TWI790761B - Image correction method and processor - Google Patents

Abstract

An image correction method and a processor are provided. The method includes the following steps: performing a feature point search on a quick response code (QR code) image to determine multiple feature points; dividing the coded area of the QR code image into multiple sub-regions according to the multiple feature points; determining a compensation vector for each sub-region according to the feature points corresponding to each sub-region; and compensating and correcting each sub-region according to the compensation vector of each sub-region to obtain a corrected image. Therefore, the solution provided by the present application use compensation vectors to correct the QR code image by region and can avoid interference between different sub-regions, thereby more accurately correcting the distortion of the QR code image.

Description

Image correction method and processor

The present application relates to the technical field of image processing, in particular to an image correction method and a processor.

Quick response matrix code, as a commonly used matrix two-dimensional code, can be applied not only to flat prints, but also to the surface of objects of various shapes. For example, a quick-response matrix code containing air-conditioner-related information can be pasted on the surface of the air-conditioner for decoding devices (such as mobile phones, tablets, etc.) to capture and decode to obtain air-conditioner-related information. However, distortion of the quick response matrix code itself or insufficient shooting capability of the decoding device may cause the decoding device to capture a deformed quick response matrix code image. In order to realize the decoding of the deformed fast response matrix code image, it needs to be corrected.

The present application provides an image correction method and a processor, which can correct the deformation of the fast response matrix code image.

The present application provides an image correction method, which includes: searching for feature points on a quick response matrix code image to determine multiple feature points in the quick response matrix code image; according to the determined multiple feature points, quickly The encoding region of the response matrix code image is divided into a plurality of subregions; and the root A compensation vector for each sub-region is determined according to the feature points corresponding to each sub-region, and compensation and correction are performed for each sub-region according to the compensation vector for each sub-region to obtain a corrected image.

The present application provides a processor, the processor executes a computer program to implement an image correction method, wherein the image correction method includes: performing feature point search on a fast response matrix code image, and determining the A plurality of feature points in the quick response matrix code image; according to the plurality of feature points, a coding area of the quick response matrix code image is divided into a plurality of sub-areas; and, according to each of the sub-areas corresponding feature points, determine a compensation vector for each of the sub-regions, and perform compensation and correction on each of the sub-regions according to the compensation vector of each of the sub-regions to obtain a corrected image.

The present application corrects the quick response matrix code image by adopting a sub-area vector compensation method, which can avoid interference between different sub-areas, thereby more accurately correcting the deformation of the quick response matrix code image.

S110, S120, S130: steps

p1,p2,p3,p4,p5,p6,p7,p8,p9,p10,p11,p12: feature points

A1,A2,A3: center point

B1, B2, B3: Outer edge points

1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17: sub-regions

01:Electronic equipment

10: Wafer

20: Memory

30: camera

100: Processor

200: interface circuit

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained based on these drawings without any creative effort.

FIG. 1 is a schematic flow chart of the image correction method provided by the present application.

FIG. 2 is an example diagram of a corrected image obtained by performing region-by-area vector compensation correction on the fast response matrix code image in the present application.

FIG. 3 is an example diagram of feature points searched in this application.

FIG. 4 is an example diagram of the central point and outer edge points of the finder pattern searched during the process of searching for the feature points shown in FIG. 3 in the present application.

Fig. 5 is an example diagram of three different regions division methods for the same part of the coding region in this application.

Fig. 6 is an example diagram of dividing the coding area into 17 sub-areas in this application.

FIG. 7 is a schematic block diagram of an electronic device 01 provided by an embodiment of the present application.

It should be noted that the principles of the present application are implemented in an appropriate application environment for illustration. The following descriptions are based on illustrated specific embodiments of the present application, which should not be construed as limiting other specific embodiments of the present application that are not described in detail here.

