der erkannten Straße abnimmt, das heißt die Größe der Felder zunimmt, s. 4 und Patentanspruch 1. Die Farbinformation der Felder wird in der Lehre der Druckschrift D1 durch Bildung des Mittelwerts oder Medians des unter dem Feld liegenden Bildausschnitts festgelegt, s. Abs. [0086]. Der Fachmann entnimmt also, dass Bildinformation mit geeigneten Netz- oder Gitterstrukturen abgetastet werden kann, um Speicherplatz zu sparen. Er entnimmt auch, dass diese Strukturen abhängig von den Bildinhalten gewählt werden sollten, und dass Felder mit dem Abstand von detaillierten Bereichen ausgehend anwachsen können.

The following is known from the prior art:

• - (a) Blurring, in the sense of consciously altering a photo, thereby reducing the drawing or contrast of an image, whereby a specific area or, for example, an object within an image is selected for editing. I also refer here to the “Gaussian Blur,” which applies a mathematical function to an image that is like holding a slightly translucent material over a photo. The “Gaussian blur” is common and can be researched anywhere.
• - (b) Depth maps, through which the already naturally existing depth of field or depth of field of a photograph can be continuously readjusted
• - (c) Vectorization of raster image graphics in the sense of an automated process of converting raster graphics into vector graphics, according to certain algorithms that scan the pixels of a raster graphic and, if necessary, create lines or shapes, making the image freely scalable without appearing blurry/pixelated
• - (d) Lossless conversion of a raster graphic into a vector graphic by recording the position, extent and color value of each individual image point (field) in the form of a block, so that the result of a conversion is 100% optically similar to the original image
• - (e) converting a raster graphic into a vector graphic in the manner described under (d), characterized in that
  • - the resolution (in the sense of number of pixels) of individual areas of an image is reduced or increased, possible in most common image editing programs
  • - repeat this action until an image is particularly well resolved in a certain area, while the resolution slowly decreases towards the edges.
• - (f) Local refinement of two-dimensional and three-dimensional objects.
• - (g) Finite element generators, which create solid elements, shell elements or line elements in the two- and three-dimensional area and in this way divide a representation or an object into a finite number of smaller elements.
• - (h) Method for encoding image files representing a geographical line DE 10 2020 122 052 B3 discloses a method for encoding image data. For this purpose, it is proposed to form an image hierarchy with different resolutions in order to achieve a compressed representation of the image content, see paragraphs [0021]-[0025]. To do this, a structure with fields is created that partitions the entire image, see. 4 . This structure is placed over the entire image. The structure itself is centered over the image depending on recognized image content such as streets, whereby the resolution decreases with the distance from the recognized street, i.e. the size of the fields increases, see. 4 and claim 1. The color information of the fields is determined in the teaching of publication D1 by forming the mean or median of the image section lying below the field, see paragraph [0086]. The person skilled in the art therefore understands that image information can be scanned with suitable network or grid structures in order to save storage space. He also notes that these structures should be chosen depending on the image content, and that fields can grow with distance from detailed areas.

Presentation of the shortcomings of previously known solutions

• - (a) Durch das Weichzeichnen bestimmter Bereiche eines Bildes wird
  • - der Speicherbedarf für dieses Bild nach geläufigen Komprimierungsverfahren nicht zwangsläufig reduziert.
  • - die Pixel-Rasterstruktur nicht verändert, d.h. die unscharfen Bereiche enthalten dieselbe Menge an Farbinformationen wie die detailliert dargestellten Bereiche.
• - (b) Es handelt sich dabei wie unter (a) lediglich um eine kosmetische Veränderung von Bildmaterial, d.h. dieser Effekt ist ebenfalls nicht dazu gedacht, das Bild zu komprimieren, sodass es weniger Speicher bedarf.
• - (c) Diese Art von Verfahren ist nicht auf jedes Bild anwendbar, Bilder natürlicher Objekte bzw. Fotografien lassen sich nur schwer auf diese Art und Weise repräsentativ vektorisieren. In den meisten Fällen macht dieses Verfahren lediglich bezogen auf Grafiken Sinn, deren Struktur sich durch Einfachheit und eine sehr kleine Farbpalette auszeichnen.
• - (d) Die Dateigröße einer auf diese Weise in ein Vektorformat konvertierten Grafik wäre deutlich größer, als die des Originalbildes. Da es optisch keine Veränderung gibt wäre diese Art der Konvertierung unnötig.
• - (e) Nach diesem Prinzip existiert kein fließender/harmonisch abgestufter Übergang zwischen gut und weniger gut aufgelösten Bereichen. Dazu kommt, dass alles entweder erst manuell eingerichtet werden muss (welche Bereiche sind scharf und in welchem Maße) oder ein sinnvoller, übertragbarer Algorithmus für die Erstellung von vektorbasierten Rastergrafiken innerhalb von vektorbasierten Rastergrafiken entwickelt werden muss.
• - (f) Die Struktur welcher ein auf diese Weise dargestelltes Objekt oder Bild unterliegt, ist unregelmäßig, der Speicherbedarf ist entsprechend hoch. Dazu kommt, dass eine Verfeinerung häufig mehr oder weniger manuell festgelegt werden muss, was entsprechend zeitintensiv sein kann.
• - (h) DE 10 2020 122 052 B3 offenbart ein Verfahren, welches Bildinhalte im Gegensatz zu meinem hier dargestellten Verfahren abhängig von dessen physischen Inhalts auswertet. Auch ist die Netzstruktur, welche über ein Bild gelegt wird, nicht punktbasiert, sondern die Aufmerksamkeit liegt viel mehr zu etwa gleichen Teilen auf der gesamten dargestellten Fläche. Das erfindungsgemäße Verfahren ist im Gegensatz dazu, schon allein bestimmt durch die Fotografien, welche aufgenommen werden, dadurch gekennzeichnet, dass in einem Punkt bzw. einem definierten Zentrum die Auflösung eines Bildes am größten ist, und das unabhängig von spezifischen Bildinhalten.

With regard to the described state of the art, the following deficiencies/problems/gaps can be identified:

• - (a) By blurring certain areas of an image
  • - The memory requirement for this image is not necessarily reduced using common compression methods.
  • - The pixel grid structure is not changed, ie the blurred areas contain the same amount of color information as the detailed areas.
• - (b) As in (a), this is only a cosmetic change to the image material, i.e. this effect is also not intended to compress the image so that it requires less memory.
• - (c) This type of process cannot be applied to every image; images of natural objects or photographs are difficult to representatively vectorize in this way. In most cases, this procedure only makes sense for graphics whose structure is characterized by simplicity and a very small color palette.
• - (d) The file size of a graphic converted to vector format in this way would be significantly larger than that of the original image. Since there is no visual change, this type of conversion would be unnecessary.
• - (e) According to this principle, there is no smooth/harmoniously graded transition between well and less well-resolved areas. In addition, everything either has to be set up manually (which areas are sharp and to what extent) or a sensible, transferable algorithm for creating vector-based raster graphics within vector-based raster graphics has to be developed.
• - (f) The structure of an object or image displayed in this way is irregular and the memory requirement is correspondingly high. In addition, a refinement often has to be determined more or less manually, which can be time-consuming.
• - (H) DE 10 2020 122 052 B3 discloses a method which, in contrast to the method presented here, evaluates image content depending on its physical content. The network structure that is placed over an image is also not point-based, but rather the attention is spread equally across the entire surface shown. In contrast, the method according to the invention, determined solely by the photographs that are taken, is characterized in that the resolution of an image is greatest at a point or a defined center, regardless of specific image content.

Solved technical problem

The method registered here for restructuring an image according to claim 1 is hereinafter referred to as the “invention”. The user of the process registered here is hereinafter referred to as the “processor”. References to the information about the state of the art and the defects are only indicated below by the small letter in brackets, e.g. (a).

• - ein Bild, unerheblich welcher Art und Größe, stark komprimieren, sodass es je nach Art und Parameter der ausgewählten Netzstruktur zu einem deutlich verringerten Speicherbedarf kommt, während die dem Bearbeiter wichtigen Bereiche/Ausschnitte des Bildes von einer Komprimierung komplett (wie unter bestimmten Voraussetzungen im Beispiel der quadratischen Netzstruktur) bzw. größtenteils verschont bleiben, und somit weiterhin die gesamten bzw. beinahe die gesamten Informationen des wichtigen Bereiches innerhalb des ursprünglichen Bildes enthalten.
• - ein sinnvoller, sowie harmonisch abgestufter Übergang zwischen gut und weniger gut aufgelösten Bereichen eines Bildes erzeugen.
• - durch Eingabe von Parametern, deren Zweck und Bedeutung einfach nachzuvollziehen ist, eine neuartige Bildstruktur erzeugen, ohne dass ein weiterer manueller Eingriff in das Bild nötig ist (wie bspw. unter (e))
• - wie unter Anspruch 2 beschrieben, ein neuartiges Dateiformat für digitale Bilder erstellen.

The invention allows

• - heavily compress an image, regardless of type and size, so that depending on the type and parameters of the selected network structure, there is a significantly reduced memory requirement, while the areas/sections of the image that are important to the editor are completely affected by compression (as under certain conditions in Example of the square network structure) or are largely spared, and thus continue to contain all or almost all of the information of the important area within the original image.
• - create a sensible and harmoniously graduated transition between well and less well-resolved areas of an image.
• - create a new image structure by entering parameters whose purpose and meaning is easy to understand, without any further manual intervention in the image being necessary (as for example under (e))
• - as described under claim 2, create a new file format for digital images.

• - der Speicherbedarf eines nach dem Prinzip der Erfindung konvertierten Bildes im Vergleich zu einem nach dem Prinzip der Erfindung erstellten gewöhnlichen Vektorformat je nach Größe der Datei erneut stark reduzieren.
• - das nach dem Prinzip der Erfindung konvertierte Bild unter Verwendung kompatibler Software tendenziell schneller anzeigen, da jenes neuartige Bildformat aus Anspruch 2 simpler strukturiert ist, somit weniger Informationen abgerufen werden müssen.

By creating an invention-specific file format specifically tailored to the method, as described in claim 2, you can

• - The memory requirement of an image converted according to the principle of the invention is again greatly reduced compared to a normal vector format created according to the principle of the invention, depending on the size of the file.
• - the image converted according to the principle of the invention tends to be displayed more quickly using compatible software, since the new image format from claim 2 is structured more simply, so less information has to be retrieved.

