A method and system for detecting map print quality

Abstract

The present invention belongs to the field of image processing technology, and specifically relates to a method and system for detecting the quality of map prints. The method comprises: partitioning all grayscale images; obtaining a first completeness of each region in the target grayscale image based on the differential representation of each region in the target grayscale image and the corresponding region in the standard grayscale image; obtaining a second completeness of each region in the target grayscale image based on the presentation of the associated map structure between each region in the target grayscale image and neighboring regions; obtaining the overall completeness of each region in the target grayscale image; recording the central defect region and the corresponding radial defect region in the target grayscale image as a defect region; and calculating a defect index for the defect region and using it to evaluate the quality of map prints. The present invention improves the accuracy and speed of map print quality detection.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and more particularly to a method and system for detecting the quality of map prints. Background Art

[0002] During the map printing process, defects such as missing prints and speckles may occur due to factors such as the printing process, mechanical precision, and the testing environment. Therefore, printed map products require inspection. Currently, machine vision technology is commonly used to inspect printed products using a reference method. The process involves selecting a standard print template, capturing a print image, pre-processing the print image, and then performing image subtraction with the standard print template. The subtraction result is then used to determine the print quality.

[0003] In related technologies, for example, Chinese patent document CN109840499B discloses a method for rapidly detecting the printing and binding quality of printed materials. This method uses a feature matching algorithm based on dynamic time warping to measure the similarity between a test image and a reference image to determine the quality of the printed material. Chinese patent document CN119130972B discloses an online method and system for detecting printed material quality. This method detects defects in printed materials by setting different scanning frame sizes and using a multi-level scanning method. Subregions of the test image are compared one-to-one with subregions of a standard image, enabling more detailed and accurate classification.

[0004] However, in the process of quality inspection of printed materials, since map products have a higher amount of information than general printed materials, taking physical geographical maps as an example, different administrative regions are distinguished by lines and color blocks. There are natural geographical features such as mountains and rivers, as well as place names in each administrative region. Therefore, map prints require higher detection accuracy, and defects will cause collateral interference, amplifying the actual impact of the defects, resulting in poor results in the current quality inspection of map prints. Summary of the Invention

[0005] In order to solve the above-mentioned technical problem of poor detection effect when using the reference method to detect map prints, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a method for detecting the quality of a printed map, comprising:

[0007] Several images of a map to be tested are obtained. Standard images are manually screened and preprocessed for all images of the maps to be tested. These images are then registered to the same coordinate system to obtain several grayscale images to be tested and standard grayscale images, collectively referred to as grayscale images. Any grayscale image to be tested is recorded as a target grayscale image. All grayscale images are partitioned based on an edge detection algorithm and grayscale distribution to obtain several regions of any grayscale image. A first completeness of each region in the target grayscale image is obtained based on the differential representation of each region in the target grayscale image and the corresponding region in the standard grayscale image. A second completeness of each region in the target grayscale image is obtained based on the representation of the associated map structure of each region in the target grayscale image and its neighboring regions in the standard grayscale image. The first completeness is combined with the second completeness to obtain the overall completeness of each region in the target grayscale image. Based on the variation pattern of the overall completeness between regions in the target grayscale image, a central defect region and a corresponding radial defect region in the target grayscale image are obtained, which are collectively recorded as a defect region. A defect index of any defect region is calculated based on the overall completeness of the regions contained in the defect region and used to evaluate the quality of the map print.

[0008] This invention divides printed maps into regional shapes based on map characteristics, enabling rapid location of defective areas during quality inspection, improving the efficiency of quality inspection of printed maps. This invention determines the overall integrity of a region based on its shape and associated map structure, allowing for more precise quality assessment of each map region, thus making quality inspection of printed maps more accurate and effective.

[0009] Preferably, the step of obtaining a plurality of regions of an arbitrary grayscale image includes:

[0010] Perform edge detection on all grayscale images to obtain the corresponding edge image of each grayscale image;

[0011] The i-th edge pixel of the edge image of the target grayscale image is recorded as , get At the left neighboring A pixels and the right neighboring A pixels in the normal direction of the edge, a first threshold is preset, and the difference between the grayscale mean of the left neighboring A pixels and the grayscale mean of the right neighboring A pixels is recorded as ,like is greater than the first threshold, then Recorded as boundary pixel point; if is less than or equal to the first threshold, then Recorded as feature pixel points;

[0012] Any grayscale image is divided into a number of closed partitions by boundary pixels, and any closed partition is recorded as a region, thereby obtaining a number of regions of the grayscale image.

