Method and system for removing large sawteeth of image, terminal and storage medium

Abstract

The invention relates to a method and system for removing large sawteeth of an image, a terminal and a storage medium, and the method comprises the following steps: carrying out the edge enhancement of an original image, carrying out the binarization conversion of brightness data, carrying out the equal-scale zooming according to a set zooming scale, comparing the zoomed image with a preset model, and carrying out the recognition of the large sawteeth of the image. A specific pixel model is detected and position and direction marking is carried out, and sawtooth related information needing to be filled is obtained; marking the starting point position of the original image by using the sawtooth starting point according to the scaling, and marking the image in the pixel range of the original image by using the distance from the sawtooth starting point as the weight in a rectangular expansion manner according to the direction of the sawtooth; carrying out interpolation calculation on the marked rectangular region, carrying out smooth filtering on data of an interpolation part by utilizing Gaussian blur, and desalting a small sawtooth edge; according to the image obtained after processing, details of the image can be reserved, filling of large sawteeth is completed, and meanwhile hardware cost is very low.

Description

Technical Field

[0001] The present invention relates to the technical field of digital image processing, and more specifically to a method, system, terminal and storage medium for removing large-scale aliasing of an image. Background Art

[0002] In the field of digital image processing, removing image aliasing is a common requirement. Traditional anti-aliasing methods mainly target smaller aliasing in the image, such as cutting the image into 5*5 pixel blocks for processing. Existing processing methods cannot detect and correct larger aliasing. If the size of the pixel block is increased, the hardware overhead will increase exponentially. A method is needed to remove large aliasing using less hardware resources. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method for removing large-scale aliasing of an image in view of the above-mentioned defects of the prior art, and also provide a system for removing large-scale aliasing of an image, a terminal for removing large-scale aliasing of an image and a computer-readable storage medium.

[0004] The technical solution adopted by the present invention to solve its technical problem is:

[0005] A method for removing large-scale aliasing of an image is constructed, which includes the following steps:

[0006] After enhancing the edge of the original image, the brightness data of the original image is binarized to obtain a binary image with clear edges;

[0007] The binary image is proportionally reduced according to the set scaling ratio, and the aliasing requirement of the reduced image can be completely within the set pixel range;

[0008] Compare the reduced image with the preset model, detect the specific pixel model and mark the position and direction, and obtain the information related to the aliasing that needs to be filled;

[0009] According to the sawtooth related information to be filled, the sawtooth starting point is marked according to the above-mentioned scaling ratio for the starting point of the original image, and the image within the pixel range of the original image is marked in a rectangular expansion manner according to the direction of the sawtooth, with the distance from the sawtooth starting point as the weight;

[0010] Interpolation calculation is performed on the marked rectangular area, and the pixels near the center of the sawtooth are weighted averaged as the pixel data to be filled, and the filling is performed according to the weight marked in the rectangular area; after interpolation processing, the large-scale sawtooth of the image has been converted into small-scale sawtooth caused by interpolation;

[0011] Use Gaussian blur to smooth the interpolated data and reduce small jagged edges.

[0012] The method for removing large-scale aliasing of an image according to the present invention, wherein after enhancing the edge of the original image, the brightness data of the original image is binarized and converted into a binary value, comprises:

[0013] An edge detection algorithm is used to obtain edge information of large-size sawtooth in the original image, and edge enhancement processing is performed on the obtained edge information;

[0014] The brightness data of the original image is binarized according to the set binarization conversion threshold.

[0015] In the method for removing large-scale aliasing of an image of the present invention, the set pixel range adopts a 5*5 pixel range;

[0016] The calculation formula for setting the scaling ratio r is:

[0017] r = A / 5;

[0018] The range of the aliasing size that needs to be processed for the original image is A*A.

[0019] In the method for removing large-scale aliasing of an image of the present invention, the calculation formula of the aliasing center coordinate point (M, N) and direction D of the original image is:

[0020] (M, N, D) = (5*m, 5*n, D 1 );

[0021] Where (m, n) is the coordinate center of the current sawtooth in the scaled image, D 1 It is the aliasing direction information in the scaled image.

[0022] The method for removing large-scale jagged edges from an image according to the present invention, wherein the weight W (X,Y) The calculation formula is:

[0023]

[0024] Where (m, n) is the coordinate center of the current sawtooth in the scaled image, D 1 To obtain the aliasing direction information in the scaled image, the aliasing size range of the original image that needs to be processed is A*A.