Means of solving this problem/technical implementation

This area discusses the considerations regarding the approach and how the computer-implemented invention can be implemented in particular. The focus here is particularly on the 9 and 10 to get expelled.

Definitions

• - Netzstruktur
  • Damit ist ein Nach Anspruch 1 erzeugtes System, bestehend aus Strecken und Bögen, zu verstehen, welches eine Fläche in mehrere Felder aufteilt.
• - Felder/Feld
  • Dieser Begriff steht für den von den Linien einer Netzstruktur eindeutig eingegrenzten Bereich und ersetzt häufig den Begriff Pixel oder Bildpunkt, welcher sich jedoch laut Definition nicht zwangsweise auf in diesem Anwendungsbereich ungebräuchliche Formen wie bspw. Kreisausschnitte übertragen lässt.
• - Ring
  • Der Begriff Ring bezieht sich auf die, sich außerhalb des Zentrums befindliche, Summe der Flächen aller jeweils eindeutig zusammengehörigen (i.d.R. gleich großen) Felder, wobei das erste Feld eines Rings jeweils an dessen letztes Feld anschließt. Ein Ring umschließt das Zentrum einer Netzstruktur und gegebenenfalls weitere, weiter innen liegende Ringe.
  • In der runden Netzstruktur handelt es sich bei einem Ring bspw. um den allgemein bekannten, sogenannten „Kreisring“.
• - Reihe
  • Dieser Begriff ist lediglich auf die quadratische, sowie die dreieckige Netzstruktur anwendbar und ist gleichzusetzen mit der Summe der Fläche der Felder, welche das Zentrum bzw. andere Reihen von einer Seite begrenzen. Dabei gilt, dass alle in einer Linie hintereinander aufgereihten Felder eine Reihe bilden, sodass einige Felder (an den Berührungspunkten der einzelnen Reihen eines Ringes) mehreren Reihen innerhalb eines Rings zugehörig sind.
• - Raster
  • Dieser Begriff beschreibt eine spezielle Art der (wie ich es hier bezeichne) Netzstruktur, dadurch gekennzeichnet, dass diese lediglich aus identischen Quadraten besteht. Diese werden gebildet aus Reihen und Spalten, welche alle denselben Abstand zueinander haben. Die auf diese Weise entstehenden „Felder“ können in diesem speziellen Fall auch als Pixel oder Bildpunkte bezeichnet werden.
• - Rastergrafik
  • Dieser Begriff beschreibt eine Form von computerlesbaren Daten, welche sich durch deren rasterförmige Anordnung von Pixeln kennzeichnet. Ein Beispielformat für eine Rastergrafik wäre das geläufige PNG-Format.
• - Pixel
  • Der Begriff Pixel steht für einen einzelnen Bildpunkt in einer Rastergrafik. Ihm ist genau ein Farbwert zugeordnet. Ggf. habe ich in dieser Beschreibung auch den Begriff Feld für einen Pixel verwendet, um die Gemeinsamkeiten zwischen den zugrundeliegenden Netzstrukturen deutlich zu machen.
• - Speicherbedarf
  • Dieser Begriff beschreibt die benötigte Menge an freiem Speicherplatz auf einem Computer oder sonstigen Informationen verarbeitenden Medium.

Common terms used in the description should be precisely defined again so that there are no misunderstandings.

• - Network structure
  • This means a system created according to claim 1, consisting of stretches and arcs, which divides an area into several fields.
• - Fields/Field
  • This term stands for the area clearly delimited by the lines of a network structure and often replaces the term pixel or image point, which, however, according to the definition, cannot necessarily be transferred to shapes that are not commonly used in this area of application, such as circular sections.
• - Ring
  • The term ring refers to the sum of the areas of all fields that clearly belong together (usually of the same size) outside the center, whereby the first field of a ring is connected to the last field. A ring encloses the center of a network structure and possibly other rings located further inside.
  • In the round network structure, a ring is, for example, the well-known so-called “circular ring”.
• - Row
  • This term is only applicable to the square and triangular network structure and is equivalent to the sum of the area of the fields that delimit the center or other rows from one side. The rule is that all fields lined up one behind the other in a line form a row, so that some fields (at the points of contact between the individual rows of a ring) belong to several rows within a ring.
• - Grid
  • This term describes a special type of network structure (as I call it here), characterized in that it consists only of identical squares. These are made up of rows and columns, all of which are the same distance apart. In this special case, the “fields” created in this way can also be referred to as pixels or image points.
• - Raster graphics
  • This term describes a form of computer-readable data that is characterized by its grid-shaped arrangement of pixels. An example format for a raster graphic would be the common PNG format.
• - Pixels
  • The term pixel stands for a single image point in a raster graphic. It is assigned exactly one color value. If necessary, I have also used the term field for a pixel in this description in order to make clear the similarities between the underlying network structures.
• - Memory requirements
  • This term describes the amount of free storage space required on a computer or other information processing medium.

implementation

The starting point is focusing on an object with a camera, choosing a high degree of optical zoom. When the photo is taken, the lens automatically begins to reduce the optical zoom level as quickly as possible, taking new images in rapid succession. All the resulting images are placed on top of each other, with the images with the higher zoom level in the foreground. In the best case, the calculation of the vector file to be generated according to the method of claim 1, or (practically) the new file format described according to the method of claim 2, can be carried out directly by a program internal to the camera.

The network structure generated during the method according to claim 1 is point-based, that is, it has a clearly determinable center point. It is specified by the “Location” or “Center Point” parameter. In addition, the network structure created using the method can be called “center-based” because, regardless of its type (square, triangular, round, spiral), it has a center that is clearly different from the surrounding “rings”. The process is adjustable, which means there are parameters through which the appearance of the generated network structure, as well as other properties of the process, can be precisely set/adjusted.

The network structure created using the process, consisting of fields that grow or shrink from the center outwards, is placed over a digital image after its creation, the original image material. Each of the fields of the generated network structure then contains clearly identifiable color material, which is evaluated according to specific parameters. The result of this evaluation decides the new color of the relevant field/area.

Finally, the resulting image is saved in a form of vector-based graphics, which can be a common format such as Scalable Vector Graphics (.svg), or, for example, a new format for image files generated according to the specified method according to claim 2.

Approach and considerations

• - Vektorisiert man Rasterbilddateien in dem Bestreben, dass während der bzw. durch die Konvertierung bestimmte Bereiche stärker berücksichtigt werden als andere, erscheint der Einsatz eines Rasters (bzw. wie hier bezeichnet einer Netzstruktur) mit unterschiedlich großen Feldern, welche über das ursprüngliche Bild gelegt wird, sinnvoll.

• - Die durch die Erfindung generierten Netzstrukturen sollten entweder punkt-, zentrums- oder aber streckenbasiert sein, um eine harmonisch stufenweise anwachsende bzw. sinkende Bildauflösung gewährleisten zu können.
• - Eine streckenbasierte Netzstruktur erscheint unpraktisch, da sich entweder (im Falle, dass sich Anfangs- und/oder Endpunkt der Strecke, welche das Zentrum der Netzstruktur bildet, innerhalb des Bereiches des ursprünglichen Bildes befinden) gar nicht oder nur schwer ein harmonischer, gleichmäßige Abstufungen hervorrufender Algorithmus finden lässt, durch welchen die Felder (im Sinne von Bildpunkten, Pixeln) sich im Verlauf der Ausdehnung der Netzstruktur vergrößern oder verkleinern, ohne dass es zu Überschneidungen oder nicht berücksichtigten Bereichen kommt oder (im Falle einer Strecke, deren Anfangs- und Endpunkt sich außerhalb des ursprünglichen Bildes befinden) nur sehr ungenau ein detaillierter bzw. weniger detaillierter Bereich festlegen lässt. Des Weiteren lassen sich nach meiner Einschätzung eine runde oder spiralförmige Netzstruktur überhaupt nicht auf ein streckenbasiertes Zentrum übertragen.
• - Andersartige Netzstrukturen wären zwar vorstellbar, jedoch mit großer Wahrscheinlichkeit schwieriger umzusetzen, als die in den Ansprüchen 4-6 beschriebenen punktbasierten (/zentrumsbasierten) Netzstrukturen.
• - Vorgeschlagene Lösung: Die Entwicklung von punktbasierten Netzstrukturen, dadurch gekennzeichnet, dass
  • - deren Mittelpunkt auch den Mittelpunkt des diesen umgebenden, in der Netzstruktur vorhandenen, Zentrums bildet,
  • - welches auf eine leicht nachvollziehbare Weise in kleinere identische Felder geteilt wird,
  • - welches umgeben wird von Reihen/Ringen aus jeweils identischen (Ausnahme: spiralförmige Netzstruktur) Feldern,
  • - welche im Vergleich zu den anderen Reihen/Ringen und zum Zentrum dieselbe Form aufweisen (bzw. im Falle der runden oder der spiralförmigen Netzstruktur Feldern ähnlicher Form)
  • - welche außerdem in alle Richtungen die gleiche bzw. annähernd gleiche Ausdehnung aufweisen (bspw. die gleichen Seitenlängen der Felder bei der quadratischen sowie der dreieckigen Netzstruktur)
• - Es kann hilfreich und sinnvoll sein, für die Anwendung auf verschiedene Arten von digitalen Bildern (bspw. Fotografien, Skizzen, Gemälden, Landkarten, Satellitenbildern), verschiedene Arten von Netzstrukturen zur Verfügung zu haben, es wäre jedoch unnütz bzw. je nach Geschmack lediglich von ästhetischem Wert, eine große Menge weiterer für das in Anspruch 1 beschriebene Verfahren vorstellbarer, durch Parameter anpassbare, Netzstrukturen zu entwickeln.

To solve the problems described, the following considerations arose:

• - If raster image files are vectorized in an effort to ensure that certain areas are taken into account more than others during or through the conversion, the use of a grid (or as referred to here, a network structure) with fields of different sizes, which is placed over the original image, appears , sensible.

