A map information point detection method, device, equipment and storage medium

Abstract

The present disclosure provides a map information point detection method and device, equipment and a storage medium, relates to the technical field of data processing, and particularly relates to the technical field of map data processing. The specific implementation scheme is: after obtaining a target information point to be detected, based on the position of the target information point, a target satellite map within a preset range of the target information point is obtained, a target feature of the target satellite map is extracted, the feature is searched in a pre-constructed no-information-point-coverage feature retrieval library, and if the target feature is found in the no-information-point-coverage feature retrieval library, it is determined that the target information point to be detected is abnormal. According to the embodiment of the present disclosure, whether the target information point is abnormal is detected through the satellite map around the target information point, and the application is not limited by the information source. Meanwhile, the satellite map has a wide coverage range, can clearly reflect the surface features, improves the coverage range of the map information point anomaly detection, and improves the detection accuracy.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of data processing, and particularly relates to the technical field of map data processing. BACKGROUND

[0002] In the field of map application, it is generally required to update information points in a map to ensure consistency between the map and the real geographical environment. The information points in the map generally refer to any meaningful points on the map without geographical significance, such as shops, scenic spots, public toilets, gas stations, hotels and the like.

[0003] Offline of abnormal information points is an important part of updating map information. Therefore, each information point in the map needs to be detected to obtain abnormal information points for offline, and then the map is updated. SUMMARY

[0004] The present disclosure provides a map information point detection method and device, an electronic device and a storage medium for detecting whether a map information point is abnormal.

[0005] According to an aspect of the present disclosure, a map information point detection method is provided, comprising:

[0006] obtaining a target information point to be detected in a general map;

[0007] obtaining a target satellite map within a preset range of the target information point based on the position of the target information point;

[0008] extracting a target feature of the target satellite map;

[0009] searching for the target feature in a pre-constructed information point free feature retrieval library;

[0010] if the target feature is found in the information point free feature retrieval library, determining that the target information point is an abnormal information point.

[0011] According to another aspect of the present disclosure, a map information point detection device is provided, comprising:

[0012] a target information point obtaining module configured to obtain a target information point to be detected in a general map;

[0013] a target satellite map obtaining module configured to obtain a target satellite map within a preset range of the target information point based on the position of the target information point;

[0014] a target feature extracting module configured to extract a target feature of the target satellite map;

[0015] a target feature searching module, configured to search the target feature in a pre-constructed feature retrieval library without information point coverage;

[0016] an abnormal information point determining module, configured to determine the target information point as an abnormal information point if the target feature is found in the feature retrieval library without information point coverage.

[0017] According to another aspect of the present disclosure, an electronic device is provided, comprising:

[0018] at least one processor; and

[0019] a memory connected with the at least one processor in communication; wherein,

[0020] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform any of the above-mentioned map information point detection methods.

[0021] According to another aspect of the present disclosure, a non-transitory computer readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to perform any of the above-mentioned map information point detection methods.

[0022] According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements any of the above-mentioned map information point detection methods.

[0023] It should be understood that the content described in this part is not intended to identify key or important features of the embodiments of the present disclosure, nor to limit the scope of the present disclosure. Other features of the present disclosure will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0024] The accompanying drawings are used to better understand the present scheme, and do not constitute a limitation on the present disclosure. Among them:

[0025] Figure 1 is a schematic diagram of a first embodiment of the map information point detection method provided according to the present disclosure;

[0026] Figure 2 is a schematic diagram of an embodiment of constructing a feature retrieval library without information point coverage in the present disclosure;

[0027] Figure 3 is a schematic diagram of a second embodiment of the map information point detection method provided according to the present disclosure;

[0028] Figure 4 is a schematic diagram of a first embodiment of training a satellite map feature extraction model in the present disclosure;

[0029] Figure 5a is a schematic diagram of a satellite map in the present disclosure;