[0025] The method for removing large-scale aliasing in an image of the present invention, wherein the interpolation calculation is performed on the marked rectangular area, pixels near the center of the aliasing are weighted averaged as pixel data to be filled, and filling is performed according to the weights marked in the rectangular area, including:

[0026] The filling value is the nearest pixel value at the center of the sawtooth in the opposite direction of D for average calculation;

[0027] t is the number of selected pixels, and the calculation method of the brightness G to be filled is:

[0028]

[0029] The brightness value of the filled pixel G (X,Y) The calculation method is:

[0030] G (X,Y) =W (X,Y) *G.

[0031] A system for removing large-scale aliasing in an image, wherein the system comprises an image preprocessing module, a scaling module, a aliasing detection module, an address mapping module, an interpolation calculation module and a Gaussian blur module;

[0032] The image preprocessing module is used to perform binary conversion on the brightness data of the original image after enhancing the edge of the original image to obtain a binary image with clear edges;

[0033] The scaling module is used to scale down the binary image in proportion to the set scaling ratio, and the aliasing requirement of the scaled image can be completely within the set pixel range;

[0034] The jaggies detection module is used to compare the reduced image with a preset model, detect a specific pixel model and mark its position and direction, and obtain information related to the jaggies that need to be filled;

[0035] The address mapping module is used to mark the starting point of the original image according to the above-mentioned scaling ratio according to the sawtooth related information that needs to be filled, and mark the image of the pixel range of the original image in a rectangular expansion manner according to the direction of the sawtooth, taking the distance from the starting point of the sawtooth as the weight;

[0036] The interpolation calculation module is used to perform interpolation calculation on the marked rectangular area, weighted average the pixels near the center of the sawtooth as pixel data to be filled, and fill according to the weights marked in the rectangular area; after the interpolation processing, the large-scale sawtooth of the image has been converted into small-scale sawtooth caused by interpolation;

[0037] The Gaussian blur module is used to perform smoothing filtering on the interpolation part data by using Gaussian blur to reduce small jagged edges.

[0038] In the system for removing large-scale aliasing of an image of the present invention, after enhancing the edge of the original image, the brightness data of the original image is binarized, comprising:

[0039] An edge detection algorithm is used to obtain edge information of large-size sawtooth in the original image, and edge enhancement processing is performed on the obtained edge information;

[0040] The brightness data of the original image is binarized according to the set binarization conversion threshold.

[0041] In the system for removing large-scale aliasing of an image of the present invention, the calculation formula of the aliasing center coordinate point (M, N) and the direction D of the original image is:

[0042] (M, N, D) = (5*m, 5*n, D 1 );

[0043] Where (m, n) is the coordinate center of the current sawtooth in the scaled image, D 1 It is the aliasing direction information in the scaled image;

[0044] Weight W (X,Y) The calculation formula is:

[0045]

[0046] Where (m, n) is the coordinate center of the current sawtooth in the scaled image, D 1 To obtain the aliasing direction information in the scaled image, the aliasing size range of the original image that needs to be processed is A*A;

[0047] Interpolate the marked rectangular area, and weight the pixels near the center of the sawtooth as the pixel data to be filled. Fill the area according to the weights marked in the rectangular area, including:

[0048] The filling value is the nearest pixel value at the center of the sawtooth in the opposite direction of D for average calculation;

[0049] t is the number of selected pixels, and the calculation method of the brightness G to be filled is:

[0050]

[0051] The brightness value of the filled pixel G (X,Y) The calculation method is:

[0052] G (X,Y) =W (X,Y) *G.

[0053] A terminal for removing large aliasing of an image comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0054] A computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the above method when executed by a processor.

[0055] The beneficial effects of the present invention are as follows: after the edge enhancement of the original image, the brightness data of the original image is binarized to obtain a binary image with clear edges; the binary image is proportionally reduced according to a set zoom ratio, and the jagged requirements of the reduced image can completely fall within the set pixel range; the reduced image is compared with a preset model, a specific pixel model is detected and the position and direction are marked, and the jagged information that needs to be filled is obtained; according to the jagged information that needs to be filled, the jagged starting point is marked according to the above-mentioned zoom ratio for the starting point position of the original image, and according to the direction of the jagged, the jagged starting point is expanded in a rectangular manner. The image within the pixel range of the original image is marked with the distance from the starting point of the jaggies as a weight; the marked rectangular area is interpolated and the pixels near the center of the jaggies are weighted averaged as the pixel data to be filled, and filled according to the weights marked in the rectangular area; after the interpolation process, the large jaggies of the image have been converted into small jaggies caused by the interpolation; the data of the interpolated part is smoothed and filtered using Gaussian blur to fade the edges of small jaggies; the image obtained after the above steps can retain the details of the image and complete the filling of large jaggies, and the hardware overhead is also very low, which can well solve the current problem encountered in the industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. The drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work:

