Method and device for tracing source of problem map

Abstract

The present invention provides a method and device for tracing the source of a problem map, including: obtaining a problem map and a picture to be detected; performing feature extraction on the problem map to obtain a first feature data set; performing feature extraction on the picture to be detected to obtain a second feature data set; performing homology judgment on the first feature data set and the second feature data set to determine whether the problem map and the picture to be detected are homology map picture sets; extracting features of the problem map, greatly reducing the category of the problem map sample library for image recognition, effectively reducing the dependence on quantity and quality, being able to efficiently and automatically match to similar maps, and improving the efficiency of tracing the problem map.

Description

Technical Field

[0001] The present invention relates to the field of geographic information technology, and in particular to a method and device for tracing the source of a problem map. Background Art

[0002] Internet maps are an important carrier of geographic information. With the booming development of emerging technologies such as mobile Internet, cloud computing, and artificial intelligence, and the continuous deepening of geographic information applications, new Internet map products such as real-life maps and autonomous driving maps are becoming increasingly popular, and static maps are widely used on social platforms, news media and other websites.

[0003] In the supervision and management of Internet maps, it is often necessary to trace the source of problematic maps and find the websites where they were originally published and cited.

[0004] At present, the methods used to search or identify problem maps are mainly based on keywords, colors, textures or deep learning, while the methods used to trace the source of problem maps are mainly through identification through complex problem map sample libraries and then manual judgment, lacking automatic discrimination technology.

[0005] However, the above-mentioned image recognition-based methods have high requirements on the construction quality of the problem map sample library; the vertical search engines based on keywords, colors or textures have a high proportion of non-map images and are less capable of discovering problem maps. Summary of the invention

[0006] In view of this, the purpose of the present invention is to provide a method and device for tracing problem maps, extract the features of problem maps, greatly reduce the categories of image recognition on problem map sample libraries, effectively reduce the dependence on quantity and quality, and can efficiently and automatically match similar maps, thereby improving the efficiency of tracing problem maps.

[0007] In a first aspect, an embodiment of the present invention provides a method for tracing the source of a problem map, the method comprising:

[0008] Get the problem map and the image to be tested;

[0009] Performing feature extraction on the problem map to obtain a first feature data set;

[0010] Extracting features from the image to be detected to obtain a second feature data set;

[0011] The first feature data set and the second feature data set are subjected to homology judgment to determine whether the problem map and the to-be-detected picture are homology map picture sets.

[0012] Furthermore, the first feature data set includes a first road network feature data set, a first annotation feature data set, a first target map location set and a first image map location set, and the second feature data set includes a second road network feature data set, a second annotation feature data set, a second target map location set and a second image map location set.

[0013] Furthermore, feature extraction is performed on the problem map to obtain a first feature data set, including:

[0014] Performing image enhancement processing on the problem map and then performing edge extraction to obtain a first linear feature set;

[0015] Performing text recognition on the first linear feature set to obtain a first annotation feature data set;

[0016] Performing a preprocessing of removing impurities on the first linear feature set to obtain a feature set after removing impurities;

[0017] Topologically connecting the feature sets after the removal of impurities within a set threshold to obtain a first road network feature data set;

[0018] Calculating the direction value of each linear feature in the first road network feature data set to obtain a statistical result;

[0019] A first target spot position set and a first image map position set are determined according to the category of the statistical result.

[0020] Further, determining the first target spot position set and the first image map position set according to the category of the statistical result includes:

[0021] If the category of the statistical result is clustering, then the problem map is a road network map;

[0022] If the category of the statistical result is discretization, the problem map is subjected to rapid supervised classification and impurity removal for cement material and water material to obtain airports and seawater, and the first target patch position set is recorded;

[0023] The first image map location set is obtained by searching using the mobile set target image.

[0024] Furthermore, the method further comprises:

[0025] If the first image map location set is empty, the problem map is a map of administrative division lines;

[0026] If the first image map location set is not empty, the problem map is an image map.

