A point of interest failure identification method, device, equipment and storage medium

Abstract

The application discloses a point of interest failure identification method, device and equipment and a storage medium. The application embodiment can be applied to various scenes such as artificial intelligence. The point of interest failure identification method comprises the following steps: obtaining a point of interest image corresponding to a to-be-identified point of interest and first position information corresponding to at least one point of interest identifier in the point of interest image; inputting the point of interest image into a failure feature detection network to perform failure feature detection, and obtaining second position information corresponding to at least one failure target in the point of interest image; and performing failure identification on the to-be-identified point of interest based on the first position information and the second position information, and obtaining failure identification information of the to-be-identified point of interest. By using the technical solution provided in the application, the failure feature of an intuitive point of interest image is mined, and the failure target corresponding to the point of interest identifier is determined based on the position information, so that the accuracy and convenience of point of interest failure identification can be improved, and the identification rate of the failure point of interest can be improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device and storage medium for identifying point of interest failure. Background Technology

[0002] With the development of information technology, electronic maps have brought convenience to people's lives, and Points of Interest (POIs) are an indispensable component of electronic maps. In practical applications, the geographic entities corresponding to POIs may change over time due to closure or information updates, causing the corresponding POIs in the electronic map data to become invalid. To avoid the negative impact of invalid POIs on map functionality and user experience, it is necessary to identify and handle invalid POIs in a timely and effective manner.

[0003] Currently, the identification of invalid points of interest (POIs) mainly relies on user feedback data. For example, when a user reports on the platform that a POI has been relocated or demolished, it can be identified as an invalid POI. However, this identification method is inefficient. Therefore, a more convenient and accurate technical solution is needed. Summary of the Invention

[0004] This application provides a method, apparatus, device, and storage medium for identifying point-of-interest (POI) failures. By mining failure features from intuitive POI images and determining the failure target corresponding to the POI identifier based on location information, it can improve the accuracy and convenience of POI failure identification while increasing the identification rate of failed POIs. The technical solution of this application is as follows:

[0005] On the one hand, a method for identifying point-of-interest (POI) failures is provided, the method comprising:

[0006] Obtain the interest point image corresponding to the interest point to be identified and the first location information corresponding to at least one interest point identifier in the interest point image;

[0007] The point of interest image is input into a failure feature detection network to detect failure features, thereby obtaining the second location information corresponding to at least one failed target in the point of interest image.

[0008] Based on the first location information and the second location information, failure identification is performed on the point of interest to be identified to obtain failure identification information of the point of interest to be identified.

[0009] On the other hand, an interest point failure identification device is provided, the method comprising:

[0010] The information acquisition module is used to acquire the interest point image corresponding to the interest point to be identified and the first location information corresponding to at least one interest point identifier in the interest point image;

[0011] The failure feature detection module is used to input the point of interest image into the failure feature detection network to detect failure features and obtain the second location information corresponding to at least one failure target in the point of interest image.

[0012] The failure identification module is used to identify the failure of the point of interest to be identified based on the first location information and the second location information, and obtain the failure identification information of the point of interest to be identified.

[0013] On the other hand, an interest point failure identification device is provided, the device including a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the interest point failure identification method as described in the first aspect.

[0014] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction or at least one program is stored therein, the at least one instruction or the at least one program being loaded and executed by a processor to implement the point of interest failure identification method as described in the first aspect.

[0015] On the other hand, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the point-of-interest failure identification method as described in the first aspect.

[0016] The method, apparatus, device, and storage medium for identifying point-of-interest (POI) failures provided in this application have the following technical advantages:

[0017] In the application scenario of point of interest (POI) failure identification, this application obtains the POI image corresponding to the POI to be identified and the first location information corresponding to at least one POI identifier in the POI image. Then, the POI image is input into a failure feature detection network for failure feature detection to obtain the second location information corresponding to at least one failure target in the POI image. Based on the first and second location information, failure identification is performed on the POI to be identified to obtain the failure identification information of the POI to be identified. The technical solution of this application can perform failure feature mining on intuitive POI images. When the POI image includes at least one POI identifier and at least one failure target, the failure target corresponding to each POI identifier is determined based on the location information. This can improve the accuracy and convenience of POI failure identification while increasing the identification rate of failure POIs, thereby improving the quality of POI data in electronic maps and enhancing the user experience. Attached Figure Description

[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of an application environment provided in an embodiment of this application;

[0020] Figure 2 This is a flowchart illustrating a method for identifying point-of-interest failures provided in an embodiment of this application;

[0021] Figure 3 This is a schematic diagram of a point of interest identifier provided in an embodiment of this application;

[0022] Figure 4 This is a flowchart illustrating a failure feature detection network training method provided in an embodiment of this application;

[0023] Figures 5a-5d This is a schematic diagram of various initial failure targets provided in the embodiments of this application;

[0024] Figure 6 This is a flowchart illustrating a method for screening failed targets according to an embodiment of this application;

[0025] Figure 7 This is a schematic diagram of a process provided in this application embodiment to perform failure identification on a point of interest to be identified based on first location information and second location information, and obtain failure identification information of the point of interest to be identified;

[0026] Figure 8 This is a schematic diagram of a process provided in this application embodiment to perform position analysis on at least one point of interest identifier and at least one failed target based on first location information and second location information, and obtain positional relationship information between at least one point of interest identifier and at least one failed target;

[0027] Figure 9 This is a schematic diagram of a process for identifying failure information of a point of interest based on location relationship information, provided in an embodiment of this application.