[0013] The present invention effectively divides the regions according to the regional color block differences and edge pixel points of the map, thereby ensuring the integrity of the map region on the basis of reducing the quality detection range and improving the quality detection effect.

[0014] Preferably, obtaining the first completeness of each region in the target grayscale image includes:

[0015] Any area of ​​the target grayscale image is recorded as the target area, and the grayscale matrix of the target area is subtracted from the grayscale matrix of the corresponding area in the standard grayscale image to obtain the differential grayscale matrix of the target area;

[0016] The first integrity of the target area satisfies the expression:

[0017] ;

[0018] Where, Indicates the first degree of completeness of the target area; The DTW distance between the chain code of the target area and the chain code of the corresponding area of ​​the target area in the standard grayscale image; Represents the average value of the differential grayscale matrix of the target area; Indicates the number of values ​​greater than 0 in the differential grayscale matrix of the target area.

[0019] The present invention compares the target area with the corresponding area in the standard grayscale image based on the difference technology. The difference technology shows that the larger the difference result value is, the greater the difference between the two is, thereby reflecting that the first integrity of the target area is lower.

[0020] Preferably, obtaining the chain code of the target area includes:

[0021] Taking the pixel point closest to the origin of the coordinate system in the target area as the starting point, the eight-direction chain code of the boundary pixel points of the target area is obtained in a clockwise direction.

[0022] Preferably, obtaining the second completeness of each region in the target grayscale image includes:

[0023] Get the natural features of any grayscale image; take the pixel point closest to the origin of the coordinate system as the starting point, and get the eight-direction chain code of the natural feature in clockwise direction; record the vth region to which the uth natural feature of the standard grayscale image belongs as , get the target grayscale image with The corresponding area is recorded as the target grayscale image about control area;

[0024] Calculate the target grayscale image about the second completeness of the control area of ​​the target area; if the target area is the control area, the second completeness of the target grayscale image with respect to the control area of ​​the area to which it belongs is recorded as the second completeness of the target area; if the target area is not the control area, the second completeness of the target area is recorded as 1.

[0025] The present invention takes natural features as key areas and calculates the second integrity of each area, thereby ensuring the quality detection accuracy of map prints and improving the accuracy of map print quality detection from the perspective of being beneficial to practical use.

[0026] Preferably, the obtaining of natural features of any grayscale image includes:

[0027] For any grayscale image, boundary pixels adjacent to feature pixels are obtained and recorded as equivalent feature pixels. Any feature pixel is recorded as a growth seed, and regional growth is performed. The growth condition is that there are feature pixels or equivalent feature pixels between adjacent pixels. The growth stop condition is that there are no adjacent pixels and no equivalent feature pixels between adjacent pixels. After the growth stops, if there are still feature pixels that have not grown, growth seeds are selected from the feature pixels that have not grown until all feature pixels have their own growth regions. Any growth region is recorded as a natural feature, and several natural features of the grayscale image are obtained.

[0028] Preferably, the target grayscale image is about The second integrity of the control area satisfies the expression:

[0029] ;

[0030] Where, Represents the target grayscale image about The second completeness of the control area; Represents the target grayscale image about The number of pixels in the control area that belong to natural features; express The number of pixels belonging to natural features in ; Represents the target grayscale image about the number of natural features in the control area; Represents the target grayscale image about The chain code of the rth natural feature of the control area is DTW distance of the chain code of natural features in .

[0031] Preferably, obtaining the overall integrity of each region in the target grayscale image includes:

[0032] The first integrity of the target area and the second integrity of the target area are added together and normalized for positive correlation to obtain the overall integrity of the target area.

[0033] Preferably, the defect index of any defect area satisfies the expression:

[0034] ;

[0035] Where, represents the defect index of the h-th defect area; represents the number of regions of the h-th defective region; The number of pixels in the sth region representing the hth defect area; represents the overall integrity of the sth region of the hth defect region.