[0027] Further, performing homology judgment on the first feature data set and the second feature data set to determine whether the problem map and the to-be-detected image are homology map image sets includes:

[0028] Filtering out a set of pictures of the same type as the problem map from the pictures to be detected;

[0029] If the problematic map is a road network map, then when the matching degree between the first road network feature data set and the second road network feature data set is greater than a first preset threshold, the problematic map is a suspected homologous map;

[0030] Continue to perform semantic similarity matching on the first annotation feature data set and the second annotation feature data set;

[0031] If the matching degree is greater than a second preset threshold, the problem map and the picture set are the homologous map picture set.

[0032] Further, performing homology judgment on the first feature data set and the second feature data set to determine whether the problem map and the to-be-detected image are homology map image sets includes:

[0033] If the problematic map is an image map, then when the matching degree between the first image map position set and the second image map position set is greater than a third preset threshold, the problematic map is a suspected homologous map;

[0034] Continue to perform semantic similarity matching on the first target spot position set and the second target spot position set;

[0035] If the matching degree is greater than a fourth preset threshold, the problem map and the picture set are the homologous map picture set.

[0036] Further, performing homology judgment on the first feature data set and the second feature data set to determine whether the problem map and the to-be-detected image are homology map image sets includes:

[0037] If the problem map is a map of administrative division lines, extract the problem map to obtain the shape characteristics and positions of the border error-prone areas or the border line key areas, and record the annotation semantics and annotation positions of the problem map;

[0038] Determining the suspected homologous map through the similarity of shape features and positions of the border error-prone areas or the border key areas;

[0039] If the similarity of the annotation semantics and the difference of the annotation positions both meet the fifth preset threshold, it is determined to be the homologous map picture set.

[0040] In a second aspect, an embodiment of the present invention provides a device for tracing the source of a problem map, the device comprising:

[0041] The acquisition module is used to obtain the problem map and the image to be detected;

[0042] A first extraction module, used for performing feature extraction on the problem map to obtain a first feature data set;

[0043] A second extraction module is used to extract features from the image to be detected to obtain a second feature data set;

[0044] The homology judgment module is used to judge the homology of the first feature data set and the second feature data set to determine whether the problem map and the image to be detected are homology map image sets.

[0045] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor implements the method described above when executing the computer program.

[0046] The embodiment of the present invention provides a method and device for tracing the source of a problem map, including: obtaining a problem map and a picture to be detected; performing feature extraction on the problem map to obtain a first feature data set; performing feature extraction on the picture to be detected to obtain a second feature data set; performing homology judgment on the first feature data set and the second feature data set to determine whether the problem map and the picture to be detected are homologous map picture sets; extracting features of the problem map, greatly reducing the category of the problem map sample library for image recognition, effectively reducing the dependence on quantity and quality, being able to efficiently and automatically match to similar maps, and improving the efficiency of tracing the problem map.

[0047] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A flow chart of a method for tracing the source of a problem map provided in the first embodiment of the present invention;

[0051] Figure 2 A schematic diagram of a source tracing device for a problem map provided in Embodiment 2 of the present invention.

[0052] icon:

[0053] 1-acquisition module; 2-first extraction module; 3-second extraction module; 4-homologous judgment module. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] In this application, the problem maps include road network maps, image maps, and administrative division line maps. The error-prone point of road network maps lies in the annotations. If the problem map to be searched is a road network map, it is only necessary to compare the similarity of the main routes and the annotation content to determine whether they are of the same source;

[0056] The common error point of image maps is moving military targets. If the problem map to be searched is an image map, the position of the moving military targets and the size of other landforms can be used to determine whether they are of the same origin; the common error point of administrative division line maps is the border areas or key border lines. If the problem map to be searched is an administrative division line map, the shape, position and annotation content of the feature lines can be compared to determine whether they are of the same origin.

[0057] To facilitate understanding of this embodiment, the embodiment of the present invention is described in detail below.