• - The network structures generated by the invention should be either point-, center- or distance-based in order to be able to ensure a harmoniously gradually increasing or decreasing image resolution.
• - A route-based network structure appears impractical because there is either no or only difficulty in achieving harmonious, uniform gradations (in the event that the start and/or end point of the route, which forms the center of the network structure, is within the area of the original image). The resulting algorithm can be found through which the fields (in the sense of image points, pixels) increase or decrease as the network structure expands, without there being any overlaps or areas not taken into account or (in the case of a route, its start and end points). are outside the original image) only very slightly a more detailed or less detailed area can be defined. Furthermore, in my opinion, a round or spiral network structure cannot be transferred to a route-based center at all.
• - Different network structures would be conceivable, but would most likely be more difficult to implement than the point-based (/center-based) network structures described in claims 4-6.
• - Proposed solution: The development of point-based network structures, characterized in that
  • - the center of which also forms the center of the center surrounding it in the network structure,
  • - which is divided into smaller identical fields in an easy to understand way,
  • - which is surrounded by rows/rings of identical (exception: spiral network structure) fields,
  • - which have the same shape compared to the other rows/rings and to the center (or fields of a similar shape in the case of the round or spiral network structure)
  • - which also have the same or approximately the same extent in all directions (e.g. the same side lengths of the fields in the square and triangular network structure)
• - It can be helpful and useful to have different types of network structures available for application to different types of digital images (e.g. photographs, sketches, paintings, maps, satellite images), but it would be useless or merely depending on taste of aesthetic value, to develop a large number of other network structures that are conceivable for the method described in claim 1 and can be adjusted by parameters.

This rough definition led to four network structures with their associated parameters, which can be found under “Structural implementation”.

Structural implementation

In this area, the network structures described in claims 4-6, the spiral network structure not mentioned in the patent claims, and associated conceivable parameters are discussed.

General parameters

In this context, we must first point out seven [(a)-(g)] parameters that are generally applicable to all network structures. The exact implementation of the parameters described here and below in the respective network structures are to be understood as possible, not mandatory, implementation. It is intended to be illustrated by the fact that they belong to the method according to claim 1 per se or to the network structures described in claims 4-6, as well as the spiral network structure. In addition, by determining the meaning of the respective input value of a parameter, the principle of how the data is created can be determined 1 - 7 understand better.

- (a) The extent

• - Ausdehnung = 0: Lediglich das Zentrum einer Netzstruktur wird dargestellt.
• - Ausdehnung = 10: Das Zentrum, sowie 10 umgebende Ringe werden dargestellt
• - Ausdehnung = 0.5: Das Zentrum, sowie ein exakt in der Mitte geteilter Ring wird dargestellt.

The network structures created using the method described in claim 1 can, in theory, be expanded outwardly to infinity. However, once the network reaches a certain extent, it is conceivable to change certain other parameters or to apply the method described in claim 1 only to part of the original image. The extent can, for example, be specified in the form of a positive rational number, which indicates how many whole or sketched rows/rings should be created around the center of the respective network structure. For the 1 - 7 applies:

• - Extent = 0: Only the center of a network structure is displayed.
• - Extent = 10: The center and 10 surrounding rings are shown
• - Extent = 0.5: The center and a ring divided exactly in the middle are shown.

- (b) The coloring

• - Eine zufällige im jeweiligen Gitterfeld vorhandene Farbe bestimmt die Farbe des kompletten Feldes
• - Die innerhalb der Grenzen des Feldes am häufigsten auftretende Farbe bestimmt die Farbe des kompletten Feldes
• - Der sich nach geläufigen Methoden ergebende „Durchschnitt“ des innerhalb des jeweiligen Feldes befindlichen Farbmaterials bestimmt die Farbe des kompletten Feldes.

The four network structures are created by a program/computer according to an algorithm and certain variable parameters and placed over an image file. Clearly identifiable color material from the original image can be assigned to the individual fields of the network structure created in this way. The fields are recolored according to certain procedures, whereby the binding, selectable algorithm responsible for recoloring the fields of the respective network system also determines the duration that a program implementing the method according to claim 1 takes to convert the original images into those resulting from the Vector graphics resulting from the method according to claim 1 (or the possible graphic file formats described in claim 2 and specifically tailored to the invention) are required. The following options are possible for specifying the color of a field:

• - A random color in the respective grid field determines the color of the entire field
• - The color that occurs most frequently within the boundaries of the field determines the color of the entire field
• - The “average” of the color material within the respective field, resulting from common methods, determines the color of the entire field.

- (c) The size

This parameter stands for the ratio of the pixel size of the original image to be converted to the size of a field in the center of the network structure generated by the method according to claim 1 or the size of the entire center of a network structure generated by the method according to claim 1, whereby it is proposed to have a value of 1 means that the individual fields in the center of the network structure have the same maximum extent as the maximum extent of a pixel in the original image (for example, an identical area could also be crucial). Furthermore, it is proposed to assign a value of 2 for the "Size" parameter the meaning that the individual fields in the center of the network structure generated according to the method according to claim 1 in their maximum extent each correspond to the sum of the maximum extent of 2 pixels from the original image corresponds. Any positive rational number is conceivable for the size of a network structure.

- (d) The location/center

This parameter represents the position of the center of a network structure within the original image. It can be determined, for example, by specifying the horizontal or vertical pixels/coordinates (e.g. x = 1000.5; y = 2000.5 for the center of the 1001st pixel from the left, as well as the 2001st pixel from above or below - depending on the approach ), whereby any rational number is conceivable for both the x-value and the y-value - so in theory the origin can also be outside the original image.

- (e) The rotation

This parameter represents the rotation of the center of the selected network structure in relation to the original image, although a specification in degrees or % seems sensible

- (f) The growth change

This parameter results from another parameter specific to the respective network structures, which I call growth here; information about this can be found in the descriptions of the respective network structures. The parameter indicates how much the growth changes per new ring. The value to be set for the change in growth could be specified in absolute or relative terms (e.g. in %).

- (g) The divisional growth

• - Teilungswachstum = 1: Bei einer Teilung von 0 (im Zentrum, die Teilung bezieht sich jeweils auf die Verhältnisse im Zentrum der Netzstruktur) beträgt die Teilung im ersten umgebenden Ring 1
• - Teilungswachstum = (-2): Bei einer Teilung von 10 beträgt die Teilung im ersten umgebenden Ring 8; im zweiten Ring 6
• - Teilungswachstum = 1,5: Bei einer Teilung von 0 beträgt die Teilung im ersten umgebenden Ring 2 (bei ...,5 wird aufgerundet); im zweiten 3

This parameter results from another parameter specific to the respective network structures, which I call the division here. Information about this can be found in the descriptions of the respective network structures. The parameter indicates how much the pitch changes per new ring. The division growth must be a rational number. The following definition is proposed:

• - Division growth = 1: With a division of 0 (in the center, the division refers to the conditions in the center of the network structure) the division in the first surrounding ring is 1
• - Pitch Growth = (-2): With a pitch of 10, the pitch in the first surrounding ring is 8; in the second ring 6
• - Pitch growth = 1.5: With a pitch of 0, the pitch in the first surrounding ring is 2 (if ....5 is rounded up); in the second 3

A percentage division growth would be conceivable to enable exponential growth.

Network structures

The network structures described in claims 4-6, as well as the spiral network structure that does not appear in the claims, are discussed here.

The square network structure

(a) Description

• - besteht aus einem einwandfrei rekonstruierbaren Prinzip von Strecken bzw. Quadraten.
• - Es handelt sich um eine Zusammensetzung von
  • - (1) einer quadratischen Grundfläche (das Zentrum), unterteilt in eine bestimmte Anzahl (Quadratzahl) von kleineren Quadraten und
  • - (2) einem nach außen hin unendlich erweiterbaren System aus immer größer bzw. kleiner werdenden Quadraten.
• - Dabei gilt, dass sich die Anzahl an Quadraten pro Reihe mit jedem das Zentrum umgebenden Ring von Quadraten je nach Wachstum ändert (statt dass zwei Quadrate pro „Reihe“ hinzu kommen wie bei einer Tabelle oder bspw. einer png Rastergrafik).
• - Zwischen den Quadraten befindet sich kein Zwischenraum, die gesamte Fläche wird bedeckt.
• - Außerdem überschneiden die Quadrate sich nicht.

The square network structure

• - consists of a perfectly reconstructable principle of lines or squares.
• - It is a composition of
  • - (1) a square base (the center), divided into a certain number (square number) of smaller squares and
  • - (2) a system of ever larger or smaller squares that can be infinitely expanded towards the outside.
• - The number of squares per row changes with each ring of squares surrounding the center depending on the growth (instead of two squares per “row” being added as in a table or, for example, a png raster graphic).
• - There is no space between the squares, the entire area is covered.
• - In addition, the squares do not overlap.

(b) Using the parameters within the square network structure:

- The Resolution parameter

• - Auflösung = 10: Das Zentrum dieser quadratischen Netzstruktur besteht aus 10x10 Feldern, d.h. 100 Feldern.
• - Auflösung = 2: Das Zentrum dieser quadratischen Netzstruktur besteht aus 2x2 Feldern, d.h. 4 Feldern.

Based on this network structure, this parameter indicates how many squares a row or column consists of in the center of a network structure. In this case the result would be:

• - Resolution = 10: The center of this square network structure consists of 10x10 fields, ie 100 fields.
• - Resolution = 2: The center of this square network structure consists of 2x2 fields, i.e. 4 fields.

In the square structure, the resolution value must be a positive integer. The entire set of positive rational numbers would be conceivable. However, this would result in fragmented fields and wouldn't make much sense in the first place. It would unnecessarily complicate the algorithm.

- The growth parameter

• - Wachstum = 2: Die Nummer an Quadraten reduziert sich um 2 pro Reihe d.h. Die Anzahl der das Zentrum umgebenden Quadrate bleibt von Reihe zu Reihe unverändert (Bei einer Auflösung = 10 Netzstruktur entspräche die Länge/Breite von 8 Feldern im ersten Ring der Länge/Breite der 10-Felder-Reihe im Zentrum)
• - Wachstum = 3: Die Nummer an Quadraten reduziert sich um 3 pro Reihe d.h. die Anzahl der das Zentrum umgebenden Quadrate reduziert sich von Reihe zu Reihe um 1 (Bei einer Auflösung = 10 Netzstruktur entspräche die Länge/Breite von 7 Feldern im ersten Ring der Länge/Breite der 10-Felder-Reihe im Zentrum). Wachstumswerte >2 bestimmen gleichzeitig die maximale Ausdehnung der Netzstruktur.
• - Wachstum = (-1): Die Nummer an Quadraten erhöht sich um 1 pro Reihe, d.h. die Anzahl der das Zentrum umgebenden Quadrate erhöht sich pro Reihe um 3 (Bei einer Auflösung = 10 Netzstruktur entspräche die Länge/Breite von 11 Pixeln im ersten Ring der Länge/Breite der 10-Pixel-Reihe im Zentrum).