The solutions provided by the embodiments of the present application relate to the field of image processing technology, and specifically relate to the correction of images of fast response matrix codes, which will be described through the following embodiments. Please refer to FIG. 1. FIG. 1 is a schematic flow chart of the image correction method provided in the embodiment of the present application. As shown in FIG. 1, the method may include:

In step S110, a feature point search is performed on a quick response matrix code image to determine a plurality of feature points in the quick response matrix code image.

The following takes the image correction method implemented in an electronic device as an example for illustration. Here, there is no specific limitation on the physical display form of the electronic device, which may be mobile electronic devices such as smart phones and tablet computers.

In this embodiment, the electronic device includes a processor, which implements the image correction method of the present invention by executing a computer program, and the processor can perform a search on a The quick response matrix code image is searched for feature points, so as to determine multiple feature points in the quick response matrix code image. It should be noted that there is no specific limitation on the configuration of the feature point search strategy in this embodiment, and it can be configured by those skilled in the art according to actual needs. For example, the processor can use the corner detection algorithm to search for points with corner points (corner points are extreme points, that is, points with particularly prominent attributes) from the QRC image as feature points. Among them, the corner detection algorithm includes but not limited to Harris (Harris) algorithm, scale-invariant feature transform (Scale-invariant feature transform, SIFT) algorithm, accelerated robust features (Speeded Up Robust Features, SURF) algorithm and Binary feature with rotation (Oriented FAST and Rotated BRIEF, ORB) algorithm, etc.

In addition, in this embodiment, there is no specific limitation on the source of the above quick response matrix code image. For example, when the electronic device is equipped with a camera, the electronic device can obtain the quick response matrix code by photographing the quick response matrix code through the configuration of the camera, or it can be obtained by the electronic device Obtained from the Internet, etc.

Step S120: Divide a coding area of the quick response matrix code image into multiple sub-areas according to multiple feature points.

Wherein, the processor divides a coded area of the quick response matrix code image into multiple sub-areas according to the configured area division strategy and based on the determined multiple feature points. There is no specific limitation on the configuration of the area division policy here, and it can be configured by those skilled in the art according to actual needs.

Step S130, determine a compensation vector for each sub-region according to the feature points corresponding to each sub-region, and perform compensation and correction for each sub-region according to the compensation vector for each sub-region to obtain a corrected image.

Among them, the processor, for each sub-region of the coding region, through the determined compensation to The amount compensates for the offset caused by the deformation, and finally obtains a corrected image for decoding. For example, please refer to FIG. 2 , after the processor performs region-by-region vector compensation correction on a deformed fast response matrix code image, a corrected image in which the symbol offset is compensated is obtained.

In an embodiment, searching for a plurality of finding patterns in the quick response matrix code image; and searching and determining a plurality of feature points according to the searching of the plurality of finding patterns.

Image-finding graphics (also known as position detection graphics, detection graphics, etc.) are used to locate the quick response matrix code image. The current quick response matrix code usually includes three image finding patterns, which are respectively located on the three corners of the quick response matrix code, ie, the lower left corner, the upper left corner and the upper right corner of the quick response matrix code. For example, please refer to Fig. 2, in which three "back"-shaped graphics are image-finding graphics.

It should be noted that the width ratio of the black and white pixels of the image finding pattern is 1:1:3:1:1. According to this feature, the processor can search for the three image finding patterns in the quick response matrix code image.

Exemplarily, in order to realize the search for the image finding pattern in the quick response matrix code image, the processor first establishes a coordinate system, for example, establishes a pixel coordinate system with the upper left corner vertex of the quick response matrix code image as the origin. Then, the processor scans the quick response matrix code image according to the feature that the width ratio of the black and white pixels of the finding pattern is 1:1:3:1:1, so as to obtain three finding patterns. For example, the processor can perform progressive scanning on the quick response matrix code image, and can also perform interlaced scanning on the quick response matrix code image. Among them, the row refers to a row of pixels of the quick response matrix code image. There is no specific limit to the number of interlaced rows here, and it can be configured by those skilled in the art according to actual needs. For example, the quick response can be paired every 3 rows. Scan the matrix code image to quickly search out the three image-finding graphics.

