Basics Digital Image Proces 1g means processing digital image by means of digital computer, We n order to get enhanced image either to Digital Image Proces ‘can also say that itis a use of computer algorithn extract some useful information Image processing mainly inelude the following steps: importing the image via image acqdisitjon tools; 2.Analysing and manipulating the image; 3.Output in which result can be altered image or a report which is based on analysing that image. What is an image? An image is defined as a two-dimensional functic AE(x,y), where x and y are spatial tensity of coordinates, and the amplitude of F at any pair of coordinates (x,y) is called the t! Whén xy, ahd amplitude values of F are finite, we call it a digital that image at that poi image. In other words, an image can be defined by a two-dimensional array specifically arranged in rows and columns. Digital Image is composed of @ finite" dumber: ofl elemeitts, each of which clements particular value at a particular location. These elements are referred 10 as picture elements, image elements, ane pixeis. A Pixel is mest widely used to denote the elements of a Digital Image. ‘Types of an image : 1. BINARY IMAGE- The binary image as its name suggests, elements i.e 0 &!1,!whéke 0 refersito black snd 1 refers to white. This imag contain only «wo pixel also known, ‘as Monochrome. BLACK AND WHITE IMAGE- The image w' is called BLACK AND WHITE IMAGE. 3, 8 bit COLOR FORMAT- [ts the most famous image format, It has 256 In this format, 0 stands for Black, hhich consists of only black and white color different shades of colors in it and commonly known as Grayseale Im: and 255 stands for white, and 127 stands for gray. @ scanned with OKEN Scanner4.16 bit COLOR FORMAT- itis 4 color image Format H has 65,536 different colors in it IL is also known as High Color Format, In this format the distribution of color is not as same as Grayscale image. divided into three further formats whieh are Red, Green and Blue. RGB A.16 bit format is actually format. That famous Image as a M ix ’As we know. images are represented in rows and columns we have the following Sys!2% in which images are represented: foo) OA) 102) ss OND fino) faa) 022) we END fley)= Fase) (wena) Haz) ve ANA) the right side of this equation is digital image by definition. Every element of this matrix is called image element , picture clement , ot pixel DIGITAL IMAGE REPRESENTATION IN MATLAB: (1,4) F(4,2) « FC1,N) fe £(2,1) (2,2) - F(2,N) #(M, 1) (uy 2) oo £(M, N) stead of 0. ‘Therefore, f(1,1) = (0,0). henceforth the two representation of image are identical, except for the shift in origin In MATLAB, matrices are stored in a variable i.e X,X,input_image , and so on. The variables must be a letter as same as other programming languages. In MATLAB the start index is from @ scanned with OKEN ScannerIMAGE PROC 1. IMAGE ACQUISITION- It could be hs simple as being given an image which is in digital form, The main work involves: i | a) Scaling ty b) Color conversion (RGB to Gray ory ng in areas of 2. IMAGE ENHANCEMENT: ind is ssing itis also used to ( umongst the simplest and most ap taclisomd hidlfen details tr E RESTORATION- It also deals with appealing of an image bat it is obj daotoation is based on tainginitcl or pTeeHb lic model or image degradation), ING le to digital image proc 5, WAVELETS AND MULTI-RESOLUTION PROCESS representing images in various degrees. 4, COLOR IMAGE PROCE processing color models are appli Itdeals with pseudocolor and full color image in ING- I is foundation of folves in developing some functions to perform this 6. IMAGE COMPRESSION-| ¢ oF resolution. operation. It mainly deals with image,s 7, MORPHOLOGICAL PROCESSING-I deals with tools for extracting image components that are useful in the representation & description of shape. 8, SEGMENTATION PROCEDURE-It includes partitioning an image into its constituent pari or objects. Autonomous segmentation is the most dificult task in lage Processing SCRIPTION: It follows output of segmentation stage, 9, REPRESENTATION & D of solution for transforming raw data into processed choosing a representation is only the par data. 10, OBJECT DETECTION AND RECOGNITION -Itisa process that assigns « label to an object based on its descriptor. My OVERLAPPING FIELDS WITH IMAGE PROCESSING Mp , i te i t do @ scanned with OKEN ScannerOUTPUT INPUT IMAGE DESCRIPTION COMMUTER VISION, DIGMAL Manet PROCESSING IMAGE DESCRIPTION ARTIFICIAL INTELLIGENCE] COMPUTER GHAPHCS According to block {if as Digital Image Pros put is an image and we gel out image as a output, then itis termed ng. According (o block 2, if input is an image and we get some kind of information or de 8.4 output, then it is termed as Computer Visidin, iption ‘ording to block 3,if input i it is termed as Computer Graphi some description or code and we get image as an output, then According to block 4,iF input is description or some keywords or some code and we get descri some keywords as a output, then it is termed as Artif{gial Intelligence Applications of Digital Image Processing ‘Some of the major fields in which digital image processing is widely used are mentioned below Image sharpening and restoration * Medical field + Remote sensing + Transmission and encoding + Machine/Robot vision + Color'processing «Patten recognition + Video processing mt i + Microscopic Imaging J « Others vid @ scanned with OKEN ScannerImage sharpening and restoration Image sharpening and restoration refers here to process images that have been captured from the modern camera to make them a better image or to manipulate those images in way to achieve desired result. It refers to do what Photoshop usually does. This includes Zooming, blurring . shafpehing . gray scale to color conversion. detecting edges and vice versa , Image retrieval and Image recognition. The common examples are: The original image Blurrimage = @ scanned with OKEN ScannerMedical fivkd ‘The common applications of DIP in the field of medical is + Gamma ray imaging @ scanned with OKEN Scannerie iy Hal re iit ma i + + PET scan © XRaylmaging | * Medical CT + UV imaging UY imaging Ha In the field of remote sensing , the area of the earth is scanned by a satellite or from a very high ground and then it is analyzed to obtain information about it. One particular application of digital image processing in the field 'éf remote sensing is to detect infrastructure damages caused'by'an earthquake, if [As it takes longer time to grasp damage, even if serious damages are focused on. Since the area effected by the earthquake is sometimes so wide , that it not possible to examine it with human eye in order to estimate damages, Even if it is , then it is very hectic and time consuming procedure, So a solution to this is found in digital image processing. An image of the effected area is captured from the above ground and then it is analyzed to detect the various types of damage done by the earthquake. ‘The key steps include in the analysis are + The extraotion of edges + Analysis and enhancement of yarious types of edges Transmission and encoding, iat ‘The very first image that has been transmitted over the wire was from London to New York via 1 submarine cable.The picture that was sent is shown below. ind z @ scanned with OKEN Scanneranother. The picture that was sent took three hours to reach from, one place 10 ety Seti fom on: Now just imagine , that today we are able to see live video feed , “ 7 ‘work has been done in continent to another with just a delay of seconds. It means that a but also oftfencoding. Many this field too. This field doesnot only focus on transmission . eode photos and then different formats have been developed for high or low bandwith to er ‘stream it over the intemet or e.t.c. ine Robot vision Apart form the many challenges that a robot face today . one of the biggest challenge still fod increase the vision of the robot. Make robot able to sce things , identify them , identify the hurdles e.te. Much work has been contributed by this field and a complete other field of n has been introduced to work on i ile detection o Hurdle detection is one of the common task that has been done through image processing, by identifying different type of objects in the image and then calculating the distance between robot and hurdles, Line follower robot ait yi Most of the robots today work by following the line and thus are called line fot This help a robot to move on its path and perform Some tasks. This has also hae through image processing. i jak i ti Lig i i i AF @ scanned with OKEN ScannerColor processing ‘colored imaiges and different color spaces that are used. Color processing includes processing of es HSV. It also involves studying transmission , storage For example RGB'color model , YCBCr, ‘and encoding of these color images. Pattern recognition i image processing and from various other fields that jneludes machine Jearning ( a branch of artificial intelligence). In pattern recognition , image processing is used for identifying the'objects in an images and then machine learning is used te brain the systemi-for the change’ in'patiem. Pattem recognition is used in computer aided diagnosis, recognition of handwriting ; recognition of images esc Pattern recognition involves study from: Video processing ‘A video is nothing. but just the very fast movement of pictures. The quality of the video depends on the number of frames/pictures pet minute and the quality of each frame being used. Video processing involves noise reduction , detail enhancement , motion detection , frame rate conversion aspect ratio conversion , Color space conversion e.t.c. Image Acquisition In image processing and miachine vision, mage acquisition is the action of retrieving an image from a source, usualy hardware systems, ike cameras, sensors, etc. It is the first and the most important step in the workflow sequence because, without an image, no actual processing is possible by the system. The image that is acquired by the system is usually completely unprocessed. : fot In the image acquisition process, ‘il fig light @ergy ‘from’ an objedt is converted into an electrical signal by the combination of sensors that are sensitive to the particular type of energy. ‘These minute subsystems work together to provide your machine vision algorithm with the most accurate representation of the object. / *! While the sensor system and cameras mostly rely on the technology available, users have complete control over illumination. {) me t @ scanned with OKEN ScannerIMAGE ENHANCEMENT ** ; ity of an image as perceived fy Image enhancement techniques are designed to