Based on this network structure, this parameter indicates how many squares are added per row, suggesting that a value of 0 represents a common grid of identical squares. In this case the result would be:

• - Growth = 2: The number of squares is reduced by 2 per row, i.e. the number of squares surrounding the center remains unchanged from row to row (With a resolution = 10 network structure, the length/width of 8 fields in the first ring would correspond to the length/ Width of the 10-field row in the center)
• - Growth = 3: The number of squares is reduced by 3 per row, i.e. the number of squares surrounding the center is reduced by 1 from row to row (with a resolution = 10 network structure, the length/width of 7 fields in the first ring would correspond to the Length/width of the 10-field row in the center). Growth values >2 simultaneously determine the maximum expansion of the network structure.
• - Growth = (-1): The number of squares increases by 1 per row, i.e. the number of squares surrounding the center increases by 3 per row (With a resolution = 10 network structure, the length/width would correspond to 11 pixels in the first Ring the length/width of the 10-pixel row in the center).

In the quadratic structure, the growth value must be an integer. (The entire set of rational numbers would be conceivable, but this would result in incomplete/fragmented fields at the end of each row.)

- The parameter division

Based on this network structure, this parameter indicates how many smaller, identical squares/parts the individual fields of the network structure are divided into. It has no influence on the “Size” parameter; the division only occurs after the size ratio of the original image to the network structure has been determined. When determining the division, the conditions in the center of the network structure always apply.

The division value must be a natural number so that the dimensions of the resulting individual parts of the fields are identical, among other things to ensure that there are no ambiguities in the order in which the color values are stored for a new file format described in claim 2.

• - Teilung = 1: Es gibt keine Teilung
• - Teilung = 2: Jedes Quadrat wird in 4 kleinere Quadrate unterteilt (2x2 Felder)
• - Teilung = 5 Jedes Quadrat wird in 25 kleinere Quadrate unterteilt (5x5 Felder)

However, a positive rational (not natural) number is also conceivable, but complicated and primarily not useful. The following specifications are proposed:

• - Division = 1: There is no division
• - Division = 2: Each square is divided into 4 smaller squares (2x2 fields)
• - Division = 5 Each square is divided into 25 smaller squares (5x5 fields)

(c) Example

According to this system, a center consisting of 10 fields in a square (resolution = 10), i.e. 100 small squares, could be surrounded by rows of 11 squares each (growth = 1). The sum of the edge lengths of 9 of these 11 squares would then correspond to the length of the sum of the edge lengths of the 10 squares in a row from the center. Overall, however, this 10x10 square center is only surrounded by 40 squares (since the first square in a row is also the last in one of the other 3 surrounding rows). Following this special algorithm, the next 4 surrounding rows consist of 12 squares each, forming a square ring of 44 squares.

• „Die quadratische Netzstruktur; Auflösung = 10; Wachstum = 1; Wachstumsänderung = 0; Ausdehnung = 9; Drehung = 0; Teilung = 1; Teilungswachstum = 0“

It should be here 1 to get expelled:

• “The square network structure; resolution = 10; growth = 1; growth change = 0; extent = 9; rotation = 0; division = 1; Division growth = 0"

The triangular network structure

(a) Description

• - besteht aus einem einwandfrei rekonstruierbaren Prinzip von Linien bzw. gleichseitigen Dreiecken.
• - Es handelt sich um eine Zusammensetzung von
  • - (1) einer Grundfläche in Form eines gleichseitigen Dreiecks (das Zentrum), unterteilt in eine bestimmte Anzahl von kleineren Dreiecken und
  • - (2) einem nach außen hin unendlich erweiterbaren System aus immer größer werdenden gleichseitigen Dreiecken.
• - Dabei gilt, dass sich die Anzahl an Dreiecken pro Reihe mit jedem das Zentrum umgebenden Ring von Dreiecken je nach Wachstum ändert.
• - Zwischen den Dreiecken befindet sich kein Zwischenraum, die gesamte Fläche wird bedeckt.
• - Außerdem überschneiden die Dreiecke sich nicht.

The triangular network structure

• - consists of a perfectly reconstructable principle of lines or equilateral triangles.
• - It is a composition of
  • - (1) a base in the form of an equilateral triangle (the center), divided into a certain number of smaller triangles and
  • - (2) a system of ever-increasing equilateral triangles that can be infinitely expanded towards the outside.
• - The number of triangles per row changes with each ring of triangles surrounding the center depending on the growth.
• - There is no space between the triangles, the entire area is covered.
• - In addition, the triangles do not overlap.

(b) Using the parameters within the triangular network structure

- The Resolution parameter

• - Auflösung = 10: Das Zentrum dieser dreieckigen Netzstruktur besteht aus 10x10 Feldern, d.h. 100 Feldern.
• - Auflösung = 2: Das Zentrum dieser dreieckigen Netzstruktur besteht aus 2x2 Feldern, d.h. 4 Feldern.

Based on this network structure, this parameter indicates how many triangles in a square the center of a network structure consists of. In this case the result would be:

• - Resolution = 10: The center of this triangular network structure consists of 10x10 fields, i.e. 100 fields.
• - Resolution = 2: The center of this triangular network structure consists of 2x2 fields, i.e. 4 fields.

In the triangular structure, the resolution value must be a positive integer. The entire set of positive rational numbers would be conceivable. However, this would result in fragmented fields and wouldn't make much sense in the first place. It would unnecessarily complicate the algorithm.

- The growth parameter

Based on this network structure, this parameter indicates how many triangles are added per row, whereby it is suggested that a value of 0 stands for a network structure consisting of identical triangles.

In the triangular structure, the growth value must be an integer (the entire set of rational numbers would be conceivable, but this would result in incomplete/fragmented fields at the end of each row).

- The parameter division

Based on this network structure, this parameter indicates how many smaller, identical triangles/parts the individual fields of the network structure are divided into. It has no influence on the “Size” parameter; the division only occurs after the size ratio of the original image to the network structure has been determined.

When determining the division, the conditions in the center of the network structure always apply.

The division value must be a natural number so that the dimensions of the resulting individual parts of the fields are identical, among other things to ensure that there are no ambiguities in the order in which the color values are stored for a new file format described in claim 2.

• - Teilung = 1: Es gibt keine Teilung
• - Teilung = 2: Jedes Dreieck wird in 4 kleinere identische Dreiecke unterteilt
• - Teilung = 5: Jedes Dreieck wird in 25 kleinere identische Dreiecke unterteilt

However, a positive rational (non-natural) number is also conceivable; the coloring method would have to be adjusted accordingly. The following specifications are proposed:

• - Division = 1: There is no division
• - Division = 2: Each triangle is divided into 4 smaller identical triangles
• - Division = 5: Each triangle is divided into 25 smaller identical triangles

(c) Example

According to this system, a center with the external dimensions of 10, i.e. 100 small triangles, could be surrounded by rows of 23 interlocking triangles, which would result in a growth value of 1 if one follows the procedure described under “Using the parameters within the triangular network structure”. Principle for assigning values for growth follows. The sum of the side lengths of 9 of these 23 triangles would then correspond to the sum of the side lengths of 10 triangles from the center. In total, however, this center, which has 100 triangles, is only surrounded by 63 triangles (since the first two triangles in a row are also the last ones in one of the other two surrounding rows). det). According to this principle, the next 3 surrounding rows consist of 27 triangles each, forming an equilateral “ring” of 75 triangles.

• „Die dreieckige Netzstruktur; Auflösung = 10; Wachstum = 1; Wachstumsänderung = 0; Ausdehnung = 6; Drehung = 0; Teilung = 1; Teilungswachstum = 0“

It should be here 2 to get expelled:

• “The triangular network structure; resolution = 10; growth = 1; growth change = 0; extent = 6; rotation = 0; division = 1; Division growth = 0"

The round network structure

(a) Description

• - besteht aus einem einwandfrei rekonstruierbaren Prinzip von Kreisen und Diameterabschnitten.
• - Ist zusammengesetzt aus
  • - (1) einem runden Zentrum, welches unterteilt wird in eine bestimmte Anzahl von Kreissegmenten (der Wert Auflösung, welcher weiter unten näher beschrieben wird) und
  • - (2) umgebenden Kreisringen, deren innerer und äußerer Umfang sich aus der Anzahl der Felder innerhalb eines Kreisringes ergibt (da ein innerhalb eines Feldes gemessener Umfangsabschnitt dessen radialer Ausdehnung entspricht)
• - Weiterhin gilt, dass mit jedem neuen Ring eine veränderliche Anzahl von Feldern hinzu kommt (der Wert Wachstum).
• - Es handelt sich um ein nach außen hin unendlich erweiterbares System aus immer größer bzw. kleiner werdenden Kreisen, unterteilt in jeweils gleich große Abschnitte.
• - Das jeweilige innere Umfangssegment eines Feldes (andere Orte zur Ableitung der Länge des jeweiligen Umfangsegmentes sind denkbar bspw. das mittlere/durchschnittliche Umfangssegment) innerhalb eines Kreisringes (nicht innerhalb des Zentrums), entspricht in der Länge der radialen Ausdehnung, sodass die einzelnen Felder, besonders in den äußeren Bereichen der runden Netzstruktur, annähernd quadratisch sind.
• - Zwischen den Segmenten befindet sich kein Zwischenraum, die gesamte Fläche wird bedeckt. Außerdem überschneiden die Segmente sich nicht.

The round network structure

• - consists of a perfectly reconstructable principle of circles and diameter sections.
• - Is composed of
  • - (1) a round center, which is divided into a certain number of circle segments (the value resolution, which is described in more detail below) and
  • - (2) surrounding circular rings, the inner and outer circumference of which results from the number of fields within a circular ring (since a circumferential section measured within a field corresponds to its radial extent)
• - Furthermore, with each new ring a variable number of fields are added (the value growth).
• - It is a system that can be infinitely expanded outwardly and consists of ever-larger or smaller circles, each divided into equally sized sections.
• - The respective inner circumferential segment of a field (other locations for deriving the length of the respective circumferential segment are conceivable, for example the middle/average circumferential segment) within a circular ring (not within the center), corresponds in length to the radial extent, so that the individual fields, especially in the outer areas of the round network structure, are approximately square.
• - There is no space between the segments, the entire surface is covered. In addition, the segments do not overlap.