In an embodiment, when searching for a plurality of feature points, the processor further searches and determines a plurality of feature points based on the plurality of imaging patterns determined by searching. For example, when the search identifies three When there are three image-finding graphics, the processor uses the three image-finding graphics as search objects to search and determine a plurality of feature points.

In one embodiment, after searching and determining a plurality of image-finding patterns, the processor does not immediately search for feature points according to the searched-out plurality of image patterns, but first responds quickly to the matrix code image according to the plurality of image-finding patterns like decoding. For example, the processor may decode the quick response matrix code image in a perspective transformation manner, and if the decoding is successful, correspondingly obtain a decoding result. If the decoding fails, the processor determines that the image of the quick response matrix code is deformed, and at this time searches and determines a plurality of feature points according to a plurality of imaging patterns for correcting the deformation of the image of the quick response matrix code.

In an embodiment, the corrected image is decoded to obtain a decoding result.

For example, the processor decodes the corrected image shown in FIG. 2 to obtain the decoding result "20210113 QR code 1".

In one embodiment, when a plurality of feature points are determined according to multiple image-finding patterns, the processor can search for multiple vertices of each image-finding pattern close to the coding area to determine as feature points, and search for the quick response matrix code map A center point of a calibration reference pattern in the image is determined as a feature point. The correction reference pattern is similar to the image finding pattern, which is also a black and white rectangular block, but its black and white pixel ratio is 1:1:1:1:1. According to this feature, the processor can search for the correction in the fast response matrix code image Reference graphic, and search for the center point of the calibration reference graphic.

For example, please refer to Fig. 3, the processor finds that in the quick response matrix code image shown in the illustration, the image-seeking figure in the upper-left corner is close to the upper-right vertex p1, the lower-right vertex p4, and the lower-left vertex p3 of the coding area, and the image-seeking figure in the upper right corner is close to The upper left vertex p2, the lower left vertex p5, the lower right vertex p6 of the encoding area, the image finding figure in the lower left corner is close to the upper left vertex p7, the upper right vertex p8, the lower right vertex p10 of the encoding area, and the center of the correction reference graphic in the lower right corner Point p9, and use p1, p2, p3, p4, p5, p6, p7, p8, p9 and p10 as feature points.

It should be noted that this embodiment does not make specific restrictions on how to perform vertex search, and the vertex search method can be configured by those skilled in the art according to actual needs. For example, the processor can use morphological erosion and expansion, edge detection and Hough (Hough) transform To search out multiple vertices of each image-seeking pattern close to the coding area, and by locating the edge of each image-seeking pattern, calculate the boundary equation to deduce the multiple vertices of each image-finding pattern close to the coding area, etc.

In one embodiment, when searching for multiple vertices of each imaging pattern close to the coding area to determine as feature points, the processor can use the center point of each imaging pattern as a starting point to search along a search direction to determine each Find an outer edge point of the image-finding graphic, and start from the outer edge point of each image-finding graphic, and search along the outer edge of each image-finding graphic to determine a plurality of vertices.

In practice, the search direction can be configured by those skilled in the art according to actual needs, for example, either the x-axis direction or the y-axis direction can be configured as the search direction.

For example, please refer to Fig. 4, taking the x-axis direction as an example for the search direction, for the image-seeking pattern in the upper left corner of the quick response matrix code image, the processor takes the center point A1 of the image-seeking pattern as the starting point, and moves toward the negative Search in the x-axis direction, determine an outer edge point B1, then use the outer edge point B1 as a starting point, search along the outer edge of the image finding pattern to determine that the image finding pattern is close to the vertices p1, p4 and p3 of the coding area; for Quickly respond to the image-seeking pattern in the upper right corner of the matrix code image, the processor takes the central point A2 of the image-seeking pattern as the starting point, searches in the direction of the negative x-axis, determines an outer edge point B2, and then uses the outer edge point B2 is the starting point, search along the outer edge of the image-finding graph to determine that the image-finding graph is close to the vertices p2, p5 and p6 of the coding area; The center point A3 of the image-seeking pattern is the starting point, search toward the negative x-axis direction, determine an outer edge point B3, and then use the outer edge point B3 as the starting point, search along the outer edge of the image-seeking pattern to determine the Like graph vertices p7, p8 and p10 near the coding region.

