CN116343253A - CAD drawing length unit and pixel value proportion identification, acquisition and calculation method - Google Patents

Abstract

The invention discloses a method for identifying, acquiring and calculating the ratio of a CAD drawing length unit to a pixel value, which comprises the following operation steps: cutting out the area to be identified, corroding and binarizing to make the scale line on the image obvious, detecting the outline, filtering the outline with too small and too large, judging whether the outline is rectangular, and reserving the rectangle to be predicted; feeding into a model for detection; after the detection is finished, the non-compliance number and the rectangle are filtered, the pixel ratio is calculated, and the comparison value result is taken as the average value of three random samples, so that the result is obtained. The invention relates to a method for identifying, acquiring and calculating the ratio of a CAD drawing length unit to a pixel value, which automatically acquires the pixel value on the CAD drawing and the number of corresponding pixel points on the drawing by using an AI method, thereby calculating the ratio of the pixel points to the length of an actual space and automatically calculating the length of the actual space represented by a building CAD image pixel.

Description

A Method for Recognition, Acquisition and Calculation of Length Unit and Pixel Value Ratio of CAD Drawings

technical field

The invention relates to the field of computer software, in particular to a method for identifying, acquiring and calculating the ratio between length units and pixel values of CAD drawings.

Background technique

The identification, acquisition and calculation method of the length unit and pixel value ratio of CAD drawings is a method for calculating the ratio between CAD drawings and the actual ratio. Using the AI method, the pixel values on the CAD drawings are automatically obtained and their correspondence on the drawings is obtained. The number of pixels, so as to calculate the length ratio of pixels to the actual space. With the continuous development of science and technology, people have more and more requirements for the manufacturing process of identifying, obtaining and calculating the length unit and pixel value ratio of CAD drawings. high.

The existing CAD drawing length unit and pixel value ratio identification, acquisition and calculation methods have certain disadvantages in use. At present, the calculation of the ratio of CAD drawing pixel value to the actual space length depends on professional software such as photoshop, which requires Manual calibration of pixel values is difficult to meet the existing automation needs, and has brought certain adverse effects to the actual use process. Therefore, we propose a method for identifying, acquiring and calculating the ratio between length units and pixel values of CAD drawings method.

Contents of the invention

Aiming at the deficiencies of the prior art, the present invention provides a method for identifying, acquiring and calculating the ratio between the length unit and the pixel value of a CAD drawing, using the method of AI to automatically acquire the pixel value on the CAD drawing and obtain its corresponding value on the drawing. The number of corresponding pixels, so as to calculate the length ratio of the pixels to the actual space, and automatically calculate the length of the actual space represented by the pixels of the architectural CAD image, can effectively solve the problems in the background technology.

In order to achieve the above object, the technical solution adopted by the present invention is: a method for identifying, obtaining and calculating the ratio of length unit and pixel value of a CAD drawing, comprising the following steps:

S1: Arranging ideas: the area of the scale line with numbers is mostly bounded by a rectangle, locate and identify the rectangle, and then identify the numbers in the rectangle;

S2: Region Segmentation: Segment the region that needs to be identified, on the one hand for regional positioning, and on the other hand to save computing resources;

S3: Image enhancement preprocessing: After corrosion and binarization, the scale line on the image becomes obvious, then detects the contour, filters the too small and too large contours, and judges whether the contour is a rectangle. For non-matrix contours and no middle The outline of the content is filtered, and the rectangle to be predicted is reserved;

S4: Perform digital detection: In order to improve the accuracy of recognition, change the aspect ratio and fill the rectangles one by one, and send them to the model detection;

S5: Acquisition and calculation: After the detection is completed, the non-compliant numbers and rectangles are filtered out, and the pixel ratio is calculated, and the comparison value is obtained by taking the average of three random samplings.

As a preferred technical solution of the present application, in the step S4, the input model detection adopts the paddlepaddleocr model of the Paddle framework, including text detection and recognition parts.

As a preferred technical solution of the present application, the input model detection in the step S4 can adopt the tuned ctpn+crnn model or other character/number recognition models of other deep learning frameworks.

As a preferred technical solution of the present application, it specifically includes the following steps:

A1: Pass the CAD drawings with scale lines and scale line values into this algorithm;

A2: Using the method of this algorithm, first segment out the area to be identified;

A3: Do contour detection after image enhancement, detect and filter out the rectangle contour, transform and fill the rectangle shape and send them to PaddlePaddleocr detection numbers one by one;

A4: After filtering out the irregular CAD numbers, according to the number of pixels on the number and the length of the rectangle, the proportional relationship between the distance between the unit pixel in the CAD drawing and the actual building can be calculated, and the proportional relationship results of multiple numbers are sampled three times to calculate the average , to get the final result.

As a preferred technical solution of the present application, the processes in the steps A1-A4 are image cutting, preprocessing, finding contours, filtering and judging contour shapes, filtering rectangles without content, model detection, calculating pixel ratios, and screening pixel ratios, Sampling takes the mean value and ends.