[0036] The present invention obtains the defect index of the defect area in a relatively comprehensive manner based on the area of ​​the defect area and the overall integrity of each area, making it possible to judge the map quality from multiple dimensions and further improving the accuracy of map print quality detection.

[0037] In a second aspect, the present invention provides a map print quality detection system, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned map print quality detection method is implemented.

[0038] By adopting the above technical solution, the above-mentioned map print quality detection method is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is manufactured based on the memory and the processor for easy use.

[0039] The beneficial effects of the present invention are:

[0040] (1) Based on the image characteristics of maps, the present invention analyzes the quality of maps from the perspective of the shape and natural geographical features of map prints, thereby improving the accuracy of quality detection of map prints;

[0041] (2) The present invention also divides the map grayscale image into regions, so that when defects are found, the defect area can be quickly located and referenced, thereby improving the quality inspection efficiency of map prints;

[0042] (3) The present invention merges the areas to which a defect belongs, takes into account the associated interference, and further improves the accuracy of quality detection of map prints. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart schematically illustrating a method for detecting the quality of map prints in the present invention;

[0044] Figure 2 is a schematic diagram schematically illustrating boundary pixels and feature pixels. DETAILED DESCRIPTION

[0045] The embodiment of the present invention discloses a method for detecting the quality of map prints, referring to Figure 1 , including steps S1 to S4:

[0046] S1: Obtain images of several maps to be tested of a map, manually screen standard images, preprocess all images of the maps to be tested, and align them to obtain several grayscale images to be tested and standard grayscale images, collectively referred to as grayscale images; any grayscale image to be tested is recorded as a target grayscale image.

[0047] It should be noted that the map printing environment is relatively complex and differs greatly from the map templates in the computer environment. Therefore, in order to more accurately obtain the quality of map prints, the present invention manually screens the images of all maps to be tested to obtain map detection results that are more in line with the real environment.

[0048] Specifically, an industrial camera is used to capture images of all printed copies of a map. Each image is recorded as an image of the map to be tested, and then a standard image is manually selected. It should be noted that the standard image maximizes the presentation of map details and eliminates areas such as missing parts and shadows that would affect visual recognition.

[0049] All images of the maps to be tested are preprocessed, including grayscale conversion and geometric correction. Grayscale conversion can maximize color preservation through weighted averaging. For example, the weights of the RGB space can be set to 0.3, 0.6, and 0.1. The geometric correction method can use the SIFT algorithm to perform feature matching between the image of the map to be tested and the standard image, placing all matching points in the same coordinate system to complete image registration of the map to be tested, and obtain several grayscale images to be tested and standard grayscale images, collectively referred to as grayscale images. The origin of the coordinate system is the first pixel in the lower left corner of the standard image, and the positive directions of the horizontal and vertical axes of the coordinate system are to the right and above the first pixel in the lower left corner of the standard image, respectively.

[0050] At this point, a grayscale image set consisting of several grayscale images to be tested and a standard grayscale image is obtained.

[0051] S2: Based on the edge detection algorithm and grayscale distribution, all grayscale images are partitioned to obtain several regions of any grayscale image; according to the differential performance of each region in the target grayscale image and the corresponding region in the standard grayscale image, the first completeness of each region in the target grayscale image is obtained.

[0052] It should be noted that the defects of map prints have a greater impact than those of text prints. For example, if ink marks appear at the boundaries of regions, the viewing of all adjacent regions will be affected. Therefore, considering the uniqueness of map prints, the present invention divides the map according to regions, thereby establishing a map print quality detection model from a more precise dimension, quickly locating defects and determining the impact caused.

[0053] It should be further explained that by comparing the area of ​​the target grayscale image with the corresponding area in the standard grayscale image, the relative difference of the area of ​​the target grayscale image can be obtained. The function of the map print is to accurately display the shape of the region and the natural geographical features. Therefore, the difference in non-feature pixel points will not affect the quality of the map print. Therefore, the present invention first determines the characteristic performance of the pixel points with large relative differences between the area of ​​the target grayscale image and the corresponding area in the standard grayscale image, and then calculates the first completeness of each area in the target grayscale image.