In one embodiment, when a plurality of vertices are searched and determined along the outer edge of each image-seeking pattern with the outer edge point of each image-finding pattern as the starting point, the processor uses the outer edge of each image-finding pattern Point is the starting point, and a plurality of vertices are determined by searching along the outer edge of each image-finding graph through an eight-neighborhood contour tracking algorithm.

Among them, the 8-neighborhood edge tracking algorithm is also called 8-neighborhood boundary tracking algorithm, Moore neighborhood tracking algorithm (Moore neighborhood tracking algorithm) and so on. In this embodiment, the search for the aforesaid multiple vertices is achieved by using an 8-neighborhood edge tracking algorithm to track the outer edges of each image-finding pattern, which can reduce the amount of computation required for the search.

In one embodiment, the multiple image-finding patterns include the image-finding pattern in the lower left corner, the image-finding pattern in the upper-left corner, and the image-finding pattern in the upper right corner. The lower right corner vertex of the corner image-seeking figure and the lower-left corner vertex of the upper right corner image-finding figure are searched for an anchor point to determine as a feature point, wherein the anchor point is connected with the upper right corner vertex, the lower right corner vertex and the lower left corner vertex to form a rectangle.

It should be noted that the Quick Response Matrix Code currently includes 40 versions, from 21 symbols*21 symbols (version 1) to 177 symbols*177 symbols (version 40), wherein the latter version of the Quick Response Matrix Code phase Compared with the previous version of the quick response matrix code, 4 symbols are added on each side. For example, the image of the quick response matrix code shown in Figure 2 corresponds to the version 2 of the quick response matrix code, which is 25 symbols * 25 symbols . However, not all versions of the quick response matrix code have correction reference patterns, for example, the version 1 quick response matrix code does not have correction reference patterns. Therefore, when the feature point searching unit 110 searches for the correction reference pattern, the search may be successful or the search may fail.

In this embodiment, when the processor fails to search and correct the reference pattern, it can further search for the image according to the upper right vertex of the image finding pattern in the lower left corner, the lower right vertex of the image finding pattern in the upper left corner, and the upper right corner vertex of the image finding pattern in the upper left corner. For the vertex at the lower left corner of the graph, search for a certain point and determine it as a feature point. Wherein, the anchor point is connected with the upper right vertex, the lower right vertex and the lower left vertex to form a rectangle.

In one embodiment, when dividing a coded area of the quick response matrix code image into multiple sub-areas, the processor takes at least one feature point as a vertex of a sub-area as a constraint condition, and divides the coded area according to multiple feature points for multiple subregions. Here, there is no specific limitation on the way of dividing the sub-regions, which can be configured by those skilled in the art according to actual needs. For example, please refer to FIG. 5, which shows three different division methods. According to the feature points p4, p5, p8 and p9, the processor can divide the middle part of the coding region into a sub-region (division method 1), and can also divide The middle part of the coding region is divided into two sub-regions (division method 2), and the middle part of the coding region can also be divided into four sub-regions (division method 3).

In one embodiment, when determining a compensation vector for each sub-region according to the feature points corresponding to each sub-region, and performing compensation and correction for each sub-region according to the compensation vector for each sub-region, the processor A feature point corresponding to a sub-region, determine a horizontal unit compensation vector of each sub-region in a horizontal direction, a vertical unit compensation vector in a vertical direction; and according to the horizontal unit compensation vector and vertical unit compensation of each sub-region A vector to perform compensation corrections for each subregion.