[0058] Embodiment 1:

[0059] Figure 1 This is a flow chart of a method for tracing the source of a problem map provided in Example 1 of the present invention.

[0060] Reference Figure 1 , the method comprises the following steps:

[0061] Step S101, obtaining a problem map and a picture to be detected;

[0062] Here, after obtaining the pictures to be detected, the pictures to be detected are uniformly aligned according to the resolution and size of the problem map.

[0063] Step S102, extracting features from the problem map to obtain a first feature data set;

[0064] Step S103, extracting features from the image to be detected to obtain a second feature data set;

[0065] Step S104, performing homology judgment on the first feature data set and the second feature data set to determine whether the problem map and the image to be detected are from the same source map image set.

[0066] Furthermore, the first feature data set includes a first road network feature data set, a first annotation feature data set, a first target map location set and a first image map location set, and the second feature data set includes a second road network feature data set, a second annotation feature data set, a second target map location set and a second image map location set.

[0067] Further, step S102 includes the following steps:

[0068] Step S201, performing image enhancement processing on the problem map and then performing edge extraction to obtain a first linear feature set;

[0069] Step S202, performing text recognition on the first linear feature set to obtain a first annotation feature data set;

[0070] Step S203, performing a preprocessing of removing impurities on the first linear feature set to obtain a feature set after removing impurities;

[0071] Here, the preprocessing of removing impurities refers to removing line segments whose length is less than a set threshold (ie, shorter linear feature data) and leaving longer feature data, thereby reducing the amount of feature data and the amount of subsequent calculation of feature directions.

[0072] Step S204, topologically connecting the feature sets after the removal of impurities within a set threshold to obtain a first road network feature data set;

[0073] Step S205, calculating the direction value of each linear feature in the first road network feature data set to obtain a statistical result;

[0074] Specifically, the statistical result is the direction value of each linear feature in the first road network feature data set, that is, 3 / π, 60°. If the problem map is a road network map, the position relationship of the longer line segments (linear features) tends to be parallel or vertical, so the direction value will tend to two values, so the statistical result will finally be expressed as two categories.

[0075] Step S206, determining a first target spot position set and a first image map position set according to the category of the statistical result.

[0076] Here, the process of extracting features from the image to be detected and obtaining the second feature data set is similar to the process of extracting features from the problem map, which will not be described here. Among them, the classification of the image to be detected may be more than the classification of the "problem map", and the wrongly classified maps are all classified into the category of administrative division line map.

[0077] Further, step S206 includes the following steps:

[0078] Step S301, if the category of the statistical result is clustering, then the problem map is a road network map;

[0079] Step S302: if the category of the statistical result is discretization, the problem map is subjected to rapid supervised classification and impurity removal for cement material and water material to obtain the airport (cement) and seawater (water body), and the first target spot position set is recorded;

[0080] Here, removing noise refers to removing spots whose size (area) is within a set threshold.

[0081] Step S303: searching by moving the set target image to obtain a first image map location set.

[0082] Among them, the clue image of the problem map is already a map with errors, so it only needs to be classified in the problem map. This is also where the existing map recognition technology is distinguished, and the classification accuracy is improved without relying on the quality of the sample library.

[0083] Furthermore, the method further comprises the following steps:

[0084] Step S401, if the first image map location set is empty, the problem map is a map of administrative division lines;

[0085] Step S402: If the first image map location set is not empty, the problem map is an image map.

[0086] Further, step S104 includes the following steps:

[0087] Step S501, selecting a set of pictures of the same type as the problem map from the pictures to be detected;

[0088] Step S502: If the problematic map is a road network map, then when the matching degree between the first road network feature data set and the second road network feature data set is greater than a first preset threshold, the map is suspected to be a homologous map;

[0089] Step S503, continue to perform semantic similarity matching on the first annotation feature data set and the second annotation feature data set;

[0090] Step S504: If the matching degree is greater than the second preset threshold, the problem map and the picture set are the same source map picture set.