[0057] Figure 1 is a flow chart of a method for removing large-scale jagged edges from an image according to a preferred embodiment of the present invention;

[0058] Figure 2 This is an image preprocessing effect diagram of a method for removing large-scale jagged edges in an image according to a preferred embodiment of the present invention;

[0059] Figure 3 This is a schematic diagram of image scaling of a method for removing large-scale jagged edges from an image according to a preferred embodiment of the present invention;

[0060] Figure 4 2. It is a schematic diagram of aliasing detection of a method for removing large aliasing of an image according to a preferred embodiment of the present invention;

[0061] Figure 5This is a schematic diagram of the interpolation effect of a method for removing large-scale aliasing in an image according to a preferred embodiment of the present invention;

[0062] Figure 6 Schematic diagram of Gaussian blur effect of a method for removing large-scale jagged edges of an image according to a preferred embodiment of the present invention;

[0063] Figure 7 This is a de-aliasing effect diagram of a method for removing large-scale aliasing of an image according to a preferred embodiment of the present invention;

[0064] Figure 8 The figure is a block diagram of the system principle for removing large-scale aliasing of an image in a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the following will be described clearly and completely in combination with the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are partial embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0066] The method for removing large-scale jagged edges of an image in a preferred embodiment of the present invention is as follows: Figure 1 See also Figure 2-Figure 7 , the method comprises the following steps:

[0067] S01: After enhancing the edge of the original image, the brightness data of the original image is binarized to obtain a binarized image with clear edges;

[0068] Pretreatment effect reference Figure 2 , this method can reduce the hardware calculation complexity; the edge detection algorithm can use the Sobel operator and other methods to perform convolution operations to obtain the edge information of large-size sawtooth; the binarization conversion threshold can be reasonably set according to the actual image, and there is no limitation on this;

[0069] S02: scaling down the binary image in proportion to the set scaling ratio, and the jagged edges of the image after scaling down must fall completely within the set pixel range;

[0070] Compare before and after scaling Figure 3 As shown, according to the range A*A of the aliasing size that needs to be processed, it is reduced according to a specific ratio r, so that the aliasing of the reduced image can completely fall within the 5*5 pixel range.

[0071] Then: r = A / 5;

[0072] In this example, the sawtooth size is about 20*20 pixels, so r = 20 / 5 = 4. Of course, it can be understood that this scaling ratio can be defined by oneself and is not limited here.

[0073] S03: Compare the scaled-down image with a preset model, detect a specific pixel model, mark its position and direction, and obtain information related to the sawtooth to be filled.

[0074] Still taking a 5*5 pixel image as an example, as Figure 4 shown, when processing, compare the input 5*5 pixel image with the preset model, detect a specific pixel model, mark its position and direction, obtain information related to the sawtooth to be filled, and input the result into the address mapping module. More specifically:

[0075] First, represent the 5 rows of 25 pixels of the input image with 25-bit length data. Set the black part of the image to 0 and the white part to 1. The preset model is also 25-bit data. Therefore, the 5*5 data to be compared in this example is 0X0008FFF. If this data is included in the preset model, output the sawtooth coordinates (m,n) of the corresponding model (where (m,n) is the coordinate center of the current sawtooth in the scaled-down image) and the sawtooth direction information D to the next module to complete sawtooth detection. Of course, the above data description is only an example and is not used for limitation. Other parameter forms can also be adopted according to the situation.

[0076] S04: According to the information related to the sawtooth to be filled, mark the starting point of the sawtooth at the starting position of the original image according to the above scaling ratio, and mark the pixels in the range of the original image in the form of a rectangle expansion with the distance from the starting point of the sawtooth as the weight according to the direction of the sawtooth.

[0077] Specifically, the address mapping module marks the starting point of the sawtooth at the starting position of the original image according to the output result (m,n,D) of the sawtooth detection module according to the above scaling ratio r. According to the above direction of the sawtooth, mark the pixels in the range of A*A pixels of the original image in the form of a rectangle expansion with the distance from the starting point of the sawtooth as the weight.

[0078] The center coordinate point (M,N) and direction D of the sawtooth in the original image 1 The calculation method is:

[0079] (M,N,D 1 ) = (5*m, 5*n, D)

[0080] The weight is related to the distance between the pixel (X,Y) to be calculated and the sawtooth center point (M,N). The maximum pixel range that needs to be interpolated is equal to the sawtooth size A*A to be removed. In this example, it is 20*20. The weight value range is (1~0.5).