[0030] Figure 5b is Figure 5a is a schematic diagram of a satellite map corresponding to the satellite map shown in the present disclosure;

[0031] Figure 6 is a schematic diagram of a second embodiment of the satellite map feature extraction model in the present disclosure;

[0032] Figure 7 is a schematic diagram of a first embodiment of a map information point detection device according to the present disclosure;

[0033] Figure 8 is a schematic diagram of a second embodiment of a map information point detection device according to the present disclosure;

[0034] Figure 9 is a block diagram of an electronic device for implementing a map information point detection method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to assist in understanding, which should be considered in their context only. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0036] Currently, abnormal information points in the map are detected by means such as intelligence (e.g., user feedback) and vehicle image acquisition, and then the abnormal information points are offline, which is an important means for POI offline. However, due to the limitations of the acquisition method and the low coverage, it is not possible to accurately offline some POIs. That is, the POI offline based on intelligence or vehicle image acquisition is limited by the intelligence source and has small coverage.

[0037] To solve the above problems, the present disclosure provides a map information point detection method, device, equipment and storage medium. First, the map information point detection method provided by the present disclosure is described below.

[0038] Referring to Figure 1 , Figure 1 is a flowchart of a first embodiment of a map information point detection method according to the present disclosure. The method can include the following steps:

[0039] Step S110, obtaining a target information point to be detected in a general map.

[0040] In the embodiments of the present disclosure, the target information point can be a POI (point of interest) in the general map. The POI in the map generally refers to any meaningful point on the map that has no geographical meaning, such as shops, scenic spots, public toilets, gas stations, hotels, and the like.

[0041] The general map is a map that comprehensively and comprehensively reflects the general characteristics of natural elements and social and economic phenomena in a certain mapping area. The map contains terrain, water system, soil, vegetation, residential area, transportation network, boundary line and the like. For example, the navigation map used in map navigation.

[0042] The information points and traffic network and the like in the general map are generally collected by vehicles or unmanned aerial vehicles and the like equipped with image collection devices, and then the map is drawn based on the collected information. However, the information collected in this way may have a certain lag, which will cause the general map to not completely reflect the characteristics of each region in real time. Therefore, it is necessary to update the general map, such as detecting abnormal information points and offline abnormal information points.

[0043] In the embodiments of the present disclosure, the target information point (query POI) can be manually specified by a person according to actual needs, or can be automatically selected by an electronic device. For example, the electronic device can periodically traverse each information point on the map, and the information point currently selected by the electronic device can be the target information point.

[0044] The target information point can be a shop, a scenic spot, a gas station, and the like, which is not limited in the present disclosure.

[0045] Step S120, based on the position of the target information point, obtaining a target satellite map within a preset range of the target information point.

[0046] In the embodiments of the present disclosure, after the target information point is obtained, the position coordinates of the target information point in the map can be determined. As a specific implementation, the position coordinates of the target information point can include the longitude and latitude information of the target information point. Then, based on the position coordinates of the target information point, the satellite map within the preset range of the target information point can be obtained as the target satellite map.

[0047] The satellite map, also known as satellite remote sensing image or satellite image, is an image that uses satellite as a medium to feed back the surface features of the earth to the user. Unlike traditional maps (such as general maps), the surface features seen on the satellite map are real and real-time.

[0048] In the embodiments of the present disclosure, the range of the target satellite map can be pre-set by a person according to actual needs. As a specific implementation of the embodiments of the present disclosure, a square satellite map region with a radius of 128 meters centered on the position coordinates of the target information point can be obtained as the target satellite map.

[0049] In step S130, a target feature of the target satellite map is extracted.

[0050] In the embodiments of the present disclosure, the target feature of the target satellite map can be extracted by a pre-trained satellite map feature extraction model.

[0051] In step S140, the target feature is searched in a pre-constructed information point-free coverage feature retrieval library.