[0054] Specifically, based on the edge detection algorithm and grayscale distribution, all grayscale images are partitioned to obtain several regions of any grayscale image, including:

[0055] Perform edge detection on all grayscale images to obtain the corresponding edge image of each grayscale image.

[0056] It should be noted that in the corresponding edge image of the grayscale image, edges are divided into boundary edges and feature edges. Boundary edges are used to distinguish different regions, and feature edges are used to represent natural features such as rivers. Rivers flow through many places, so feature edges will intersect with multiple boundary edges. Partitioning based solely on edge detection results cannot guarantee that each partition is a complete region, and the map color blocks of adjacent regions are different. Therefore, the grayscale distribution can be combined to divide the region.

[0057] The i-th edge pixel of the edge image of the target grayscale image is recorded as , get At the left neighboring A pixels and the right neighboring A pixels in the normal direction of the edge, a first threshold is preset, and the difference between the grayscale mean of the left neighboring A pixels and the grayscale mean of the right neighboring A pixels is recorded as ,like is greater than the first threshold, then Recorded as boundary pixel point; if is less than or equal to the first threshold, then Recorded as feature pixel points. A is a preset value, which can be set to 5, and the first threshold can be set to 10. It should be noted that boundary pixels are pixels that distinguish different areas of the map, and feature pixels are used to express natural features. Figure 2 Schematic diagram of boundary pixels and feature pixels.

[0058] Any grayscale image is divided into a number of closed partitions by boundary pixels, and any closed partition is recorded as a region, thereby obtaining a number of regions of the grayscale image.

[0059] So far, several regions of each grayscale image have been obtained.

[0060] What is needed is to differentiate the grayscale matrix of the area of ​​the target grayscale image from the corresponding area in the standard grayscale image. When the shape of the area of ​​the target grayscale image is complete and there are no defects, the differential result should be a zero matrix. The more serious the defects of the area of ​​the target grayscale image, the more there are in the matrix of the differential result. When there are interfering defects in the area of ​​the target grayscale image, the data in the matrix of the differential result will also be regularly distributed. Therefore, the present invention obtains the first completeness of each area in the target grayscale image based on the differential performance and data distribution of each area in the target grayscale image and the corresponding area in the standard grayscale image.

[0061] Preferably, obtaining the first completeness of each region in the target grayscale image according to the differential expression between each region in the target grayscale image and the corresponding region in the standard grayscale image includes:

[0062] Any area of ​​the target grayscale image is recorded as the target area, and the grayscale matrix of the target area is subtracted from the grayscale matrix of the corresponding area in the standard grayscale image to obtain the differential grayscale matrix of the target area.

[0063] Obtaining the chain code for any region: Taking the target region as an example, starting from the pixel closest to the origin of the coordinate system, obtain the chain codes for the boundary pixels of the target region in eight directions in a clockwise direction. It should be noted that the chain code is a numerical sequence that describes the shape; similar chain codes indicate similar shapes.

[0064] The first integrity of the target area satisfies the expression:

[0065] ;

[0066] Where, Indicates the first degree of completeness of the target area; The DTW distance between the chain code of the target area and the chain code of the corresponding area of ​​the target area in the standard grayscale image; Represents the average value of the differential grayscale matrix of the target area; Indicates the number of values ​​greater than 0 in the differential grayscale matrix of the target area. It should be noted that the DTW distance is an indicator used in the dynamic time warping algorithm to evaluate the similarity between two sequences.

[0067] Where, Indicates the proportion of each value greater than 0 in the average value of the differential grayscale matrix of the target area. The larger the value, the greater the average value and the fewer the values ​​greater than 0. In this case, the defects in the target area are more concentrated and more noticeable to the human eye, thus affecting map viewing, indicating a lower first integrity of the target area. Indicates the difference between the shape of the target area and the shape of the corresponding area of ​​the target area in the standard grayscale map. The larger the value, the greater the shape difference, which means that the target area is difficult to meet the map's description of the shape of the relevant area, and thus the first completeness of the target area is lower.