As a preferred technical solution of the present application, the steps A1-A4 automatically obtain the pixel value ratio of the CAD drawing.

As a preferred technical solution of the present application, in the steps S1-S5, the length of the actual space represented by the pixels of the architectural CAD image is automatically calculated.

As a preferred technical solution of the present application, the steps S1-S5 are mainly to locate and recognize the rectangular frame, and then recognize the numbers in the rectangle.

Compared with the prior art, the present invention provides a method for identifying, acquiring and calculating the ratio of the length unit of the CAD drawing to the pixel value, which has the following beneficial effects: the identification, acquisition and calculation of the ratio of the length unit of the CAD drawing to the pixel value The calculation method uses the method of AI to automatically obtain the pixel value on the CAD drawing and the number of corresponding pixel points on the drawing, so as to calculate the length ratio between the pixel point and the actual space, and represent the pixel of the architectural CAD image The length of the actual space can be automatically calculated, and the pixel value ratio of the CAD drawing can be automatically obtained. Abandoning the traditional method of requiring professional software, the user needs to import the CAD drawing with the ruler line and the value of the ruler line into this algorithm. Using this Algorithm method, first cut out the area that needs to be recognized, and then do contour detection after image enhancement, after detecting and screening out the rectangular contour, transform and fill the rectangular shape and send them to PaddlePaddleocr to detect numbers one by one, and filter out irregular CAD numbers Finally, according to the number of pixels on the number and the length of the rectangle, the proportional relationship between the distance between the unit pixel in the CAD drawing and the actual building can be calculated, and the result of the proportional relationship between multiple numbers is sampled three times to calculate the mean value, and the final result can be obtained. The identification, acquisition and calculation method of the length unit and pixel value ratio of CAD drawings is simpler, more convenient to operate, and the effect of use is better than the traditional method.

Description of drawings

Fig. 1 is a schematic diagram of the overall process of the identification, acquisition and calculation method of the length unit and pixel value ratio of a CAD drawing according to the present invention.

Implementation

The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments, but those skilled in the art will understand that the following described embodiments are some embodiments of the present invention, rather than all embodiments. It is only used to illustrate the present invention and should not be construed as limiting the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention. Those who do not indicate the specific conditions in the examples are carried out according to the conventional conditions or the conditions suggested by the manufacturer. The reagents or instruments used were not indicated by the manufacturer, and they were all conventional products that could be purchased from the market.

In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" etc. The indicated orientation or positional relationship is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, or in a specific orientation. construction and operation, therefore, should not be construed as limiting the invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and should not be construed as indicating or implying relative importance.

In the description of the present invention, it should be noted that, unless otherwise clearly stipulated and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection or a detachable connection. Connected, or integrally connected; it can be mechanically connected or electrically connected; it can be directly connected or indirectly connected through an intermediary, and it can be the internal communication of two components. Those of ordinary skill in the art can understand the specific meanings of the above terms in the present invention in specific situations.

As shown in Figure 1, a method for identifying, acquiring and calculating the ratio between the length unit and the pixel value of a CAD drawing includes the following steps:

S1: Arranging ideas: the area of the scale line with numbers is mostly bounded by a rectangle, locate and identify the rectangle, and then identify the numbers in the rectangle;

S2: Region Segmentation: Segment the region that needs to be identified, on the one hand for regional positioning, and on the other hand to save computing resources;

S3: Image enhancement preprocessing: After corrosion and binarization, the scale line on the image becomes obvious, then detects the contour, filters the too small and too large contours, and judges whether the contour is a rectangle. For non-matrix contours and no middle The outline of the content is filtered, and the rectangle to be predicted is reserved;

S4: Perform digital detection: In order to improve the accuracy of recognition, change the aspect ratio and fill the rectangles one by one, and send them to the model detection;

S5: Acquisition and calculation: After the detection is completed, after filtering out the non-compliant numbers and rectangles, calculate the pixel ratio, and take the average of three random samplings for the comparison value results, and obtain the results, and use the AI method to automatically obtain the CAD drawings. Pixel value and obtain its corresponding pixel number on the drawing, so as to calculate the length ratio of the pixel point to the actual space, and automatically calculate the length of the actual space represented by the pixel of the architectural CAD image.

Further, the input model detection in step S4 adopts the paddlepaddleocr model of the Paddle framework, including text detection and recognition parts.

Further, in the step S4, the input model detection can adopt the tuned ctpn+crnn model of other deep learning frameworks or other character/number recognition models.

Further, it specifically includes the following steps:

A1: Pass the CAD drawings with scale lines and scale line values into this algorithm;

A2: Using the method of this algorithm, first segment out the area to be identified;

A3: Do contour detection after image enhancement, detect and filter out the rectangle contour, transform and fill the rectangle shape and send them to PaddlePaddleocr detection numbers one by one;

A4: After filtering out the irregular CAD numbers, according to the number of pixels on the number and the length of the rectangle, the proportional relationship between the distance between the unit pixel in the CAD drawing and the actual building can be calculated, and the proportional relationship results of multiple numbers are sampled three times to calculate the average , to get the final result.