[0052] In the embodiments of the present disclosure, the information point-free coverage feature retrieval library can be features of a satellite map of a region pre-acquired without coverage of information points (such as shops, scenic spots, gas stations, public toilets, and the like).

[0053] In the embodiments of the present disclosure, the satellite feature extraction model can be used to pre-extract features of a satellite map without coverage of information points from the satellite map.

[0054] In step S150, if the target feature is found in the information point-free coverage feature retrieval library, it is determined that the target information point is an abnormal information point.

[0055] In the embodiments of the present disclosure, if the feature of the target satellite map is searched in the information point-free coverage feature retrieval library, it is indicated that there is no coverage of information points in the target satellite map, that is, the target information point can not be located at the position displayed on the map, that is, the target information point can be data that needs to be removed because of being demolished or data that needs to be changed because of being wrong, and the like.

[0056] As described above, the ground surface seen on the satellite map is real and real-time. Therefore, demolition or wrong data (such as a shopping type shop POI falling into water or being in a green land) can be clearly found through the satellite map, and the detection accuracy of the map information point is improved.

[0057] It can be seen that in the map information point detection method provided by the embodiment of the present disclosure, after the target information point to be detected is acquired, the target satellite map within the preset range of the target information point is acquired based on the position of the target information point, then the target feature of the target satellite map is extracted, and the target feature is searched in the no-information-point-covered feature retrieval library constructed in advance. If the target feature is found in the no-information-point-covered feature retrieval library, it can be determined that the target information point to be detected is abnormal. By using the embodiment of the present disclosure, whether the target information point is abnormal is detected through the satellite map around the target information point, which is not limited to intelligence sources. Meanwhile, since the satellite map has a wide coverage range and can clearly reflect the surface features, the coverage range of the map information point abnormality detection is improved, and the detection accuracy is improved, so that the abnormal information point can be accurately offline.

[0058] In an embodiment of the present disclosure, as shown in Figure 2 the no-information-point-covered feature retrieval library can be constructed by the following steps:

[0059] Step S210: manually labeling each no-information-point-covered area on the satellite map.

[0060] As an embodiment of the present disclosure, the no-information-point-covered area on the satellite map can be labeled manually. The no-information-point-covered area can include empty land, green land, river, etc.

[0061] Step S220: extracting the features of each no-information-point-covered area on the satellite map based on the result of the manual labeling.

[0062] In the embodiment of the present disclosure, the features of the satellite map corresponding to the manually labeled no-information-point-covered area can be extracted. The features of the satellite map corresponding to the manually labeled no-information-point-covered area can be extracted based on the result of the manual labeling, so that the satellite map features of the no-information-point-covered area on the satellite map can be accurately acquired.

[0063] Step S230: constructing the no-information-point-covered feature retrieval library based on the features of each no-information-point-covered area.

[0064] As a specific implementation of the embodiment of the present disclosure, after the satellite map features of each no-information-point-covered area are acquired, the features can be stored to construct the no-information-point-covered feature retrieval library.

[0065] In an embodiment of the present disclosure, as shown in Figure 3 the step S140 shown in Figure 1 may be refined as:

[0066] Step S141, similarity between the target feature and each feature in the feature search library without information points is calculated respectively.

[0067] As a specific embodiment of the present disclosure, the similarity can be cosine distance, Mahalanobis distance, Euclidean distance, etc. between the target satellite map feature and each feature in the feature search library without information points. The present disclosure does not make specific limitations on this.

[0068] Correspondingly, as shown in Figure 3 Step S150 shown in Figure 1 may be refined as:

[0069] Step S151, if there is a feature in the feature search library without information points that has similarity exceeding a preset similarity threshold with the target feature, it is determined that the target information point is an abnormal information point.

[0070] In the present disclosure, the preset similarity threshold can be set by human beings in advance. As a specific example, the preset similarity threshold can be 0.83.