(b) Using the parameters within the round network structure

- The Resolution parameter

• - Auflösung = 10: Das Zentrum dieser runden Netzstruktur besteht aus 10 Kreissegmenten, d.h. 10 Feldern
• - Auflösung = 2: Das Zentrum dieser runden Netzstruktur besteht aus 2 Kreissegmenten, d.h. 2 Feldern In der runden Netzstruktur muss der Wert für die Auflösung eine positive, ganze Zahl sein, (wobei alle positiven rationalen Zahlen denkbar wären)

Based on this network structure, this parameter indicates how many circle segments the center of a network structure consists of. In this case the result would be:

• - Resolution = 10: The center of this round network structure consists of 10 circle segments, ie 10 fields
• - Resolution = 2: The center of this round network structure consists of 2 circle segments, ie 2 fields. In the round network structure, the value for the resolution must be a positive, integer (whereby all positive rational numbers would be conceivable)

- The growth parameter

This parameter indicates, based on this network structure, how many fields are added per ring, whereby it is suggested that a value of 0 means that the number of fields per ring remains constant.

In the round structure, the growth value must be an integer (the entire set of rational numbers would be conceivable, but this would result in incomplete/fragmentary fields at the end of each ring).

- The parameter division

Based on this network structure, this parameter indicates how many smaller identical or similar parts the individual fields of the network structure are divided into. It applies that each part of a field created in this way has the same radial extent and the limiting sections of the radius are at the same angle to one another.

It has no influence on the “Size” parameter; the division only occurs after the size ratio of the original image to the network structure has been determined.

• - Teilung = 1: Es gibt keine Teilung
• - Teilung = 2: Jedes Feld wird in 4 kleinere Teile unterteilt
• - Teilung = 5: Jedes Feld wird in 25 kleinere Teile unterteilt

When determining the division, the conditions in the center of the network structure always apply. The division value must be a natural number so that the dimensions of the resulting individual parts of the fields are identical, among other things to ensure that there are no ambiguities in the order in which the color values are stored for a new file format described in claim 2. However, a positive rational (not natural) number is also conceivable; the coloring method would have to be adapted accordingly. The following specifications are proposed:

• - Division = 1: There is no division
• - Division = 2: Each field is divided into 4 smaller parts
• - Division = 5: Each field is divided into 25 smaller parts

- The twist parameter

This parameter is only applicable to the round network structure and should be understood more as an aesthetic effect; it indicates how much the respective circular rings are twisted relative to one another. It could be specified in degrees, for example.

In combination with a specific division of the individual fields, the question arises as to whether the already divided rings are twisted relative to one another. Here it is possible to divide the total twist evenly among all divided rings.

- The twist growth parameter

This parameter is also only applicable to the round network structure and should also be understood as an aesthetic effect. It indicates by how many degrees or % (for exponential changes) the twist changes from ring to ring.

• „Die runde Netzstruktur; Auflösung = 4; Wachstum = 4; Wachstumsänderung = 0; Ausdehnung = 8; Drehung = 0; Teilung = 1; Teilungswachstum = 0; Verdrehung = 0; Verdrehungswachstum = 0“

It should be here 3 to get expelled:

• “The round network structure; resolution = 4; growth = 4; growth change = 0; extent = 8; rotation = 0; division = 1; division growth = 0; twist = 0; Twist growth = 0"

• „Darstellung 4: Andeutung der runden Netzstruktur; Auflösung = 4; Wachstum = 1; Wachstumsänderung = 1; Ausdehnung = 5; Drehung = 0; Teilung = 1; Teilungswachstum = 1; Verdrehung = 0; Verdrehungswachstum = 0“

It should be here 4 to get expelled:

• “Illustration 4: Indication of the round network structure; resolution = 4; growth = 1; growth change = 1; extent = 5; rotation = 0; division = 1; division growth = 1; twist = 0; Twist growth = 0"

• „Die runde Netzstruktur; Auflösung = 4; Wachstum = 4; Wachstumsänderung = 0; Ausdehnung = unbekannt; Drehung = 0; Teilung = 4; Teilungswachstum = 0; Verdrehung = 0; Verdrehungswachstum = 0“

It should be here 5 to get expelled:

• “The round network structure; resolution = 4; growth = 4; growth change = 0; extent = unknown; rotation = 0; division = 4; division growth = 0; twist = 0; Twist growth = 0"

• „Die runde Netzstruktur; Auflösung = 4; Wachstum = 0; Wachstumsänderung = 0; Ausdehnung = 4; Drehung = 0; Teilung = 2; Teilungswachstum = 0; Verdrehung = 3 Grad; Verdrehungswachstum = 100%“

It should be here 6 to get expelled:

• “The round network structure; resolution = 4; growth = 0; growth change = 0; extent = 4; rotation = 0; division = 2; division growth = 0; twist = 3 degrees; Twist growth = 100%"

The spiral network structure

This type of network structure is the most difficult to describe and create, as well as the most difficult to generalize, among those presented here.

I would like to point out that the network structure I developed is a fake spiral, which consists of individual arcs strung together. An irrational, real spiral is certainly also conceivable. The principle I used could be transferred to such a situation.

(a) Description

• - besteht aus einem einwandfrei rekonstruierbaren Prinzip von Bögen und „radial einteilenden Strecken“
• - ist zusammengesetzt aus
  • - (1) einem runden Zentrum,
    • - welches durch einen bestimmten Wert (Auflösung) geteilt wird.
    • - Teilt man den Durchmesser, des das Zentrum bildenden Kreises durch diesen Wert und multipliziert ihn mit einem um 1 kleineren Wert, teilt dann dieses Ergebnis wiederum durch 2, so erhält man den Radius des ersten, von dem die äußere Begrenzung des Zentrum bildenden Kreises nach innen verlaufenden Bogen.
    • - Die Mittelpunkte der nach innen führenden Bögen sind jeweils so zu wählen, dass der Abstand zweier benachbarter Bögen konstant bleibt.
    • - Die Hälfte der Differenz des unter dem unter Auflösung genannten Wertes und 2 ergibt dann den Radius des zweiten, weiter nach innen führenden Bogens.
    • - Dieses Prinzip wird nach innen fortgesetzt, bis der Durchmesser eines Bogens dessen Abstand zum nächsten Ring entspricht
    • - Die innerhalb des kreisförmigen Zentrums, auf diese Weise entstandene, Spirale wird in einzelne Felder aufgeteilt, dabei gilt, dass
      • - mit der Teilung von außen begonnen wird, d.h. an dem Punkt, an welchem zwei, einen Ring bildende, Bögen sich von innen aus betrachtet das erste Mal annähern bzw. weiter voneinander entfernen, beginnt das erste Feld.
      • - Dessen radiale Ausdehnung dem Quotienten aus dem Gesamtdurchmesser des Zentrums und dem für die Auflösung festgelegten Wertes entspricht
      • - dessen radiale Ausdehnung (Seitenlängen) der Länge des durch die Mittelpunkte der das jeweilige Feld tangential einteilenden Radiusabschnitte verlaufenden, denselben Ursprung aufweisenden, Kreisbogens entspricht.
      • - Fällt ein Feld auf die durch die Aneinanderreihung zweier Bögen entstandene Knickstelle, so entspricht die Länge des weiter außen in der Spirale liegenden, das jeweilige Feld begrenzenden, Radiusabschnittes der Summe der durch die Mittelpunkte der das jeweilige Feld tangential einteilenden Radiusabschnitte verlaufenden, denselben Ursprung aufweisenden, Kreisbögen. Wobei die Schnittstellen der begrenzenden, sowie der zur Bemessung der tangentialen Ausdehnung herangezogenen, Kreisbögen auf einer Geraden liegen.
      • - Dieses Prinzip wird bis ins innerste der Spirale fortgesetzt.
  • - (2) das Zentrum umgebenden, eine Spirale bildenden, Kreisbögen,
    • - wobei gilt, dass
      • - der Abstand zwischen zwei benachbarten Kreisbögen sich bis zur jeweiligen Änderung des zugrundeliegenden Radius' (bei halbkreisförmigen Kreisbögen nach jeweils 180 Grad) gleichmäßig erhöht/reduziert, wodurch eine Verschiebung des jeweiligen Mittelpunktes des Kreisbogens von Bogen zu Bogen bewirkt wird.
      • - sich der Abstand zwischen zwei benachbarten Kreisbögen mit jedem neuen Kreisbogen (bei halbkreisförmigen Kreisbögen nach jeweils 180 Grad) um einen durch den Parameter Wachstum bezifferbaren Wert verändert, wobei absolutes Wachstum sowie exponentielles Wachstum denkbar sind.
      • - das Prinzip der Einteilung der einzelnen Felder aus dem Zentrum nach außen hin fortgesetzt wird, wobei die jeweils innen liegende, durch einen Radiusabschnitt begrenzte Seite eines Feldes der Länge eines theoretischen, das Feld mittig teilenden Umfangsabschnittes entspricht
• - Es handelt sich um ein nach außen hin unendlich erweiterbares System aus immer größer bzw. kleiner werdenden Kreisbögen.
• - Der begrenzende Radiusabschnitt zwischen dem letzten Feld des Zentrums und dem ersten (von innen heraus betrachtet in seiner Größe abweichenden) Feldes der umgebenden Spirale, nennt sich der Wendepunkt.
• - Zwischen den Segmenten befindet sich kein Zwischenraum, die gesamte Fläche wird bedeckt. Außerdem überschneiden die Segmente sich nicht.