Further, the process in steps A1-A4 is image cutting, preprocessing, finding contours, filtering and judging contour shapes, filtering rectangles without content, model detection, calculating pixel ratios, screening pixel ratios, sampling to get the mean value, and ending.

Further, steps A1-A4 automatically obtain the pixel value ratio of the CAD drawing.

Further, in steps S1-S5, the length of the actual space represented by the pixels of the architectural CAD image is automatically calculated.

Further, the steps S1-S5 are mainly to locate and recognize the rectangular frame, and then recognize the numbers in the rectangle.

Working principle: Use AI method to automatically obtain the pixel value on the CAD drawing and obtain the corresponding pixel number on the drawing, so as to calculate the length ratio between the pixel point and the actual space, and represent the pixel of the architectural CAD image Automatically calculate the length of the actual space;

The user needs to import the CAD drawing with the ruler line and the value of the ruler line into this algorithm. Using the method of this algorithm, first cut out the area to be recognized, and then perform contour detection after image enhancement, and detect and filter out the rectangle. After the outline, transform and fill the rectangle shape and send it to PaddlePaddleocr to detect the numbers one by one. After filtering out the irregular CAD numbers, the distance between the unit pixel in the CAD drawing and the actual building can be calculated according to the number of pixels on the number and the length of the rectangle. Proportional relationship, the proportional relationship results of multiple numbers are sampled three times to calculate the mean, and the final result can be obtained.

It should be noted that in this article, relational terms such as first and second (number one, number two), etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between such entities or operations. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

The basic principles and main features of the present invention and the advantages of the present invention have been shown and described above. Those skilled in the industry should understand that the present invention is not limited by the above-mentioned embodiments. What are described in the above-mentioned embodiments and the description only illustrate the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will also have Variations and improvements are possible, which fall within the scope of the claimed invention.

Claims (8)

1. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing is characterized by comprising the following steps of: the method comprises the following operation steps:

s1: and (3) idea arrangement: the scale line area with the numbers is mostly defined by a rectangular frame, the rectangular frame is positioned and identified, and then the numbers in the rectangle are identified;

s2: region segmentation: the areas to be identified are segmented, and the areas are used for positioning the areas on one hand and saving computing resources on the other hand;

s3: image enhancement preprocessing: after corrosion and binarization, the scale line on the image becomes obvious, then the outline is detected, whether the outline is rectangular or not is judged after the too small and too large outline is filtered, the outline which is not a matrix and the outline without content in the middle are filtered, and the rectangle to be predicted is reserved;

s4: digital detection is carried out: in order to improve the recognition accuracy, changing the length-width ratio and filling the rectangles one by one, and then sending the rectangles into a model for detection;

s5: and (3) obtaining and calculating: after the detection is finished, the non-compliance number and the rectangle are filtered, the pixel ratio is calculated, and the comparison value result is taken as the average value of three random samples, so that the result is obtained.

2. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing according to claim 1, wherein the method comprises the following steps: the step S4 is to send the model detection to adopt a paddlepaddleocr model which is a Paddle frame, and the method comprises a character detection and recognition part.

3. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing according to claim 1, wherein the method comprises the following steps: the model detection sent in the step S4 can adopt a ctpn+crnn model or other character/number recognition models of the tuning of other deep learning frameworks.

4. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing according to claim 1, wherein the method comprises the following steps: the method specifically comprises the following operation steps:

a1: the CAD drawing with the scale line and the scale line value is transmitted into the algorithm;

a2: the method of the algorithm is used for firstly cutting out the area to be identified;

a3: performing contour detection after image enhancement, detecting and screening rectangular contours, transforming and filling the rectangular shapes, and sending the rectangular shapes into PaddlePaddleocr detection numbers one by one;

a4: and after the denormal CAD numbers are filtered, according to the numbers and the number of pixels on the rectangular length, the proportional relation between the unit pixels and the distance of the actual building in the CAD drawing can be calculated, and the proportional relation result of a plurality of numbers is sampled three times to calculate the average value, so that the final result can be obtained.

5. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing according to claim 4, wherein: the process of the step A1-A4 is image cutting, preprocessing, contour finding, filtering and judging contour shape, filtering rectangle without content, detecting a model, calculating pixel ratio, screening pixel ratio, sampling and taking average value, and ending.

6. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing according to claim 4, wherein: and the steps A1-A4 automatically acquire the pixel value proportion of the CAD drawing.

7. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing according to claim 1, wherein the method comprises the following steps: and in the steps S1-S5, automatically calculating the length of the actual space represented by the building CAD image pixels.

8. The method for identifying, acquiring and calculating the ratio of the length unit to the pixel value of the CAD drawing according to claim 1, wherein the method comprises the following steps: the steps S1-S5 mainly comprise positioning and identifying rectangular frames and then identifying numbers in the rectangles.

Images