[0071] Correspondingly, in an embodiment of the present disclosure, after the feature search library without information points is searched by using the target satellite map feature, if there is a return result, i.e., if there is a feature having similarity greater than a preset similarity (e.g., cosine distance greater than 0.83) with the target satellite map feature, it can be determined that the target information point is abnormal. For example, the target information point is suspected to be offline or has position error and needs to be modified. If there is no return result, the target information point is a normal information point.

[0072] In the present disclosure, by searching the feature search library without information points pre-constructed by using the target satellite map feature, the determination result of whether the target information point is abnormal is obtained, the coverage is wide, each map information point can be detected without omission, and the accuracy is high.

[0073] In an embodiment of the present disclosure, based on Figure 1 As shown in Figure 3 the method can further include:

[0074] Step S360, the abnormal information point is offline in the normal map.

[0075] In an embodiment of the present disclosure, after it is determined that the target information point is an abnormal information point, the information point can be deleted in the normal map (e.g., navigation map). Then, the current geographical information of the corresponding position can be re-collected by image collection, and the normal map is updated based on the collected information.

[0076] For example, if the target information point is a shopping POI that has been removed, the corresponding location is currently an empty land, and the shopping POI is still displayed in the corresponding location in the map. Then the shopping POI can be deleted in the general map, and the information of the location in the general map can be modified to an empty land.

[0077] Of course, in other embodiments of the present disclosure, the general map can also be modified based on the satellite map corresponding to the location of the abnormal information point, and the like. The present disclosure does not make specific limitations in this regard.

[0078] In the embodiments of the present disclosure, by removing the detected POI, the user can be timely prevented from being misled by the incorrect POI in the general map (such as a navigation map), and the map use experience can be improved.

[0079] As described above, in an embodiment of the present disclosure, the target features of the target satellite map extracted in step S130 and the features of the area covered by each information point on the satellite map extracted based on the result of manual labeling in step S220 can be extracted by using a pre-trained satellite map feature extraction model to speed up the feature extraction.

[0080] In an embodiment of the present disclosure, the satellite map feature extraction model can be a CNN (Convolutional Neural Networks, convolutional neural network). For example, it can be a ResNet-50 model or the like.

[0081] In an embodiment of the present disclosure, as shown in the following table, the satellite map feature extraction model can be pre-trained by the following steps: Figure 4

[0082] Step S410, based on the satellite map, the corresponding satellite base map is obtained.

[0083] In the embodiments of the present disclosure, the satellite base map can be obtained by processing the satellite map according to the actual needs of the relevant personnel. For example, only the main lines in the satellite map can be retained as the satellite base map. For example, only the roads in the satellite map can be retained as the satellite base map, and then the road network can be analyzed and processed based on the satellite base map. Alternatively, the satellite map can be processed by using software with elevation data processing capability to obtain the corresponding topographic map, and then the topographic map is Gaussian blurred to obtain the corresponding satellite base map, and then the topography and location of a specific area can be labeled based on the satellite base map to more directly reflect the geographical conditions of the specific area, and the like. The above examples are only used to illustrate the way of obtaining the satellite base map in the present disclosure, and do not specifically limit the present disclosure.

[0084] ​As a specific implementation of the embodiments of the present disclosure, the contrast of the satellite map can be adjusted until only the main lines in the satellite map appear, and then the required lines are outlined as the satellite base map. The main lines can include the lines of rivers, roads, green spaces, and buildings, etc.

[0085] As shown in Figure 5a With Figure 5b As shown in Figure 5a is a satellite map of a certain area, Figure 5b is the satellite base map corresponding to the satellite map of the area. Figure 5b The satellite base map shown in Figure 5a The main lines in the satellite map shown in

[0086] Step S420, based on the satellite base map, obtaining a satellite base map data set.

[0087] In the embodiments of the present disclosure, the satellite map data set can include positive samples and negative samples.

[0088] Step S430, based on the satellite base map data set, training the satellite map feature extraction model to be trained.