For example, for a sub-area, the processor divides the coordinate vectors of the two feature points corresponding to the sub-area in the horizontal direction by the number of symbols between the two feature points to obtain a horizontal unit compensation vector, and the sub-area The vertical unit compensation vector is obtained by dividing the coordinate vectors of the two corresponding feature points in the vertical direction by the number of symbols between the two feature points.

In an embodiment, when performing compensation and correction on each sub-region according to the horizontal unit compensation vector and the vertical unit compensation vector of each sub-region, the processor can obtain a matching template corresponding to the QRM code image, and Taking the feature points included in each sub-area as the starting point, according to the level The unit compensation vector and the vertical unit compensation vector search each symbol in the sub-region, and map the searched pixel value of each symbol to a matching template to obtain a corrected image.

Wherein, the processor can identify the version number of the quick response matrix code corresponding to the quick response matrix code image, and then obtain the corresponding matching template according to the version number. For example, please refer to Figure 4, the processor can calculate the number of pixels occupied by the side length of a symbol according to the image-finding pattern, and then calculate the quick response shown in Figure 4 through the coordinates of the center points A1, A2, and A3 of the image-finding pattern The version number of the matrix code is expressed as: V=(((A1A2+A1A3)/2+7)/s-17)/4; wherein, V represents the version number, A1A2 represents the coordinate distance from A1 to A2, and A1A3 represents A1 The coordinate distance to A3, s represents the number of pixels occupied by the side length of a symbol.

In this embodiment, the processor takes the feature points included in each sub-area as the starting point, searches each symbol in the sub-area according to the horizontal unit compensation vector and the vertical unit compensation vector, and converts each code element found The pixel value of the element is mapped to the matching template to obtain the corrected image.

The vector compensation and correction method provided by the present application will be described below by taking the sub-region division method shown in FIG. 6 as an example.

As shown in FIG. 6 , the quick response matrix code image of version 2 has searched out 10 feature points, namely p1, p2, p3, p4, p5, p6, p7, p8, p9 and p10. Among them, use p1, p2, p4, and p5 to divide the middle and upper part of the coding area into four sub-areas, which are respectively sub-area 1, sub-area 2, sub-area 3, and sub-area 4; use p3, p4, p7, and p8 to divide the coding area The middle left part of the area is divided into 4 sub-areas, namely sub-area 5, sub-area 6, sub-area 10, and sub-area 11; use p4, p5, p8, and p9 to divide the central part of the coding area into 4 sub-areas, respectively for subregion 7, subregion 8, subregion 12, and subregion 13; use p5, p6, p8, and p9 to The middle right part of the coding region is divided into two subregions, subregion 9 and subregion 14 respectively; the lower middle part of the coding region is divided into two subregions by using p5, p8, p9 and p10, respectively subregion 15 and subregion 16 ; Use p5, p8 and p9 to divide the lower right part of the coding region into one sub-region, ie sub-region 17.

其中，

表示子區域7的水平單位補償向量，

表示子區域 7的垂直單位補償向量，px表示特徵點的橫坐標，py表示特徵點的縱坐標。 The vector compensation of sub-area 7 is described below as an example: First, the horizontal unit compensation vector and vertical unit compensation vector of sub-area 7 are calculated, wherein the feature point corresponding to sub-area 7 in the horizontal direction (that is, the x-axis direction) is p4 and p5, there are 11 code elements between p4 and p5, and the corresponding feature points in the vertical direction (that is, the y-axis direction) are p4 and p8, and there are 11 code elements between p4 and p8, which can be calculated according to the following formula Horizontal unit compensation vector and vertical unit compensation vector of sub-area 7:

in,

represents the horizontal unit compensation vector for subregion 7,

Represents the vertical unit compensation vector of sub-region 7, px represents the abscissa of the feature point, and py represents the ordinate of the feature point.