[0091] Furthermore, step S104 further includes the following steps:

[0092] Step S601: If the problem map is an image map, then when the matching degree between the first image map position set and the second image map position set is greater than a third preset threshold, the problem map is a suspected homologous map;

[0093] Step S602, continue to perform semantic similarity matching on the first target spot position set and the second target spot position set;

[0094] Step S603: If the matching degree is greater than a fourth preset threshold, the problem map and the picture set are the same source map picture sets.

[0095] Furthermore, step S104 further includes the following steps:

[0096] Step S701, if the problem map is a map of administrative division lines, extract the problem map to obtain the shape characteristics and positions of the border error-prone areas or the border line key areas, and record the annotation semantics and annotation positions of the problem map;

[0097] Step S702, determining suspected homologous maps based on the shape features and position similarity of the border error-prone areas or the border key areas;

[0098] Step S703: If the similarity of the annotation semantics and the difference of the annotation positions both meet the fifth preset threshold, it is determined to be a homologous map image set.

[0099] This application classifies the features of problematic maps, greatly reduces the categories of sample libraries for image recognition, effectively reduces dependence on quantity and quality, can efficiently match similar maps, improves the efficiency of tracing problematic maps, and can effectively replace the manual judgment part of current map supervision.

[0100] Embodiment 2:

[0101] Figure 2 A schematic diagram of a source tracing device for a problem map provided in Embodiment 2 of the present invention.

[0102] Reference Figure 2 , the device comprises:

[0103] Acquisition module 1, used to obtain the problem map and the image to be detected;

[0104] The first extraction module 2 is used to extract features from the problem map to obtain a first feature data set;

[0105] The second extraction module 3 is used to extract features of the image to be detected to obtain a second feature data set;

[0106] The homology judgment module 4 is used to judge the homology of the first feature data set and the second feature data set to determine whether the problem map and the image to be detected are homology map image sets.

[0107] In this application, map classification is constructed based on the problem feature points of the problem map, including road network maps, remote sensing images and maps of administrative division lines. There are fewer categories, low computing power requirements and fast classification speed.

[0108] This application does not rely on sample libraries, and has no requirements on the quality, update timeliness, content richness, etc. of sample libraries, and has high stability. The location of map category features is used to automatically determine whether it is a homologous map, without the need for manual judgment.

[0109] An embodiment of the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the problem map tracing method provided in the above embodiment are implemented.

[0110] An embodiment of the present invention also provides a computer-readable medium having a non-volatile program code executable by a processor, wherein a computer program is stored on the computer-readable medium, and when the computer program is executed by the processor, the steps of the problem map tracing method of the above embodiment are executed.

[0111] The computer program product provided in the embodiment of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be found in the method embodiment and will not be repeated here.

[0112] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0113] In addition, in the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0114] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0115] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0116] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for tracing the source of a problem map, characterized in that: The method comprises: Get the problem map and the image to be tested; Performing feature extraction on the problem map to obtain a first feature data set; Extracting features from the image to be detected to obtain a second feature data set; Perform homology judgment on the first feature data set and the second feature data set to determine whether the problem map and the image to be detected are homology map image sets; The first feature data set includes a first road network feature data set, a first annotation feature data set, a first target spot location set and a first image map location set, and the second feature data set includes a second road network feature data set, a second annotation feature data set, a second target spot location set and a second image map location set; Perform feature extraction on the problem map to obtain a first feature data set, including: Performing image enhancement processing on the problem map and then performing edge extraction to obtain a first linear feature set; Performing text recognition on the first linear feature set to obtain a first annotation feature data set; Performing a preprocessing of removing impurities on the first linear feature set to obtain a feature set after removing impurities; Topologically connecting the feature sets after the removal of impurities within a set threshold to obtain a first road network feature data set; Calculating the direction value of each linear feature in the first road network feature data set to obtain a statistical result; Determining a first target spot position set and a first image map position set according to the category of the statistical result; Determining a first target spot position set and a first image map position set according to the category of the statistical result includes: If the category of the statistical result is clustering, then the problem map is a road network map; If the category of the statistical result is discretization, the problem map is subjected to rapid supervised classification and impurity removal for cement material and water material to obtain airports and seawater, and the first target patch position set is recorded; The first image map location set is obtained by searching using the mobile set target image.