[0089] In an embodiment of the present disclosure, the satellite map feature extraction model to be trained can be trained based on the satellite base map data set by using a metric learning method such as arcface algorithm. The present disclosure does not make specific limitations on this.

[0090] For example, the satellite map feature extraction model to be trained can be trained in the embodiments of the present disclosure by using the following steps:

[0091] ①, input each picture in the satellite base map data set to the satellite map feature extraction model to be trained in turn.

[0092] ②, obtaining the probability that the picture output by the satellite map feature extraction model to be trained belongs to each preset type.

[0093] In the embodiments of the present disclosure, the each preset type can be information points such as gas stations, shops, and public toilets, and rivers, open spaces, and green spaces, etc.

[0094] ③, based on the input picture and the probability output by the satellite map feature extraction model to be trained, calculating the value of the arcface loss function.

[0095] ④ Based on the value of the arcface loss function calculated in step ③, adjust the parameters of the satellite map feature extraction model to be trained until the value of the arcface loss function is less than the preset error.

[0096] In this embodiment of the disclosure, the above-mentioned satellite base map feature extraction model is trained based on a satellite base map dataset. Since the satellite base map only contains the necessary information and has little unnecessary information, the model training is more convenient and the effect is better.

[0097] As described above, in one embodiment of this disclosure, the aforementioned satellite base map dataset may include positive samples and negative samples.

[0098] Specifically, based on Figure 4 ,like Figure 6 As shown, step S420 above can be further refined into the following steps:

[0099] Step S421: Based on the satellite base map, obtain the base map of each sub-satellite of a preset size.

[0100] In this embodiment of the disclosure, the dimensions of each sub-satellite base map can be preset manually. For example, the dimensions of each sub-satellite base map can be set to a square area with a side length of 256 meters.

[0101] In this embodiment, the satellite map can be converted into a satellite base map first, and then a sub-satellite base map can be obtained according to the aforementioned preset size. Of course, the satellite map can also be obtained according to the aforementioned preset size, and the corresponding satellite base map can be obtained. This disclosure does not specifically limit this method.

[0102] In this embodiment of the disclosure, the number of sub-satellite base maps acquired is not specifically limited.

[0103] Step S422: For each sub-satellite base map, randomly translate and rotate the sub-satellite base map to obtain the transformed base map corresponding to the sub-satellite base map.

[0104] In this embodiment of the disclosure, the acquired sub-satellite base map can be randomly translated and rotated with overlaps to obtain the transformed base map.

[0105] In this embodiment of the disclosure, the range for translating the aforementioned sub-satellite base map can be predetermined manually. As described above, in this embodiment of the disclosure, overlapping translations of the aforementioned sub-satellite base maps can be performed. Therefore, as a specific implementation, if the size of the aforementioned sub-satellite base map is a square area of ​​256 meters, then the aforementioned translation range can be set to 128 meters. That is, each sub-satellite base map can be randomly translated within 128 meters.

[0106] In the embodiments of the present disclosure, the range of rotation of each sub-satellite base map can be 360°.

[0107] In the embodiments of the present disclosure, after random translation and random rotation of each sub-satellite base map, a plurality of transformed base maps can be obtained. That is, one sub-satellite base map can correspond to a plurality of transformed base maps.

[0108] In step S423, for each sub-satellite base map, the corresponding preset number of transformed base maps are taken as positive samples, and the remaining transformed base maps corresponding to the sub-satellite base map are taken as negative samples.

[0109] In the embodiments of the present disclosure, the preset number can be less than the number of transformed base maps corresponding to the sub-satellite base map. As a specific embodiment, for each sub-satellite base map, 16 transformed base maps can be selected as positive samples, and the other transformed base maps corresponding to the sub-satellite base map are taken as negative samples.

[0110] The above-mentioned positive samples and negative samples constitute the above-mentioned satellite base map data set. Then, the satellite map feature extraction model can be trained based on the positive and negative samples.