The spiral network structure

• - consists of a perfectly reconstructable principle of arcs and “radially dividing lines”
• - is composed of
  • - (1) a round center,
    • - which is divided by a certain value (resolution).
    • - If you divide the diameter of the circle that forms the center by this value and multiply it by a value that is 1 smaller, then divide this result again by 2, you get the radius of the first circle that forms the outer boundary of the center internal arch.
    • - The center points of the arcs leading inwards must be chosen so that the distance between two adjacent arcs remains constant.
    • - Half of the difference between the value mentioned under resolution and 2 then results in the radius of the second arc that leads further inwards.
    • - This principle is continued inwards until the diameter of an arc corresponds to its distance to the next ring
    • - The spiral created in this way within the circular center is divided into individual fields, the following applies:
      • - the division is started from the outside, ie at the point at which two arcs forming a ring, viewed from the inside, approach each other or move further away from each other for the first time, the first field begins.
      • - Its radial extent corresponds to the quotient of the total diameter of the center and the value specified for the resolution
      • - whose radial extent (side lengths) corresponds to the length of the circular arc running through the centers of the radius sections tangentially dividing the respective field and having the same origin.
      • - If a field falls on the kink created by the juxtaposition of two arcs, the length of the radius section lying further out in the spiral and delimiting the respective field corresponds to the sum of the radius sections running through the centers of the radius sections tangentially dividing the respective field and having the same origin , circular arcs. The interfaces of the limiting circular arcs and those used to measure the tangential extent lie on a straight line.
      • - This principle is continued into the innermost part of the spiral.
  • - (2) circular arcs surrounding the center, forming a spiral,
    • - whereby it applies that
      • - the distance between two adjacent circular arcs increases/reduces uniformly until the underlying radius changes (in the case of semicircular arcs after every 180 degrees), which causes the respective center of the circular arc to shift from arc to arc.
      • - the distance between two adjacent circular arcs changes with each new circular arc (in the case of semicircular circular arcs after every 180 degrees) by a value that can be quantified by the growth parameter, with absolute growth as well as exponential growth being conceivable.
      • - the principle of dividing the individual fields from the center to the outside is continued, with the inner side of a field delimited by a radius section corresponding to the length of a theoretical circumferential section dividing the field in the middle
• - It is a system of circular arcs that are infinitely expandable towards the outside.
• - The limiting radius section between the last field of the center and the first field of the surrounding spiral (which differs in size when viewed from the inside) is called the turning point.
• - There is no space between the segments, the entire surface is covered. In addition, the segments do not overlap.

(b) Using the parameters within the spiral network structure

The arc length value

The principle described here for creating spiral network structures for use in the sense of the method described according to claim 1 uses semicircular arcs as a basis for forming the spiral. It would be conceivable to determine the number of arches of different sizes per entire ring using a parameter that specifies exactly this using a natural number. For example, the value 2 would represent 2 semicircular arcs, the value 4 would represent 4 arcs each covering 90 degrees. A particularly interesting case would be an arc length, which corresponds to the tangential extent of a field.

(c) Example

What is described here is a spiral network structure with a resolution value of 4, an expansion value of 1 (1 x 360 degrees), an arc length value of 2 (i.e. 180 degrees as in the description above) and an exponential growth value of 200%, or 2 (representing doubling - multiplying by 2).

$$\begin{array}{l}-\text{L}\ddot{\text{a}}\text{nge des mittleren Umfangsabschnittes betreffenden Feldes}=\\ \left[\text{Pi}*3\left(\text{Durchmesser d}\text{.betr}\text{.Bogens}\right)\right]:\left[360\left(\text{Grad}\right):\text{Spanne in Grad}\left(\text{siehe oben}\right)\right]\\ \to 1\end{array}$$

For these values, the tangential extent (the length of the circumferential section running through the centers of the radius sections delimiting the field) for the first field directed inwards from the “turning point” is calculated as follows: $$\begin{array}{l}-\text{range between}2\text{fields in degrees}=\\ \left[180\left(\text{Degree}\right)\right]:\left[\text{pi}\left(3.14159\right)*1.5\left(\text{Radius of the "middle" arc}\right)\right]\\ \to 38.19719\end{array}$$

$$\begin{array}{l}-\text{L}\ddot{\text{a}}\text{nge of the middle circumferential section relevant field}=\\ \left[\text{pi}*3\left(\text{diameter d}\text{.re:}\text{.Arch}\right)\right]:\left[360\left(\text{Degree}\right):\text{Range in degrees}\left(\text{see above}\right)\right]\\ \to 1\end{array}$$

This means that the center in this example consists of 11 fields (the innermost field only covers about 24 degrees and is difficult to see)

$$\begin{array}{l}-\text{L}\ddot{\text{a}}\text{nge des mittleren Umfangsabschnittes betreffenden Feldes}=\\ \left[\text{Pi}*\mathrm{4,5}\left(\text{Durchmesser d}\text{.betr}\text{.Bogens}\right)\right]:\left[360\left(\text{Grad}\right):\text{Spanne in Grad}\left(\text{siehe oben}\right)\right]\\ \to 1\end{array}$$

For these values, the tangential extent for the first field directed outwards from the “turning point” is calculated as follows: $$\begin{array}{l}-\text{range between}2\text{fields in degrees}=\\ \left[180\left(\text{Degree}\right)\right]:\left[\text{pi}\left(3.14159\right)*2.25\left(\text{Radius of the "middle" arc}\right)\right]\\ \to 25.46479\end{array}$$

$$\begin{array}{l}-\text{L}\ddot{\text{a}}\text{nge of the middle circumferential section relevant field}=\\ \left[\text{pi}*4.5\left(\text{diameter d}\text{.re:}\text{.Arch}\right)\right]:\left[360\left(\text{Degree}\right):\text{Range in degrees}\left(\text{see above}\right)\right]\\ \to 1\end{array}$$

For each additional field within the same sheet: $$\begin{array}{l}-\text{range between}2\text{fields in degrees}\\ \left\{\left[180\left(\text{Degree}\right)\right]:\left[\text{pi}\left(3.14159\right)*2.25\left(\text{Radius of the "middle" arc}\right)\right]\right\}*\left(\text{L}\ddot{\text{a}}\text{nge of the inner side}\right)\\ \to 27.09708\\ \to 31.81826\\ \to 39.11392\\ \to 46.72789\end{array}$$

$$$$

The next field is at a breaking point, but since the whole thing would only become unnecessarily complicated, the average of the radii of the “middle arcs” is simply used here. The lower the growth, the less important this simplification is: $$\begin{array}{l}-\text{range between}2\text{fields in degrees}=\\ \{\left[180\right]:[3.14159*\left(2.25+3.5\right):2]\}*\left\{1.9951\left(\text{L}\ddot{\text{a}}\text{nge of the inner side}\right)\right\}\\ \to 39.76028\end{array}$$

$$$$

For each additional field within the same sheet the following applies again: $$\begin{array}{l}-\text{range between}2\text{fields in degrees}=\\ \left\{\left[180\left(\text{Degree}\right)\right]:\left[\text{pi}\left(3.14159\right)*2.25\left(\text{Radius of the "middle" arc}\right)\right]\right\}*\left(\text{L}\ddot{\text{a}}\text{nge of the inner side}\right)\\ \to 35.97029\\ \to 46.05271\end{array}$$

This calculation is continued externally.

Note: In this example, the outer arc of a field was used instead of the middle one to select the corresponding center point.

It should be here 7 are referred to, in which exactly this calculation is clarified here.

For the application of this network structure in the manner described in the method according to claim 1, a simplification is proposed so that a computer program does not require an unnecessary amount of time and energy to execute. It would be possible to assign the same tangential extent to all fields within the length of an arc, which then results, for example, from the average distance between two circular arcs. The regulation at the kink points can also be simplified by, for example, using the inner “middle arc” as the basis for calculation.

Useful combinations/special shapes

At this point, reference should be made to patent claim number 9.

• Es wäre sinnvoll, wenn ein nach dem Verfahren erzeugtes Bild unabhängig davon, wie stark man es heranholt (wie stark man zoomt), in der Anzeige immer etwa gleich gut aufgelöst werden könnte, ohne dass sich dabei der Speicherbedarf des nach dem Verfahren in Anspruch 1 erzeugten Bildes unnötig erhöht. Wichtig ist hierbei der Parameter, welcher das Wachstum angibt, das heißt wie viele Felder pro Ring hinzu kommen.

Since common screens are tied to a grid-shaped system of pixels, it is necessary to adapt an image created using the method in claim 1 to their grid for the purpose of display. In relation to the economic application of the method according to claim 1, the following further consideration arises:

• It would make sense if an image created using the method could always be displayed with approximately the same resolution, regardless of how closely you zoom in (how much you zoom), without reducing the memory requirement of the image created using the method in claim 1 generated image is increased unnecessarily. What is important here is the parameter that indicates the growth, i.e. how many fields are added per ring.

A possible definition of a “sufficiently high display resolution” could be that a pixel in the display corresponds to at least the area of an entire field of the respective network structure.

This in turn is the case when the resolution in the area of the image that is least well resolved corresponds to at least one field of the network structure per pixel output/displayed.

It would make sense to choose the value for the “Resolution” parameter so that it corresponds to the desired final display size. If, for example, an image is to be displayed later in the format 500x500 pixels, you should set the resolution to 500 in the case of the square network structure, so that with the largest possible zoom factor, the center of that network structure is displayed 1:1, i.e. without loss. This means that according to the principle of this special form, for example in the case of a square network structure, a growth value of 2 should be used (i.e. the number of squares/fields in a row remains constant over the entire image created in this way).

Types of execution

Method for converting raster image files

The method according to claim 1 makes it possible, for example, to convert a common digital raster image, for example an image in png format (it is also conceivable to convert vector graphics in this way) into a special type of vector graphic, for example a Scalable Vector Graphics (.svg) graphic.

New file format for image files

At this point, reference should be made to patent claim number 2.

Description/Considerations

A normal raster graphic is not saved as a vector file, although in theory it also consists of square vectors and their color values. This is because it is more efficient to simply mark the pixel system (columns and rows form a certain number of identical squares) and to determine the respective color values according to certain enumeration algorithms (e.g. from top to bottom, left to right). This saves a lot of storage space. A separate file format is also conceivable for the method described in claim 1. The same program that converts image files by type of process could temporarily store the information necessary to reconstruct an image and choose a file format internal to the program. This would save a large to very large amount of storage space depending on the size and network structure selected. I would only like to discuss the simplest form of this file format, the one in which the expansion value is large enough to cover the entire original image with the respective network structure.