[0081] The weight calculation method is:

[0082]

[0083] S05: performing interpolation calculation on the marked rectangular area, taking the weighted average of pixels near the center of the sawtooth as pixel data to be filled, and filling according to the weights marked in the rectangular area;

[0084] Specifically, interpolation calculation is performed on the marked rectangular area, and the pixels near the center of the jaggies are weighted averaged as the pixel data to be filled, and the filling is performed according to the weights marked in the rectangular area. After interpolation calculation, the original image is obtained to obtain an image with less jaggies.

[0085] Among them, the filling value selects the nearest pixel value at the center of the sawtooth in the opposite direction of D for average calculation, t is the number of selected pixels, and the calculation method of the brightness G to be filled is:

[0086]

[0087] The brightness value of the filled pixel is G (X,Y) =W (X,Y) *G;

[0088] like Figure 5 As shown, after interpolation, the large aliasing of the image has been converted into small aliasing caused by interpolation;

[0089] S06: Use Gaussian blur to smooth the interpolated data and reduce small jagged edges;

[0090] Gaussian blur effect Figure 6 shown.

[0091] like Figure 7 As shown, the image obtained after the above steps can not only retain the details of the image, but also complete the filling of large aliasing. At the same time, the hardware overhead is also very low, which can well solve the current problem encountered in the industry.

[0092] A system for removing large aliasing from an image, such as Figure 8 As shown, the system includes an image preprocessing module 100, a scaling module 101, a jaggies detection module 102, an address mapping module 103, an interpolation calculation module 104 and a Gaussian blur module 105;

[0093] The image preprocessing module 100 is used to perform binary conversion on the brightness data of the original image after enhancing the edge of the original image to obtain a binary image with clear edges;

[0094] Scaling module 101, used to scale down the binary image in proportion to a set scaling ratio, and the aliasing requirement of the scaled image can be completely within a set pixel range;

[0095] The aliasing detection module 102 is used to compare the reduced image with a preset model, detect a specific pixel model and mark the position and direction, and obtain information related to the aliasing that needs to be filled;

[0096] The address mapping module 103 is used to mark the starting point of the original image according to the above-mentioned scaling ratio according to the sawtooth related information that needs to be filled, and mark the image of the pixel range of the original image in a rectangular expansion manner according to the direction of the sawtooth, taking the distance from the starting point of the sawtooth as the weight;

[0097] The interpolation calculation module 104 is used to perform interpolation calculation on the marked rectangular area, weighted average the pixels near the center of the sawtooth as pixel data to be filled, and fill according to the weights marked in the rectangular area; after the interpolation processing, the large-scale sawtooth of the image has been converted into small-scale sawtooth caused by interpolation;

[0098] A Gaussian blur module 105 is used to smooth the interpolation data by using Gaussian blur to reduce small jagged edges;

[0099] The system of the present application can retain the details of the image and complete the filling of large aliasing. At the same time, the hardware overhead is also very low, which can well solve the current problem encountered in the industry. The implementation description of each module is described in the above description and will not be repeated here.

[0100] A terminal for removing large aliasing of an image comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0101] A computer-readable storage medium storing a computer program, wherein the computer program implements the steps of the above method when executed by a processor.

[0102] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A method for removing large-scale aliasing of an image, characterized in that: The following steps are involved: After enhancing the edge of the original image, the brightness data of the original image is binarized to obtain a binary image with clear edges; The binary image is proportionally reduced according to the set scaling ratio, and the jagged edges of the reduced image must fall completely within the set pixel range; Compare the reduced image with the preset model, detect the specific pixel model and mark the position and direction, and obtain the information related to the aliasing that needs to be filled; According to the sawtooth related information to be filled, the sawtooth starting point is marked according to the above-mentioned scaling ratio for the starting point of the original image, and the image within the pixel range of the original image is marked in a rectangular expansion manner according to the direction of the sawtooth, with the distance from the sawtooth starting point as the weight; Interpolation calculation is performed on the marked rectangular area, and the pixels near the center of the sawtooth are weighted averaged as the pixel data to be filled, and the filling is performed according to the weight marked in the rectangular area; after interpolation processing, the large-scale sawtooth of the image has been converted into small-scale sawtooth caused by interpolation; Use Gaussian blur to smooth the interpolated data and reduce small jagged edges.