[0111] In an embodiment of the present disclosure, after obtaining the above-mentioned positive samples and negative samples, the positive and negative samples can be filtered and screened. As a specific embodiment, the above-mentioned positive samples and negative samples can be screened by using a gray scale statistical method.

[0112] For example, for each transformed base map, the gray scale values of the transformed base map can be counted. If there is no obvious difference between the gray scale values of the transformed base map, the transformed base map can not contain valid information, for example, the transformed base map can be a base map corresponding to an unknown area on a satellite map. For such invalid base map, it can be filtered to further ensure the training effect.

[0113] It can be seen that the map information point detection method provided by the present disclosure uses the satellite map around the target information point to be detected to detect whether the target information point is abnormal, has a wide coverage, and reduces missed detection. At the same time, the map information point detection method provided by the present disclosure is not restricted by the information source, and makes up for the deficiency of intelligence offline information points.

[0114] According to the embodiments of the present disclosure, in another aspect of the present disclosure, a map information point detection device is also provided, as shown in Figure 7 The device can include:

[0115] The target information point acquisition module 710 can be configured to acquire a target information point to be detected in a general map.

[0116] The target satellite map acquisition module 720 can be configured to acquire a target satellite map in a preset range of the target information point based on the position of the target information point.

[0117] The target feature extraction module 730 can be configured to extract a target feature of the target satellite map.

[0118] The target feature searching module 740 can be configured to search for the target feature in a pre-constructed no-information-point-covered feature retrieval library.

[0119] The abnormal information point determination module 750 can be configured to determine that the target information point is an abnormal information point if the target feature is found in the no-information-point-covered feature retrieval library.

[0120] As can be seen, in the map information point detection device provided in the embodiments of the present disclosure, after a target information point to be detected is acquired, a target satellite map in a preset range of the target information point is acquired based on the position of the target information point, then a target feature of the target satellite map is extracted, and the target feature is searched for in a pre-constructed no-information-point-covered feature retrieval library. If the target feature is found in the no-information-point-covered feature retrieval library, it can be determined that the target information point to be detected is abnormal. By using the satellite map around the target information point to detect whether the target information point is abnormal, the present disclosure is not limited to intelligence sources. Meanwhile, since the satellite map has a wide coverage range and can clearly reflect the surface features, the coverage range of the map information point abnormality detection is improved, and the detection accuracy is improved.

[0121] In an embodiment of the present disclosure, the no-information-point-covered feature retrieval library is pre-constructed by the following method:

[0122] The areas covered by each no-information point on the satellite map are manually labeled.

[0123] Based on the results of the manual labeling, the features of the areas covered by each no-information point on the satellite map are extracted.

[0124] Based on the features of the areas covered by each no-information point, the no-information-point-covered feature retrieval library is constructed.

[0125] In an embodiment of the present disclosure, the target feature searching module 740 can be specifically configured to calculate the similarity between the target feature and each feature in the no-information-point-covered feature retrieval library.

[0126] The abnormal information point determination module 750 can be specifically configured to determine that the target information point is an abnormal information point if there is a feature in the no-information-point-covered feature retrieval library that has a similarity to the target feature exceeding a preset similarity threshold.

[0127] In an embodiment of the present disclosure, the target feature extraction module 730 can be specifically configured to extract target features of the target satellite map by using a pre-trained satellite map feature extraction model. Figure 7 As shown in the above device, the device can further include an abnormal information point offline module 860. Figure 8 The abnormal information point offline module 860 can be configured to offline the abnormal information points in the normal map.

[0128] The abnormal information point offline module 860 can be configured to offline the abnormal information points in the normal map.

[0129] In an embodiment of the present disclosure, the target feature extraction module 730 can be specifically configured to extract target features of the target satellite map by using a pre-trained satellite map feature extraction model.