Necessary image information/properties to be saved

• - Die Art der Netzstruktur (quadratisch, dreieckig, rund, spiralförmig)
• - Alle nötigen Parameter (Ort, Größe, Auflösung, Färbung, Wachstum, Wachstumsänderung, Drehung, Ausdehnung, Teilung, Teilungswachstum, Verdrehung, Verdrehungswachstum, Bogenlänge)
• - Der Algorithmus, in welchem die Reihenfolge der Festlegung der Farbwerte der einzelnen Felder angegeben wird. Dieser ist jeweils einer Netzstruktur zugehörig.

Alle anderen nötigen Informationen zum Anzeigen einer solchen Datei könnten programmintern (innerhalb der Software, welche damit beauftragt wird, entsprechende Datei anzuzeigen/zu erzeugen/zu konvertieren) gespeichert werden.The following information is necessary for the unequivocal reconstruction of an image of this type converted using the method described in claim 1:

• - The type of mesh structure (square, triangular, round, spiral)
• - All necessary parameters (location, size, resolution, coloring, growth, growth change, rotation, expansion, division, division growth, twist, twist growth, arc length)
• - The algorithm in which the order in which the color values of the individual fields are determined is specified. This is assigned to a network structure.

All other information necessary to display such a file could be stored internally in the program (within the software that is tasked with displaying/creating/converting the corresponding file).

Algorithms for the order of determining color values

These are examples of possible algorithms for determining the color values of each individual field of a network structure, each belonging to a specific type of network structure.

- (a) The square network structure

The rows of the center are read/set from left to right, starting with the top one; Then you start with the first surrounding ring. There you first set the color value for the upper left square before continuing clockwise around the center. Then start with the second ring/row. So the image is saved from the inside (from the center) out.

- (b) The round network structure

It starts with the field in the center of the network structure created according to the method according to claim 1, which is located to the right of the vertically upward radius of the center, and then continues in a clockwise direction. When using the division parameter, the correspondingly smaller units/fields are taken into account and are picked up sequentially, first in a clockwise direction, then from the inside out. The process then continues with the second ring, starting again at exactly 90 degrees - and shifting it by the corresponding number of degrees when the Twist parameter is used. So the image is saved from the inside (from the center) out.

- (c) The triangular network structure

The rows of the center are read/set, starting with the top one, from left to right (whereby each row consists of rotated and non-rotated triangles, i.e. with a value of 10 for the resolution parameter, this results in exactly 10 lines), then started with the first surrounding rows/ring. There you first set the color value for the triangle at the tip of the ring before continuing in a clockwise direction. Then start with the second ring/row. So the image is saved from the inside (from the center) out.

- (d) The spiral network structure

The image is saved from the inside (from the center) out.

Further considerations

For further compression, it is conceivable to transfer the compression methods commonly used for raster image graphics to the new image format, for example in the form of run-length coding. This further saving in memory requirements is not possible with ordinary vector graphics such as Scalable Vector Graphics (.svg).

The own file format also offers the possibility of marking a center point. If you open a file saved in this format with a compatible program, the zoom can be aligned to that center and snap into place - i.e. when you zoom in/enlarge an image, the center and center of the associated network structure are automatically brought up, even if it is is not in the center of the image.

This format for image files is characterized by a high degree of individuality and could therefore have a good chance of being accepted by some common image viewing or editing programs over the years, so that it is no longer necessary to use programs specifically designed for these image files to use. Since such image files can provide advantages that the current state of the art does not provide, there should be little to no compatibility problems in the long run. Nevertheless, a generally accepted and widely known vector format for distributing and publicizing images resulting from the method in claim 1 appears to be very useful.

Effect in image editing

As well as pure conversion, the method according to claim 1 can be applied to pure digital image processing. As a pure effect, the process is primarily characterized by the fact that the grid structure underlying a digital raster image remains untouched or comes into effect again after the image processing has been completed.

Appearances in the video sector

All types of execution mentioned here can be transferred to videos. According to the method according to claim 1, videos can be converted into a vector format by applying it to each of the individual images. A separate file type is also conceivable for videos created in this way. (Novel file format according to claim 2, only transferred to videos)

When it comes to editing video material, new parameters seem to make sense, as not every image will/need to be edited individually. For example, functions could be created for changing individual parameters per frame. An interesting, simple effect would be rotating the entire network structure from one frame to the next. Video editing is much more complex and many parameters are conceivable. They should not be listed individually here.

The use of machine learning

Since it may not be worthwhile for certain applications to manually convert individual images using the method according to claim 1 or the method according to claim 2, it could make sense to use so-called “machine learning” and thereby create one or more image files completely automatically to convert. This is the artificial creation of knowledge from experience. Applied to a conversion according to the method of claim 1, there are some sensible approaches.

One possibility would be to teach the software that carries out the “machine learning” what content and information from the original images is important through supervised learning, i.e. the algorithm learns a function from given inputs and outputs. Facial recognition algorithms could be integrated, algorithms that evaluate symbols and lettering or algorithms that identify repetitions (e.g. across several images/photographs).

Application examples

Highlighting objects/people/etc. in pictures

For example, imagine large-format landscape pictures in which two people can be seen somewhere in the background. You can zoom in on the people; the resolution of the image is high enough to display them in high resolution in an enlarged form. However, the rest of the image often does not require the high resolution at all and so, using the method described in claim 1, the size, location and resolution of a detailed area can be determined while harmoniously averting attention from the rest of the image. A lot of storage space is saved (especially with your own file format), a focus is set, a harmonious transition is guaranteed and last but not least: the type of field structure can be determined depending on the area of use (square, triangular, round, spiral).

Connecting location information with local details

If the method according to claim 1 is applied to a detailed aerial photograph, which was taken, for example, by a drone or satellite, a location within the image can be represented in detail, with the surroundings blurring harmoniously and gradually towards the outside. On the one hand, it can be seen where exactly something is located, while on the other hand, you can obtain precise information about that place/object/etc. can retrieve. Here too, as the size of a digital image to be processed increases, you have the advantage of being able to save an enormous amount of storage space. Companies and private individuals can, for example, create and use an image using this process to mark the location of a company or residential building on their website, instead of embedding map data from external servers as is often the case. In combination with a zoom snap function (if you zoom using a controller, the center of a graphic file described according to claim 2 is automatically brought up), this forms a useful alternative to integrating map data.

• Wählt man für den detaillierten Bereich, das Zentrum (Beispielsweise ein Gebäude von 100 Metern x 100 Metern Ausmaße, 1ha), eine Auflösung von 500 x 500 Pixel und möchte man, dass das Gesamtbild einen Bereich von 10'000km x 10'000km abdeckt, erhält man nach dem Verfahren nach Anspruch 1 und einem harmonischem Wert für das Wachstum von 2 (siehe dazu auch Sinnvolle Kombinationen/Sonderform) im Vergleich zu geläufigen Rasterbildgrafiken folgende Werte:
  • - Anzahl Megapixel bei gleichmäßiger Auflösung (25 Pixel pro m²): ca. 250'000'000
  • - Ungefährer Speicherbedarf als 24bit png: ca. 750 TB (Terabyte) - bei 3 byte pro Pixel
  • - Anzahl Megapixel (Anzahl der „Felder“ dividiert durch 1'000'000) nach dem Verfahren nach Anspruch 1: ca. 5,78
  • - Ungefährer Speicherbedarf als 24bit svg: ca. 1006 MB (Megabyte) - bei 174 byte pro Pixel

(dies gilt für unkomprimierte Scalable Vector Grafics, der eigentliche Speicherbedarf ist deutlich geringer)

• - Unverbindliche Vermutung über den Speicherbedarf bei Verwendung eines neuartigen Dateiformates nach Anspruch 2 (ebenfalls 24bit):

ca. 20MB (Megabyte) - in unkomprimierter Form

• - Unverbindliche Vermutung über den Speicherbedarf bei Verwendung eines neuartigen Dateiformates nach Anspruch 2 (8bit - 256 Farben):

ca. 7MB (Megabyte) - in unkomprimierter FormA calculation example was prepared on this topic specifically for the square network structure:

• If you choose a resolution of 500 x 500 pixels for the detailed area, the center (for example a building measuring 100 meters x 100 meters, 1ha), and you want the overall image to cover an area of 10,000km x 10,000km, is obtained by the method according to claim 1 and a harmonic value for growth of 2 (see also Useful combinations/special form) compared to common raster image graphics:
  • - Number of megapixels with uniform resolution (25 pixels per m ² ): approx. 250,000,000
  • - Approximate storage requirement as a 24bit png: approx. 750 TB (terabytes) - at 3 bytes per pixel
  • - Number of megapixels (number of “fields” divided by 1,000,000) according to the method according to claim 1: approx. 5.78
  • - Approximate memory requirement as a 24bit svg: approx. 1006 MB (megabytes) - at 174 bytes per pixel

(this applies to uncompressed Scalable Vector Graphics, the actual memory requirement is significantly lower)

• - Non-binding assumption about the memory requirement when using a new file format according to claim 2 (also 24bit):

approx. 20MB (megabytes) - in uncompressed form

• - Non-binding assumption about the memory requirement when using a new file format according to claim 2 (8bit - 256 colors):

approx. 7MB (megabytes) - in uncompressed form

Individual adaptation of field structures (artistic value)

In addition to conveying the most important information with (almost) the smallest possible storage space consumption and at the same time completely harmonious transitions, the process can also be understood as an art form. For example, it is possible to create circular fields for an image according to an individual algorithm in order to appropriately emphasize a round object, for example.

Obscuring objects/people/etc. in pictures

As already mentioned several times in this document, the process can also resolve certain areas of an image less well, while the resolution continues to increase towards the outside. All you have to do is set the correct values for the various parameters. Using this method, attention can be diverted from certain points and people can be made unrecognizable. Even if images are converted in which areas already exist that are less well-resolved (e.g. a license plate, obscured by reducing the resolution in the relevant area), this can be harmoniously concealed using the method according to claim 1.

Disguise distance/zoom related photo stitching in general

• Es werden mehrere Fotografien eines Motivs erstellt, wobei entweder der Abstand zum Objekt, der (optische) Zoom Faktor der Kamera oder beides verändert wird. Die auf diese Weise entstandenen, sich überlagernden Fotografien, werden anschließend zusammengefügt. Dabei ist auf einen möglichst nahtlosen Übergang zwischen zwei Aufnahmen zu achten. Es gibt bereits viele Methoden, mittels derer sich mehrere digitale Bilder zusammenfügen lassen, es soll hier auf keine besondere Bezug genommen werden. Wichtig ist hierbei auch ein zoom- bzw. entfernungsbedingtes entzerren.