improve the quality of a i ial as well as in a human being. Image enhancement can be performéd both ae ae iowidge aed frequency domain, After reading this chapter, the reader should ha the following concepis: j ea |. Image enhancement in spatial and frequency domain 2. Point operations and mask operations in spatial domain 3. Different types of gray —level transformation 4, Histogram and histogram equalization a 5. Frequency domain filtering Image enhancement approaches fall into two broad categories: © spatial domain methods © frequency domain methods, ‘The term spatial domain refers to the imajfe ‘plane itself and approaches in this category are based on direct manipulation of pixels in an image. eM Frequency domain processing techniques are based n modifyirig the Fourier ‘transform of an image. Enhancing an image provides better contijst vd a ind d tailed |image compare to non- Ghhanced image. Image enhancement has Very good pplieations, I used to enhance medical images, images captured in remote sensing, images fry mn satelite ete. As indicated previously, the term spatial domain refers to the aggregate of pixels’, ih - nods re procedures that operate directly on these pele, Spatial domain processes will be denoted by the expression. @ scanned with OKEN Scannery) = bisa oe where f(x, y) is the,input ime . y) is the.input image, 2(x, ¥) is! c i wr 9) ele 9) isthe processed image, and isan operator on f defined Lie Image Restoration Restoration ss ya x Scel kno a attempts to reconstruct or recover an image that has been degraded aan amen: odes of the degradation phenomenon. Similar to the enhancement tralituch ane based bn. led fF restoration technique is to improve the image. Restoration techies oa eaal fodeling the degradation using a priori knowledge and applying the process in ofdér to testore theloriginal imagd. Fl ‘ i Restoration: A process that uses alprior knowledge to reconstruct the image that has been degraded. et ee Restoration techniques involve folnifjlating certain(ériteria of goodness that gives the optimal estimate of the desired result. ‘ot On the other hand, enhancement techniques are heuristic procedures fo manipulate the given image to produce the most pleasing image, in order.'0 take advantage of psycho-physical aspects of -hurién visual. system. For example, contest stretching is considered as an asPeeement technique because it s primarily based on the pleasing aspects it might present to Sana teerver, whereas removal of image blur by using a de-blurring function is considered as @ restoration technique. io Restoration techniques can be broadly classified into the following: «Frequency domain i + Algebraic approach. Frequency domain techniques are ej"ielatively less computational complexity, whereas the algebraic approach involves manipulation of large system of simultaneous equations. However, a eer cena conditions the computational complexity ofthe algebraic approach can be reduced athe same level as that of frequency domain restoration technique. In order to restore an image, the degradation phenomenon involved should be understood which is discussed in detail in the following section. i [Allenatural images when displayed have gone through some sort of degradation: 1. During display, mode 2, Acquisition mode, or 3. Processing mode 4.Sensor noise 5, Blur due to camera mis focus 4 i it, @ scanned with OKEN Scanner6. Relative object-camera motion 72 DEGRADATION MODEL erator Hy with an additive Figure 7.1 shows the degradation process which is modeled as an Seed image a iy). Rolse term n(x, y) operating on an input image flx, y) to priduce a dew 7 The input-output relationship of the model i given asin equation (7.1) 8,9) = HUF, yy] 4 nix, y) i: (7a) ‘ His said to be tame 10%) = 0 such that g(x, y) = Hf{s, »)] then the operator / i and k2 are constants andl (x, y) andj2(x, y) are two input images. 'en equation (7.2) can be written as. For simplicity ass linear where 1 IKI = K2=1 th HUG AG ») + Katz, )] = Kies y)-4 KALA, my ALAC + Al y= ALAC, y+ Hh») (73) | Noise Tarm 4 niyo Degraded Image fy) Degradation pesca CU | Operator H (x, y) tig ost From the equation 7.3 we infer that if H is a linear operat f, then the response to a'sum of two inputs is equal to the sum of two responses and. shit is called additivity property. If we assume /2 (x, y)= 0 equation (7.2) reduces to Mt i i a wi Et AUK Sie, 9) = KHL, vy] | i va } } “i @ scanned with OKEN Scanneritive which is x Known as the property of hombogenity Linear Operator; 1 An ope vinden equal a i neht 5 Kor is said to be linear if the response to a sum of two inputs This property sta 1 BEL that inpen anes nt espns to 4 constant muitiple of any input is equal to the response of operator por, iplied by the same congtltt, From this discussion one can conclude that the linear Possesses both the property af.additivity and homogeneity. For any image fw) and any a, values, if HIS —a, y—p)] = gle sey) -B) (7.5) then g(x, y) = Het, y)] is said to be position invariant Position Invariant: An operator is said’t6 be position invariant if it satisfies both additivity and homogeneity properties.) | | This definition indicates that the response at any point in the image depends only on the value of the input at that point and not on the position of the point, th i et @ scanned with OKEN Scanner