[0068] At this point, the first completeness of each region is obtained.

[0069] S3: Obtain a second completeness of each region in the target grayscale image based on the associated map structure of each region and the neighboring region in the target grayscale image and their presentation in the standard grayscale image; and combine the first completeness with the second completeness to obtain an overall completeness of each region in the target grayscale image.

[0070] It should be noted that since the pixels of natural features such as rivers pass through multiple regions, if there is a defect in one region, it will lead to an error in the overall perception of the natural feature, thereby affecting the integrity of the natural feature in other regions. Therefore, the present invention analyzes the defect status of all pixels of each natural feature and obtains the second integrity of all regions it passes through. Finally, the first integrity and second integrity of the region are combined to obtain the overall integrity of each region, where the first integrity is used to describe the shape and grayscale integrity of the region, and the second integrity is used to describe the integrity of the key natural features in the region.

[0071] Specifically, according to the associated map structure of each region in the target grayscale image and the neighboring regions, and the presentation of the structure in the standard grayscale image, the second completeness of each region in the target grayscale image is obtained, including:

[0072] For any grayscale image, boundary pixels adjacent to feature pixels are obtained and recorded as equivalent feature pixels. Any feature pixel is recorded as a growth seed, and regional growth is performed. The growth condition is that there are feature pixels or equivalent feature pixels between adjacent pixels. The growth stop condition is that there are no adjacent pixels and no equivalent feature pixels between adjacent pixels. After the growth stops, if there are still feature pixels that have not grown, growth seeds are selected from the feature pixels that have not grown until all feature pixels have their own growth regions. Any growth region is recorded as a natural feature, and several natural features of the grayscale image are obtained.

[0073] The vth region to which the uth natural feature of the standard grayscale image belongs is recorded as , get the target grayscale image with The corresponding area is recorded as the target grayscale image about The control area of ​​​​the target grayscale image is obtained The natural characteristics of the control area.

[0074] It should be noted that the u-th natural feature of the standard grayscale image completely records the direction of the natural feature, while if the natural feature of the control area has defects, there will be disconnected and redundant pixels. Therefore, the completeness of the natural feature of the control area can be obtained based on the difference in the direction and number of pixels between the u-th natural feature of the standard grayscale image and the natural feature of the control area.

[0075] Obtain chain codes for all natural features: Take the pixel point closest to the origin of the coordinate system as the starting point, and obtain the eight-direction chain codes of the natural feature in a clockwise direction.

[0076] Target grayscale image about The second integrity of the control area satisfies the expression:

[0077] ;

[0078] Where, Represents the target grayscale image about The second completeness of the control area; Represents the target grayscale image about The number of pixels in the control area that belong to natural features; express The number of pixels belonging to natural features in ; Represents the target grayscale image about the number of natural features in the control area; Represents the target grayscale image about The chain code of the rth natural feature of the control area is DTW distance of the chain code of natural features in .

[0079] Where, Represents the target grayscale image about The difference between the number of pixels belonging to natural features in the control area and the number of pixels belonging to natural features in the standard area, the larger the value, the better the target grayscale image is. The worse and more incomplete the natural features of the control area are, the more the target grayscale image is about The lower the second integrity of the control area; Indicates that the target grayscale image is about In the case where the natural features in the control area are split or redundant, each natural feature is The difference in the direction of the natural features in the image. The larger the value, the more the target grayscale image is about The more the direction of the natural features in the control area does not conform to the standard situation, the more the target grayscale image The second integrity of the control area is lower.

[0080] If the target area is the control area, the second completeness of the target grayscale image with respect to the control area of ​​the area to which it belongs is recorded as the second completeness of the target area; if the target area is not the control area, the second completeness of the target area is recorded as 1.

[0081] At this point, the second completeness of each area is obtained.

[0082] Preferably, the first completeness and the second completeness are combined to obtain the overall completeness of each area in the target grayscale image, including: adding the first completeness of the target area and the second completeness of the target area, and performing positive correlation normalization to obtain the overall completeness of the target area.

[0083] At this point, the overall completeness of each area is obtained.