2. The method for removing large-scale aliasing of an image according to claim 1, characterized in that: After the edge enhancement of the original image, the brightness data of the original image is binarized, including: An edge detection algorithm is used to obtain edge information of large-size sawtooth in the original image, and edge enhancement processing is performed on the obtained edge information; The brightness data of the original image is binarized according to the set binarization conversion threshold.

3. The method for removing large-scale aliasing of an image according to claim 1, characterized in that: The set pixel range adopts a 5*5 pixel range; The calculation formula for setting the scaling ratio r is: r＝A / 5； Among them, the range of the aliasing size that needs to be processed in the original image is A*A; The calculation formula of the sawtooth center coordinate point (M, N) and direction D of the original image is: (M, N, D) = (5*m, 5*n, D1); Wherein, (m, n) is the coordinate center of the current sawtooth in the scaled image, and D1 is the sawtooth direction information in the scaled image.

4. The method for removing large-scale aliasing of an image according to claim 3, characterized in that: The weight W (X,Y) The calculation formula is: Wherein, (m, n) is the coordinate center of the current aliasing in the scaled image, D1 is the aliasing direction information in the scaled image, and the range of the aliasing size that needs to be processed in the original image is A*A.

5. The method for removing large-scale aliasing of an image according to claim 4, characterized in that: The interpolation calculation is performed on the marked rectangular area, pixels near the center of the sawtooth are weighted averaged as pixel data to be filled, and filling is performed according to the weights marked in the rectangular area, including: The filling value is the nearest pixel value at the center of the sawtooth in the opposite direction of D for average calculation; t is the number of selected pixels, and the calculation method of the brightness G to be filled is: The brightness value of the filled pixel G (X,Y) The calculation method is: G (X,Y) =W (X,Y) *G。 6. A system for removing large-scale aliasing of an image, characterized in that: The system includes an image preprocessing module, a scaling module, a jaggies detection module, an address mapping module, an interpolation calculation module and a Gaussian blur module; The image preprocessing module is used to perform binary conversion on the brightness data of the original image after enhancing the edge of the original image to obtain a binary image with clear edges; The scaling module is used to scale down the binary image in proportion to the set scaling ratio, and the aliasing requirement of the scaled image can be completely within the set pixel range; The jaggies detection module is used to compare the reduced image with a preset model, detect a specific pixel model and mark its position and direction, and obtain information related to the jaggies that need to be filled; The address mapping module is used to mark the starting point of the original image according to the above-mentioned scaling ratio according to the sawtooth related information that needs to be filled, and mark the image of the pixel range of the original image in a rectangular expansion manner according to the direction of the sawtooth, taking the distance from the starting point of the sawtooth as the weight; The interpolation calculation module is used to perform interpolation calculation on the marked rectangular area, weighted average the pixels near the center of the sawtooth as pixel data to be filled, and fill according to the weights marked in the rectangular area; after the interpolation processing, the large-scale sawtooth of the image has been converted into small-scale sawtooth caused by interpolation; The Gaussian blur module is used to perform smoothing filtering on the interpolation part data by using Gaussian blur to reduce small jagged edges.

7. The system for removing large-scale aliasing of an image according to claim 6, characterized in that: After the edge enhancement of the original image, the brightness data of the original image is binarized, including: An edge detection algorithm is used to obtain edge information of large-size sawtooth in the original image, and edge enhancement processing is performed on the obtained edge information; The brightness data of the original image is binarized according to the set binarization conversion threshold.

8. The system for removing large-scale aliasing of an image according to claim 6, characterized in that: The calculation formula of the sawtooth center coordinate point (M, N) and direction D of the original image is: (M, N, D) = (5*m, 5*n, D1); Wherein, (m, n) is the coordinate center of the current sawtooth in the scaled image, and D1 is the sawtooth direction information in the scaled image; Weight W (X,Y) The calculation formula is: Wherein, (m, n) is the coordinate center of the current sawtooth in the scaled image, D1 is the sawtooth direction information in the scaled image, and the range of the sawtooth size that needs to be processed in the original image is A*A; Interpolate the marked rectangular area, and weight the pixels near the center of the sawtooth as the pixel data to be filled. Fill the area according to the weights marked in the rectangular area, including: The filling value is the nearest pixel value at the center of the sawtooth in the opposite direction of D for average calculation; t is the number of selected pixels, and the calculation method of the brightness G to be filled is: The brightness value of the filled pixel G (X,Y) The calculation method is: G (X,Y) =W (X,Y) *G。 9. A terminal for removing large-scale jagged edges from an image, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images