2. The method for tracing the source of a problem map according to claim 1, characterized in that: The method further comprises: If the first image map location set is empty, the problem map is a map of administrative division lines; If the first image map location set is not empty, the problem map is an image map.

3. The method for tracing the source of a problem map according to claim 1, characterized in that: The first feature data set and the second feature data set are subjected to homology judgment to determine whether the problem map and the to-be-detected image are homology map image sets, including: Filtering out a set of pictures of the same type as the problem map from the pictures to be detected; If the problematic map is a road network map, then when the matching degree between the first road network feature data set and the second road network feature data set is greater than a first preset threshold, the problematic map is a suspected homologous map; Continue to perform semantic similarity matching on the first annotation feature data set and the second annotation feature data set; If the matching degree is greater than a second preset threshold, the problem map and the picture set are the homologous map picture set.

4. The method for tracing the source of a problem map according to claim 1, characterized in that: The first feature data set and the second feature data set are subjected to homology judgment to determine whether the problem map and the to-be-detected image are homology map image sets, including: If the problematic map is an image map, then when the matching degree between the first image map position set and the second image map position set is greater than a third preset threshold, the problematic map is a suspected homologous map; Continue to perform semantic similarity matching on the first target spot position set and the second target spot position set; If the matching degree is greater than a fourth preset threshold, the problem map and the picture set are the homologous map picture set.

5. The method for tracing the source of a problem map according to claim 1, characterized in that: The first feature data set and the second feature data set are subjected to homology judgment to determine whether the problem map and the to-be-detected image are homology map image sets, including: If the problem map is a map of administrative division lines, extract the problem map to obtain the shape characteristics and positions of the border error-prone areas or the border line key areas, and record the annotation semantics and annotation positions of the problem map; Determining the suspected homologous map through the similarity of shape features and positions of the border error-prone areas or the border key areas; If the similarity of the annotation semantics and the difference of the annotation positions both meet the fifth preset threshold, it is determined to be the homologous map picture set.

6. A device for tracing the source of a problem map, characterized in that: The device comprises: The acquisition module is used to obtain the problem map and the image to be detected; A first extraction module, used for performing feature extraction on the problem map to obtain a first feature data set; A second extraction module is used to extract features from the image to be detected to obtain a second feature data set; A homology judgment module, used for judging the homology of the first feature data set and the second feature data set, and determining whether the problem map and the image to be detected are homology map image sets; The first feature data set includes a first road network feature data set, a first annotation feature data set, a first target spot location set and a first image map location set, and the second feature data set includes a second road network feature data set, a second annotation feature data set, a second target spot location set and a second image map location set; The first extraction module is specifically used for: Performing image enhancement processing on the problem map and then performing edge extraction to obtain a first linear feature set; Performing text recognition on the first linear feature set to obtain a first annotation feature data set; Performing a preprocessing of removing impurities on the first linear feature set to obtain a feature set after removing impurities; Topologically connecting the feature sets after the removal of impurities within a set threshold to obtain a first road network feature data set; Calculating the direction value of each linear feature in the first road network feature data set to obtain a statistical result; Determining a first target spot position set and a first image map position set according to the category of the statistical result; The first extraction module is specifically used for: If the category of the statistical result is clustering, then the problem map is a road network map; If the category of the statistical result is discretization, the problem map is subjected to rapid supervised classification and impurity removal for cement material and water material to obtain airports and seawater, and the first target patch position set is recorded; The first image map location set is obtained by searching using the mobile set target image.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

Images