[0084] S4: According to the variation pattern of the overall integrity between regions in the target grayscale image, the central defect area of ​​the target grayscale image and the corresponding radiation defect area are obtained and recorded as one defect area; according to the overall integrity of the area contained in the defect area, the defect index of any defect area is calculated and used to evaluate the quality of the map print.

[0085] It should be noted that due to the irregular distribution of defects, a defect may be distributed in several adjacent areas, including the central defect area of ​​the defect source with the lowest overall integrity, and several defect areas radiating outward with gradually decreasing overall integrity. By merging these defect areas, all defect areas can be located to avoid omissions.

[0086] Specifically, according to the change law of the overall integrity between regions in the target grayscale image, the central defect area of ​​the target grayscale image and the corresponding radiation defect area are obtained, and are recorded as one defect area, including:

[0087] Regions with an overall integrity less than the second threshold are designated as defective partitions. The minimum overall integrity value of all regions is obtained, each designated as a central defective region. A breadth-first traversal of the central defective region is performed until no defective partitions remain in its neighborhood. All traversed regions are designated as radial defective regions corresponding to the central defective region. The central defective region and all radial defective regions are merged into a single defective region. It should be noted that the second threshold is set by the implementer based on actual implementation circumstances; for example, the second threshold can be set to 0.8.

[0088] It should be noted that the larger the defect area is and the lower the overall integrity of the defect partitions it contains, the higher the defect index of the defect area is. Therefore, the defect index of any defect area is calculated.

[0089] Preferably, the defect index of any defect area satisfies the expression:

[0090] ;

[0091] Where, represents the defect index of the h-th defect area; represents the number of regions of the h-th defective region; The number of pixels in the sth region representing the hth defect area; represents the overall integrity of the sth region of the hth defect region.

[0092] Where, Indicates the number of pixels in the h-th defect area, indicating the area of ​​the h-th defect area. The larger the value, the larger the area of ​​the h-th defect area. The area of ​​the h-th defect area has a greater impact on the quality of the map print, indicating that the defect index of the h-th defect area is higher. It represents the sum of the overall integrity of all areas of the h-th defect area. The larger the value, the more serious the defect of the h-th defect area, and thus the higher the defect index of the h-th defect area.

[0093] At this point, the defect index of each defect area is obtained.

[0094] Preferably, the quality of the map print is assessed, including:

[0095] The defect index of all defective areas in all grayscale images to be tested is obtained through the method in steps S1-S3, and a third threshold is set. If a defective area in the grayscale image to be tested has a defect index greater than the third threshold, the quality of the map print corresponding to the grayscale image to be tested is marked as unqualified. It should be noted that the third threshold is set by the implementer based on the actual accuracy requirements of the map print; for example, the third threshold can be set to 0.3.

[0096] At this point, the quality inspection of map prints has been completed.

[0097] An embodiment of the present invention further discloses a map print quality detection system, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a map print quality detection method according to the present invention is implemented.

[0098] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0099] While several embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

Claims

1. A method for detecting the quality of map prints, characterized in that: include: Obtain images of several maps to be tested of a map, manually screen standard images, pre-process all images of the maps to be tested, and align them to the same coordinate system to obtain several grayscale images to be tested and standard grayscale images, collectively referred to as grayscale images; Any grayscale image to be tested is recorded as the target grayscale image; Based on the edge detection algorithm and grayscale distribution, all grayscale images are partitioned to obtain several regions of any grayscale image; Obtaining a first completeness of each region in the target grayscale image according to a differential representation of each region in the target grayscale image and a corresponding region in the standard grayscale image; Get several regions of any grayscale image, including: Perform edge detection on all grayscale images to obtain the corresponding edge images of each grayscale image; the i-th edge pixel point of the edge image of the target grayscale image is recorded as , get At the left neighboring A pixels and the right neighboring A pixels in the normal direction of the edge, a first threshold is preset, and the difference between the grayscale mean of the left neighboring A pixels and the grayscale mean of the right neighboring A pixels is recorded as ,like is greater than the first threshold, then Recorded as boundary pixel point; if is less than or equal to the first threshold, then Recorded as feature pixel points; any grayscale image is divided into several closed partitions by boundary pixels, and any closed partition is recorded as a region, thereby obtaining several regions of the grayscale image; Obtaining a second completeness of each region in the target grayscale image based on the presentation of the associated map structure of each region and the neighboring regions in the target grayscale image in the standard grayscale image; and combining the first completeness with the second completeness to obtain an overall completeness of each region in the target grayscale image; According to the variation pattern of the overall integrity between regions in the target grayscale image, the central defect area of ​​the target grayscale image and the corresponding radiation defect area are obtained and recorded as one defect area. According to the overall integrity of the areas contained in the defect area, the defect index of any defect area is calculated and used to evaluate the quality of map prints.