和垂直單位補償向 量

搜索水平方向和垂直方向上的第一個碼元spt₇，可以表示為：

如上搜索到spt₇的座標後，將spt₇對應的像素值映射至版本2的匹配模板中，之後繼續搜索子區域7中的其它碼元ept₇，可以表示為：

其中，i₇

[1,N₇-1]，j₇

[1,M₇-1]，N₇表示子區域7在水平方向的碼元數量，M₇表示子區域7在垂直方向的碼元數量。 Take p4 as the starting point, and compensate the vector according to the horizontal unit

and the vertical unit compensation vector

Search for the first symbol spt ₇ in the horizontal and vertical directions, which can be expressed as:

After searching the coordinates of spt ₇ as above, map the pixel value corresponding to spt ₇ to the matching template of version 2, and then continue to search for other code elements ept ₇ in sub-region 7, which can be expressed as:

Among them, i ₇

[1, N ₇ -1], j ₇

[1, M ₇ -1], N ₇ represents the number of symbols in the sub-region 7 in the horizontal direction, and M ₇ represents the number of symbols in the sub-region 7 in the vertical direction.

As above, after each search of ept ₇ , the pixel value corresponding to ept ₇ is mapped to the matching template of version 2 until the correction of all symbols in sub-region 7 is completed.

其中，

除了作為子區域7的垂直單位補償向量之外，還作為 子區域12的垂直單位補償向量，此外，按照如下公式計算得到子區域12的水平單位補償向量：

其中，

表示子區域12的水平單位補償向量。 The following explains the vector compensation of sub-area 12 as an example: similar to sub-area 7, the difference is that the horizontal unit compensation vector cannot be directly determined according to p8 and p9, but the new feature points p11 and p11 geometrically corresponding to P9 need to be determined first It can be understood as the center point of the block where P8 is located, and can be searched through p4, p7 and p8, expressed as:

in,

In addition to being the vertical unit compensation vector of sub-area 7, it is also used as the vertical unit compensation vector of sub-area 12. In addition, the horizontal unit compensation vector of sub-area 12 is calculated according to the following formula:

in,

represents the horizontal unit compensation vector for the subregion 12.

和垂直單位補償向 量

搜索水平方向和垂直方向上的第一個碼元spt₁₂，可以表示為：

如上搜索到spt₁₂的座標後，將spt₁₂對應的像素值映射至版本2的匹配模板中，之後繼續搜索子區域12中的其它碼元ept₁₂，可以表示為：

其中，i₁₂

[1,N₁₂-1]，j₁₂

[1,M₁₂-1]，N₁₂表示子區域12在水平方向的碼元數量，M₁₂表示子區域12在垂直方向的碼元數量。 Take p8 as the starting point, and compensate the vector according to the horizontal unit

and the vertical unit compensation vector

Search for the first symbol spt ₁₂ in the horizontal and vertical directions, which can be expressed as:

After searching for the coordinates of spt ₁₂ as above, map the pixel value corresponding to spt ₁₂ to the matching template of version 2, and then continue to search for other code elements ept ₁₂ in sub-region 12, which can be expressed as:

Among them, i ₁₂

[1, N ₁₂ -1], j ₁₂

[1, M ₁₂ −1], N ₁₂ represents the number of symbols in the sub-region 12 in the horizontal direction, and M ₁₂ represents the number of symbols in the sub-region 12 in the vertical direction.

As above, each time ept ₁₂ is found, the pixel value corresponding to ept ₁₂ is mapped to the matching template of version 2 until the correction of all symbols in the sub-region 12 is completed.

The following explains the vector compensation of sub-area 17 as an example: similar to sub-area 12, the horizontal unit compensation vector cannot be directly determined according to the feature points p8 and p9 corresponding to sub-area 17 in the horizontal direction, nor can it be determined according to the vertical position of sub-area 17 The feature points p5 and p9 corresponding to the direction determine the vertical unit compensation vector, but it is necessary to first determine the new feature point p11 corresponding to the P9 geometry in the horizontal direction, and determine the new feature point p12 corresponding to the P9 geometry in the vertical direction.