[0130] The feature extraction of each information point-free area on the satellite map based on the artificial labeling result can include:

[0131] The feature extraction of each information point-free area on the satellite map based on the artificial labeling result can include:

[0132] In an embodiment of the present disclosure, the satellite map feature extraction model is trained by using the following method:

[0133] Based on the satellite map, a corresponding satellite base map is obtained.

[0134] Based on the satellite base map, a satellite base map data set is obtained.

[0135] Based on the satellite base map data set, a satellite map feature extraction model to be trained is trained.

[0136] In an embodiment of the present disclosure, the satellite base map data set can include positive samples and negative samples.

[0137] The satellite base map data set can be obtained based on the satellite base map, and can include positive samples and negative samples.

[0138] Based on the satellite base map, each sub-satellite base map of a preset size is obtained.

[0139] For each sub-satellite base map, the sub-satellite base map is randomly translated and randomly rotated to obtain each transformed base map corresponding to the sub-satellite base map.

[0140] For each sub-satellite base map, a preset number of corresponding transformed base maps are taken as positive samples, and the remaining transformed base maps corresponding to the sub-satellite base map are taken as negative samples.

[0141] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in the technical solution comply with relevant laws and regulations and do not violate public order and good customs.

[0142] According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

[0143] Figure 9 A schematic block diagram of an example electronic device 900 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present disclosure described and / or claimed in this document.

[0144] As shown in Figure 9 The device 900 includes a computing unit 901 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 902 or a computer program loaded into a random access memory (RAM) 903 from a storage unit 908. Various programs and data required for the operation of the device 900 can also be stored in the RAM 903. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0145] Various components in the device 900 are connected to the I / O interface 905, including an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the device 900 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0146] The computing unit 901 can be various general and / or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 901 performs various methods and processes described above, such as the map information point detection method. For example, in some embodiments, the map information point detection method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded onto the RAM 903 and executed by the computing unit 901, one or more steps of the map information point detection method described above can be performed. Alternatively, in other embodiments, the computing unit 901 can be configured to perform the map information point detection method by any other appropriate means, such as by means of firmware.

[0147] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0148] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces a means for implementing the functions / acts specified in the flowcharts and / or block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0149] In the context of this disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0150] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0151] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0152] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0153] It should be understood that the various forms of flow shown above can be used to reorder, add, or delete steps. For example, the steps described in the present disclosure can be performed in parallel, in series, or in a different order, as long as the desired results of the technology disclosed in the present disclosure can be achieved, which is not limited herein.

[0154] The above detailed description does not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for detecting a map information point, comprising: obtaining a target information point to be detected in a common map; obtaining a target satellite map within a preset range of the target information point based on a position of the target information point; extracting a target feature of the target satellite map; searching for the target feature in a pre-constructed no-information-point-coverage feature retrieval library; if the target feature is found in the no-information-point-coverage feature retrieval library, determining that the target information point is an abnormal information point; the no-information-point-coverage feature retrieval library is pre-constructed by the following method: manually labeling each no-information-point-coverage area on a satellite map; based on the result of the manual labeling, extracting features of each no-information-point-coverage area on the satellite map; based on the features of each no-information-point-coverage area, constructing the no-information-point-coverage feature retrieval library.

2. The method of claim 1, wherein, the step of searching for the target feature in the pre-constructed no-information-point-coverage feature retrieval library comprises: respectively calculating similarities between the target feature and each feature in the no-information-point-coverage feature retrieval library; the step of determining that the target information point is an abnormal information point if the target feature is found in the no-information-point-coverage feature retrieval library comprises: if there is a feature in the no-information-point-coverage feature retrieval library that has a similarity to the target feature exceeding a preset similarity threshold, determining that the target information point is an abnormal information point.