Regardless of the method described in claim 1, the following procedure may be useful and easy to implement in general:

• Several photographs of a subject are created, changing either the distance to the object, the (optical) zoom factor of the camera or both. The overlapping photographs created in this way are then put together. It is important to ensure that the transition between two recordings is as seamless as possible. There are already many methods by which multiple digital images can be stitched together, no specific reference will be made here. It is also important to equalize the zoom or distance.

The method according to claim 1 is then applied. It is suggested to place the center of the respective network structure on the image section with the highest zoom factor, or the image section taken from the lowest distance.

Economic exploitation

• - A computer program (mobile applications or online applications are also possible) which implements/applies the method according to claim 1 or a combination of several or all of the methods, possible applications and products described in the patent claims can be sold or licensed.

• „Mögliche Umsetzung der Benutzeroberfläche eines das Verfahren nach Anspruch 1 ausführende Computerprogramm“.
  • - Die Verwendung des Verfahrens nach Anspruch 1 an sich lässt sich lizenzieren, Herausgeber von Bild- und Videobearbeitungsprogrammen, sowie Herausgeber von Kameras mit optischer Zoom-Funktion könnten interessiert sein.
  • - Durch das Verfahren nach Anspruch 1 entstandene digitale Bilder und Videos, sowie zugehörige Druckprodukte aller Art, lassen sich verkaufen.
  • - Kameras, mit der Option, das Verfahren nach Anspruch 1 automatisch bzw. manuell anzuwenden, lassen sich verkaufen
  • - Die Lizenz zur Verwendung von nach dem Verfahren aus Anspruch 1 entstandenen Bildmaterials kann verkauft werden. Designer, welche durch die Verwendung der in den Ansprüchen 4-6 beschriebenen, der spiralförmigen, oder auch individuellen, Netzstrukturen bestimmte Formen betonen möchten, könnten interessiert sein.

It should be here 8th to get expelled:

• “Possible implementation of the user interface of a computer program executing the method according to claim 1”.
  • - The use of the method according to claim 1 itself can be licensed; publishers of image and video editing programs, as well as publishers of cameras with an optical zoom function, could be interested.
  • - Digital images and videos created by the method according to claim 1, as well as associated printed products of all kinds, can be sold.
  • - Cameras with the option to use the method according to claim 1 automatically or manually can be sold
  • - The license to use image material created using the method of claim 1 can be sold. Designers who would like to emphasize certain shapes through the use of the spiral or individual network structures described in claims 4-6 may be interested.

Advantages achieved by the invention

Some of the advantages of the invention have already been mentioned in the description, especially under “Solved technical problem” you can find information.

• - Die tendenziell unwichtige Umgebung eines Details kann komprimiert werden, ohne die Auflösung des ausgewählten Details zu beeinträchtigen. Daraus kann sich bei Anwendung des Verfahrens nach Anspruch 1, sowie im Besonderen bei Verwendung des in Anspruch 2 beschriebenen neuartigen Dateiformats, je nach Größe des ursprünglichen Bildmaterials, ein enorm verringerter Speicherbedarf ergeben.
• - Ein fließender, harmonischer Übergang zwischen hoch und weniger hoch aufgelösten Bereichen innerhalb ein und desselben Bildes kann gewährleistet werden.
• - Wichtige Details innerhalb eines Bildes können durch eine höhere Auflösung im Vergleich zum Rest des Bildes betont/hervorgehoben werden.
• - Es sind Bilddateien vorstellbar, welche die Vorteile eines weiten Aufnahmewinkels und einer hohen Auflösung im Zentrum des Bildes vereinen. Die Informationen aus mehreren Bildern, welche entweder aus unterschiedlicher Entfernung oder aber durch die Verwendung unterschiedlicher Zoom-Faktoren

entstanden sind, werden zu diesem Zweck zu einem einzigen Bild zusammengefügt.In short, the following can be said:

• - The surroundings of a detail, which tend to be unimportant, can be compressed without affecting the resolution of the selected detail. This can result in an enormously reduced storage requirement when using the method according to claim 1, and in particular when using the novel file format described in claim 2, depending on the size of the original image material.
• - A smooth, harmonious transition between high and lower resolution areas within the same image can be guaranteed.
• - Important details within an image can be emphasized/highlighted using a higher resolution compared to the rest of the image.
• - Image files are conceivable that combine the advantages of a wide recording angle and high resolution in the center of the image. The information from several images, which are either from different distances or by using different zoom factors

are created into a single image for this purpose.

Figure list

Figure no. Depicted figure 1 Possible embodiment of the square network structure according to claim 4 2 Possible embodiment of the triangular network structure according to claim 5 3 Possibility of implementing the round network structure according to claim 6; resolution = 4; growth = 4; growth change = 0; extent = 8; rotation = 0; division = 1; division growth = 0; twist = 0; Twist growth = 0' 4 Hint of the round network structure; resolution = 4; growth = 1; growth change = 1; extent = 5; rotation = 0; division = 1; division growth = 1; twist = 0; Twist growth = 0 5 Possible embodiment of the round network structure according to claim 6: Resolution = 4; growth = 4; growth change = 0; extent = unknown; rotation = 0; division = 4; division growth = 0; twist = 0; Twist growth = 0 6 Possibility of implementing the round network structure according to claim 6; resolution = 4; growth = 0; growth change = 0; extent = 4; rotation = 0; division = 2; division growth = 0; twist = 3 degrees; Twist growth = 100% 7 Possible embodiment of the spiral network structure according to claim 7 8th Possible implementation of the user interface of a computer program executing the method according to claim 1 9 Technical implementation of the method according to claim 1 10 Technical implementation of the method according to claim 1

Claims (9)

Computer-implemented method for generating and restructuring a digital image, characterized in that - a point-based, adjustable, clearly reconstructable network structure is generated, - the individual fields of which gradually grow or shrink outwards from a central area of the network structure in such a way that the entire area is covered by those fields without there being any overlap or areas not taken into account, - the network structure in question is then placed over a digital image (hereinafter also referred to as the “original image material”), with the center point the network structure can be located somewhere on the same level, inside or outside the image, - then the respective color of a field of this network structure by calculation, for example based on the Gaussian filter model, or selection from the color information of the original image section immediately below, is determined and - the resulting image is saved in the form of a vector-based graphic, - the underlying original image material is created by manually or using a program - several photographs are taken in succession, - while the distance to the object or the The optical zoom factor is changed - whereby, for the purpose of using the method, either the photographs with the higher optical zoom factor overlay/hide the photographs with the lower zoom factor, - or all the color information contained in the photographs created in this way is evaluated . Procedure according to Claim 1 , characterized in that - the preliminary result of the method is stored losslessly in a specially tailored, also vector-based, file format for graphics, - which in turn is characterized in that the type and parameters of the underlying point-based network structure, an algorithm for determining the order, in which the individual color values are listed, as well as the color values of the fields of the network structure themselves are recorded. Procedure according to Claim 1 , characterized in that - the method according to Claim 1 Frame by frame (image by image), - or across sequences using appropriate algorithms, - is applied to the relevant video material. Procedure according to Claim 1 or 2 , characterized by the use of a square, point-based network structure, characterized in that - its center forms the exact center of a surrounding square center, - which is divided by a grid into at least four identical smaller squares, - which is surrounded by four the rows overlapping the corners, each consisting of identical squares, - where the sum of the side lengths of a certain number of surrounding squares of a row is the The sum of the side lengths of a row of squares each seen from the center corresponds to. Procedure according to Claim 1 or 2 , characterized by the use of a point-based network structure consisting of equilateral triangles, which is characterized by the fact that - its center forms the exact center of a surrounding equilateral triangle, the center, - which is divided by a grid into at least four identical smaller equilateral triangles , - which is surrounded by three rows of identical, interlocking, equilateral triangles that overlap at the corners, - where the sum of the side lengths of a certain number of surrounding triangles of a row is the sum of the side lengths of a row of each starting from the center corresponds to triangles seen in the past. Procedure according to Claim 1 or 2 , characterized by the use of a round, point-based network structure, which is characterized by the fact that - its center forms the exact center of a round center, - which is either not or is divided into a certain number of identical circular sectors, - which is surrounded by circular rings, - which are also divided into identical fields separated from each other by parts of the radius extending from the center, - with the possibility of adding a single field per circular ring that differs by up to 100% in the length of the inner circumferential section in order to increase the growth factor of the fields or To be able to precisely adjust the circular rings from ring to ring - whereby the circular rings can be rotated in relation to each other in any way with respect to the fields they contain - and the side lengths, i.e. the difference between the radius of the outer and inner circumference of the respective circular ring either correspond to the length of the inner section of the circumference of one of these fields or to the length of the circumferential section running through the middle of the respective radius section, - which results in the outer circumference of that circular ring after determining the number of fields within a circular ring. Procedure according to Claim 1 characterized by applying the method to one or more sections of a digital image, i.e. only parts of an image. Procedure according to Claim 1 , characterized in that - by means of a parameter, hereinafter referred to as “growth”, the number of fields added per row/ring is determined, and/or - by means of a parameter, hereinafter referred to as “resolution”, the number of fields which the center of a network structure consists of, is determined, and/or - by means of a parameter, hereinafter referred to as “size”, it is determined how large the area is in the original digital image to be changed, which is overlaid by the center of the created network structure, and /or - by means of a parameter, hereinafter usually referred to as the “location” or the “center point”, it is determined over which exact point in the original image the center of the created network structure is located, and/or - by means of a parameter, hereinafter referred to as “ Coloring” determines the way in which a program evaluates the colors within the boundaries of a field of the network structure superimposed on the original image in order to determine the new color of the entire field, and/or - by means of a parameter, hereinafter called “division”, the fields created after the creation of the network structure are divided, and/or - by means of a parameter, hereinafter referred to as “rotation”, the rotation of the network structure over the original image is determined before the color material is evaluated, and/ or - by means of a parameter, hereinafter referred to as extent, the extent of the network structure is determined so that only parts of the original image are covered. Procedure according to Claim 1 or 2 , characterized in that - the display resolution, i.e. the sharpness of an image generated by the method when displayed on a screen divided into a grid, remains constant to the extent that it is independent of the zoom factor, - as the network structure used allows, - whereby the resolution in the center forms the scale.

Images