由於p11和p9之間有12個碼元，p9和p12之間也有12個碼元，按照如下公式計算得到子區域17的水平單位補償向量和垂直單位補償向量：

其中，

表示子區域17的水平單位補償向量，

表示子區 域17的垂直單位補償向量。 p12 can be understood as the center point of the block where P5 is located, and can be searched through p2, p5 and p6, expressed as:

Since there are 12 symbols between p11 and p9, and there are also 12 symbols between p9 and p12, the horizontal unit compensation vector and the vertical unit compensation vector of sub-region 17 are calculated according to the following formula:

in,

represents the horizontal unit compensation vector of the sub-region 17,

Denotes the vertical unit compensation vector for subregion 17.

其中，i₁₇

[1,N₁₇-1]，j₁₇

[1,M₁₇-1]，N₁₇表示子區域17在水平方向的碼元數量，M₁₇表示子區域17在垂直方向的碼元數量。 Take p9 as the first symbol, map the pixel value corresponding to p9 to the matching template of version 2, and then continue to search for other symbols ept ₁₇ in sub-region 17, which can be expressed as:

Among them, i ₁₇

[1, N ₁₇ -1], j ₁₇

[1, M ₁₇ -1], N ₁₇ represents the number of symbols in the sub-region 17 in the horizontal direction, and M ₁₇ represents the number of symbols in the sub-region 17 in the vertical direction.

As above, each time ept ₁₇ is found, the pixel value corresponding to ept ₁₇ is mapped to the matching template of version 2 until the correction of all symbols in the sub-region 17 is completed.

As above, specific vector compensation methods for different types of sub-regions are shown. For other sub-regions, reference may be made to the above-mentioned vector compensation methods for corresponding types of sub-regions, and details will not be repeated here.

It should be noted that since the position of the image finding pattern is fixed, the corresponding pixel values of the image finding pattern can be directly filled in the matching template, and finally a complete corrected image can be obtained.

Please refer to FIG. 7 , the present application also provides an electronic device 01, including a chip 10, a memory 20, and a camera 30, wherein the camera 30 includes at least a lens and an image sensor, wherein the lens is used to capture external The light signal is projected to the image sensor, and the image sensor is used for photoelectric conversion of the light signal projected by the lens, converting the light signal into a usable electrical signal, and obtaining a digital image. The camera 30 can be used to capture the quick response matrix code image obtained by shooting any quick response matrix code.

The chip 10 includes an interface circuit 200 and a processor 100 . The interface circuit 200 is used to acquire the quick response matrix code image captured by the camera 30 and may be a Mobile Industry Processor Interface (MIPI). The processor 100 can implement the image correction method of the present invention by executing the computer program in the memory 20, divide the encoding area of the QRC image acquired by the interface circuit 200 into a plurality of sub-areas for compensation and correction, and obtain Corrected image.

The memory 20 can be a high-speed random access memory, or a non-volatile memory body, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device.

Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed through computer programs or instructions, or through instructions to control related hardware. The computer programs or instructions can be stored in a stored in a computer-readable storage medium, and loaded and executed by a chip.

The image correction method and the processor provided in the embodiments of the present application have been introduced in detail above. In this paper, specific examples are used to illustrate the principles and implementation methods of the present application, and the descriptions of the above embodiments are only used to help understand the present application. At the same time, for those skilled in the art, based on the idea of this application, there will be changes in the specific implementation and application scope. In summary, the content of this specification should not be construed as limiting the application.