3. The method of claim 1, wherein, the step of extracting the target feature of the target satellite map comprises: extracting the target feature of the target satellite map by using a pre-trained satellite map feature extraction model; the step of extracting features of each no-information-point-coverage area on the satellite map based on the result of the manual labeling comprises: based on the result of the manual labeling, extracting features of each no-information-point-coverage area on the satellite map by using the pre-trained satellite map feature extraction model.

4. The method of claim 3, wherein, the satellite map feature extraction model is trained by the following method: based on a satellite map, obtaining a corresponding satellite base map; based on the satellite base map, obtaining a satellite base map data set; based on the satellite base map data set, training a satellite map feature extraction model to be trained.

5. The method of claim 4, wherein, the satellite base map data set comprises positive samples and negative samples; the step of obtaining a satellite base map data set based on the satellite base map comprises: based on the satellite base map, obtaining each sub-satellite base map of a preset size; for each sub-satellite base map, obtaining each transformed base map corresponding to the sub-satellite base map by randomly translating and rotating the sub-satellite base map; for each sub-satellite base map, taking a preset number of corresponding transformed base maps as positive samples, and taking the remaining transformed base maps corresponding to the sub-satellite base map as negative samples. 6.The method of claim 1, further comprising: offline processing the abnormal information point in the common map. 7.A device for detecting a map information point, comprising: a target information point obtaining module configured to obtain a target information point to be detected in a common map; The target satellite map acquisition module is configured to acquire a target satellite map within a preset range of the target information point based on a position of the target information point. The target feature extraction module is configured to extract a target feature of the target satellite map. The target feature searching module is configured to search for the target feature in a pre-constructed no-information-point-coverage feature retrieval library. The abnormal information point determination module is configured to determine that the target information point is an abnormal information point if the target feature is found in the no-information-point-coverage feature retrieval library. The no-information-point-coverage feature retrieval library is pre-constructed by the following method: Artificially labeling each no-information-point-coverage area on a satellite map; Based on the result of the artificial labeling, extracting features of each no-information-point-coverage area on the satellite map; Based on the features of each no-information-point-coverage area, constructing the no-information-point-coverage feature retrieval library.

8. The apparatus of claim 7, wherein, The target feature searching module is specifically configured to calculate a similarity between the target feature and each feature in the no-information-point-coverage feature retrieval library, respectively. The abnormal information point determination module is specifically configured to determine that the target information point is an abnormal information point if there is a feature in the no-information-point-coverage feature retrieval library that has a similarity to the target feature exceeding a preset similarity threshold.

9. The apparatus of claim 7, wherein, The target feature extraction module is specifically configured to extract the target feature of the target satellite map by using a pre-trained satellite map feature extraction model. The extracting the features of each no-information-point-coverage area on the satellite map based on the result of the artificial labeling comprises: Based on the result of the artificial labeling, extracting the features of each no-information-point-coverage area on the satellite map by using the pre-trained satellite map feature extraction model.

10. The apparatus of claim 9, wherein, The satellite map feature extraction model is trained by the following method: Based on a satellite map, acquiring a corresponding satellite base map; Based on the satellite base map, acquiring a satellite base map data set; Based on the satellite base map data set, training a satellite map feature extraction model to be trained.

11. The apparatus of claim 10, wherein, The satellite base map data set includes positive samples and negative samples. The acquiring the satellite base map data set based on the satellite base map comprises: Based on the satellite base map, acquiring each sub-satellite base map of a preset size; For each sub-satellite base map, performing random translation and random rotation on the sub-satellite base map to acquire each transformed base map corresponding to the sub-satellite base map; For each sub-satellite base map, taking a preset number of corresponding transformed base maps as positive samples, and taking the remaining transformed base maps corresponding to the sub-satellite base map as negative samples.

12. The apparatus of claim 7, further comprising: An abnormal information point offline module configured to offline the abnormal information point in a normal map.

13. An electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein, The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

14. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are for causing the computer to perform the method of any one of claims 1-6.

15. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-6.

Images