2. A method for detecting the quality of map prints according to claim 1, characterized in that: The obtaining of the first completeness of each region in the target grayscale image includes: Any area of ​​the target grayscale image is recorded as the target area, and the grayscale matrix of the target area is subtracted from the grayscale matrix of the corresponding area in the standard grayscale image to obtain the differential grayscale matrix of the target area; The first integrity of the target area satisfies the expression: ； Where, Indicates the first degree of completeness of the target area; The DTW distance between the chain code of the target area and the chain code of the corresponding area of ​​the target area in the standard grayscale image; Represents the average value of the differential grayscale matrix of the target area; Indicates the number of values ​​greater than 0 in the differential grayscale matrix of the target area.

3. A method for detecting the quality of map prints according to claim 2, characterized in that: Acquisition of the chain code of the target area includes: Taking the pixel point closest to the origin of the coordinate system in the target area as the starting point, the eight-direction chain code of the boundary pixel points of the target area is obtained in a clockwise direction.

4. A method for detecting the quality of map prints according to claim 1, characterized in that: The obtaining of the second completeness of each region in the target grayscale image includes: Get the natural features of any grayscale image; take the pixel point closest to the origin of the coordinate system as the starting point, and get the eight-direction chain code of the natural feature in clockwise direction; record the vth region to which the uth natural feature of the standard grayscale image belongs as , get the target grayscale image with The corresponding area is recorded as the target grayscale image about control area; Calculate the target grayscale image about the second completeness of the control area of ​​the target area; if the target area is the control area, the second completeness of the target grayscale image with respect to the control area of ​​the area to which it belongs is recorded as the second completeness of the target area; if the target area is not the control area, the second completeness of the target area is recorded as 1.

5. A method for detecting the quality of map prints according to claim 4, characterized in that: The obtaining of natural features of any grayscale image includes: For any grayscale image, boundary pixels adjacent to feature pixels are obtained and recorded as equivalent feature pixels. Any feature pixel is recorded as a growth seed, and regional growth is performed. The growth condition is that there are feature pixels or equivalent feature pixels between adjacent pixels. The growth stop condition is that there are no adjacent pixels and no equivalent feature pixels between adjacent pixels. After the growth stops, if there are still feature pixels that have not grown, growth seeds are selected from the feature pixels that have not grown until all feature pixels have their own growth regions. Any growth region is recorded as a natural feature, and several natural features of the grayscale image are obtained.

6. A method for detecting the quality of map prints according to claim 4, characterized in that: The target grayscale image is about The second integrity of the control area satisfies the expression: ； Where, Represents the target grayscale image about The second completeness of the control area; Represents the target grayscale image about The number of pixels in the control area that belong to natural features; express The number of pixels belonging to natural features in ; Represents the target grayscale image about the number of natural features in the control area; Represents the target grayscale image about The chain code of the rth natural feature of the control area is DTW distance of the chain code of natural features in .

7. A method for detecting the quality of map prints according to claim 1, characterized in that: The obtaining of the overall integrity of each region in the target grayscale image includes: The first integrity of the target area and the second integrity of the target area are added together and normalized for positive correlation to obtain the overall integrity of the target area.

8. A method for detecting the quality of map prints according to claim 1, characterized in that: The defect index of any defect area satisfies the expression: ； Where, represents the defect index of the h-th defect area; represents the number of regions of the h-th defective region; The number of pixels in the sth region representing the hth defect area; represents the overall integrity of the sth region of the hth defect region.

9. A map print quality inspection system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a method for detecting the quality of map prints according to any one of claims 1 to 8 is implemented.

Images