S110, S120, S130: steps

Claims (11)

An image correction method, comprising: Carry out feature point search to a quick response matrix code image, determine a plurality of feature points in the described quick response matrix code image; According to the plurality of feature points, a coding area of the Quick Response Matrix Code image is divided into a plurality of sub-areas; and Determine a compensation vector for each sub-area according to the feature points corresponding to each of the sub-areas, and perform compensation and correction for each of the sub-areas according to the compensation vector for each of the sub-areas to obtain a Corrected image. The image correction method as described in claim 1, wherein the feature point search is performed on a quick response matrix code image to determine a plurality of feature points in the quick response matrix code image, including: searching for a plurality of finding patterns in the quick response matrix code image; and The multiple feature points are determined by searching according to the multiple imaging patterns. The image correction method according to claim 2, wherein, before searching and determining the plurality of feature points according to the plurality of imaging patterns, further includes: decoding the Quick Response Matrix code image according to the plurality of finding patterns; and When the decoding of the quick response matrix code image fails, the multiple feature points are searched and determined according to the multiple image finding patterns. The image correction method according to claim 2, wherein said searching and determining said multiple feature points according to said multiple imaging patterns includes: A plurality of vertices of each of the image finding patterns close to the coding area are searched for the feature points, and a center point of a calibration reference pattern in the QRM code image is searched for the feature points. The image correction method according to claim 4, wherein said searching for multiple vertices of each said imaging pattern close to said encoding area to determine as said feature points includes: Taking the central point of each image-finding pattern as a starting point, search and determine an outer edge point of each image-finding pattern along a search direction, and use the outer edge point of each image-finding pattern As a starting point, the plurality of vertices are determined by searching along the outer edge of each imaging pattern. The image correction method according to claim 5, wherein, starting from the outer edge point of each image-finding pattern, searching along the outer edge of each image-finding pattern to determine the multiple vertices, including: Taking the outer edge point of each image-finding pattern as a starting point, the plurality of vertices are determined by searching along the outer edge of each image-finding pattern through an 8-neighborhood edge tracking algorithm. The image correction method according to claim 5, wherein the plurality of image-finding patterns include a bottom-left image-finding pattern, an upper-left image-finding pattern, and an upper-right image-finding pattern, and the image correction method further includes: When searching for the correction reference pattern fails, search for a location according to the upper right vertex of the image finding pattern in the lower left corner, the lower right vertex of the image finding pattern in the upper left corner, and the lower left vertex of the image finding pattern in the upper right corner. A point is determined as the feature point, wherein the anchor point is connected with the upper right vertex, the lower right vertex, and the lower left vertex to form a rectangle. The image correction method according to claim 1, wherein, according to the plurality of feature points, dividing a coding area of the Quick Response Matrix Code image into multiple sub-areas, including: Taking at least one feature point as a vertex of a sub-area as a constraint condition, the coding area is divided into a plurality of sub-areas according to the plurality of feature points. The image correction method according to claim 8, wherein, according to the feature points corresponding to each of the sub-regions, a compensation vector for each of the sub-regions is determined, and according to the corresponding feature points of each of the sub-regions The above compensation vector performs compensation and correction for each of the sub-regions, including: determining a horizontal unit compensation vector in a horizontal direction and a vertical unit compensation vector in a vertical direction for each of the sub-regions according to the feature points corresponding to each of the sub-regions; and Compensation and correction are performed on each of the sub-areas according to the horizontal unit compensation vector and the vertical unit compensation vector of each of the sub-areas. The image correction method according to claim 9, wherein, performing compensation and correction on each of the sub-regions according to the horizontal unit compensation vector and the vertical unit compensation vector of each of the sub-regions includes : obtaining a matching template corresponding to the Quick Response Matrix Code image; and Taking the feature point included in each sub-area as a starting point, according to the horizontal unit compensation vector and the vertical unit compensation vector, search for each symbol in the sub-area, and use the searched The pixel value of each symbol is mapped to the matching template to obtain the corrected image. A processor, the processor executes a computer program to implement an image correction method, wherein the image correction method comprises: Carry out feature point search to a quick response matrix code image, determine a plurality of feature points in the described quick response matrix code image; According to the plurality of feature points, a coding area of the Quick Response Matrix Code image is divided into a plurality of sub-areas; and Determine a compensation vector for each sub-area according to the feature points corresponding to each of the sub-areas, and perform compensation and correction for each of the sub-areas according to the compensation vector for each of the sub-areas to obtain a Corrected image.

Images