[0028] Figure 10 This is a flowchart illustrating a process provided in this application embodiment of performing positional analysis on the identifier to be analyzed and the target to be analyzed based on the first positional information of the identifier to be analyzed and the second positional information of the target to be analyzed, to obtain the positional relationship information between the identifier to be analyzed and the target to be analyzed;

[0029] Figure 11 This is a schematic diagram of a point of interest image provided in an embodiment of this application;

[0030] Figure 12 This is a block diagram of a point of interest failure identification device provided in an embodiment of this application;

[0031] Figure 13 This is a schematic diagram of the structure of a point of interest failure identification device provided in an embodiment of this application. Detailed Implementation

[0032] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0033] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, products or devices.

[0034] It is understood that in the specific embodiments of this application, data such as user information are involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0035] Please see Figure 1 , Figure 1This is a schematic diagram of an application environment provided in an embodiment of this application. This application environment may include a client 10 and a server 20, which can be indirectly connected via wireless communication. A user can send a point of interest (POI) failure identification request to the server 20 through the client 10. This POI failure identification request carries an POI image corresponding to the POI to be identified and first location information corresponding to at least one POI identifier in the POI image. In response to the POI failure identification request, the server 20 inputs the POI image into a failure feature detection network for failure feature detection, obtains second location information corresponding to at least one failed target in the POI image, and then, based on the first and second location information, performs failure identification on the POI to be identified, obtaining failure identification information for the POI to be identified, and returns the failure identification information to the client 10. It should be noted that... Figure 1 This is just one example.

[0036] The client can be a physical device such as a smartphone, computer (e.g., desktop computer, tablet computer, laptop computer), digital assistant, smart voice interaction device (e.g., smart speaker), smart wearable device, in-vehicle terminal, etc., or it can be software running on the physical device, such as a computer program. The operating system corresponding to the first client can be Android, iOS (a mobile operating system developed by Apple), Linux (an operating system), Microsoft Windows, etc.

[0037] The server side can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The server may include network communication units, processors, and memory, etc. The server side can provide backend services to the corresponding clients.

[0038] The aforementioned client 10 and server 20 can be used to build a system for identifying points of interest failures. This system can be a distributed system. Taking a blockchain system as an example, the distributed system consists of multiple nodes (any form of computing device connected to the network, such as servers or user terminals) and clients. These nodes form a peer-to-peer (P2P) network. The P2P protocol is an application layer protocol running on top of the Transmission Control Protocol (TCP). In a distributed system, any machine, such as a server or terminal, can join and become a node. A node includes a hardware layer, a middleware layer, an operating system layer, and an application layer.

[0039] The functions of each node in the aforementioned blockchain system include:

[0040] 1) Routing: A basic function of nodes used to support communication between nodes.

[0041] In addition to routing capabilities, nodes can also have the following functions:

[0042] 2) Applications are deployed in the blockchain to implement specific business needs. They record data related to the implementation of functions to form record data, carry digital signatures in the record data to indicate the source of the task data, and send the record data to other nodes in the blockchain system. When other nodes successfully verify the source and integrity of the record data, they add the record data to a temporary block.

[0043] 3) A blockchain consists of a series of blocks that are sequentially generated. Once a new block is added to the blockchain, it will not be removed. The blocks contain the data submitted by the nodes in the blockchain system.

[0044] The following describes a specific embodiment of the point of interest failure identification method provided in this application. Figure 2 This is a flowchart illustrating a method for identifying point-of-interest (POI) failures according to an embodiment of this application. This application provides the operational steps described in the embodiments or flowchart, but based on conventional or non-inventive methods, it may include more or fewer operational steps. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only possible execution order. In actual systems or products, the methods can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment) as shown in the embodiments or drawings. Specifically, as... Figure 2 As shown, the method may include:

[0045] S201, obtain the interest point image corresponding to the interest point to be identified and the first location information corresponding to at least one interest point identifier in the interest point image.

[0046] In the embodiments of this specification, the point of interest to be identified can be a location marked on an electronic map. The point of interest image can be a captured image of the physical environment corresponding to the point of interest to be identified. Specifically, the point of interest image may contain point of interest markers, which may include, but are not limited to, signs, placards, and other physical landmarks that can identify the identity information of the point of interest.

[0047] In one specific embodiment, the point of interest to be identified may include at least one point of interest, which may correspond to the same point of interest image. That is, the point of interest image may display the physical environment of the at least one point of interest. Specifically, the point of interest image may include the point of interest identifier corresponding to the at least one point of interest.

[0048] In the embodiments of this specification, the first location information can be the location information of the bounding box corresponding to the interest point marker in the interest point image. In practical applications, considering the excessive workload of manual annotation, the first location information of the interest point marker can be obtained in advance by performing interest point marker detection on the interest point image based on an interest point marker detection network. Specifically, the interest point marker detection network can include, but is not limited to, R-CNN, Faster R-CNN, etc.

[0049] See Figure 3 , Figure 3 This is a schematic diagram of a point of interest identifier provided in an embodiment of this application. Figure 3 It includes a box indicating a point of interest, such as "a noodle shop".

[0050] S202, input the point of interest image into the failure feature detection network to perform failure feature detection, and obtain the second location information corresponding to at least one failed target in the point of interest image.

[0051] In the embodiments of this specification, the failed target can be a target object that can characterize the failure features of the point of interest. Specifically, the failed target may include, but is not limited to: door lock target, roller shutter lock target, and posting target containing the words "renovation / transfer / relocation / demolition / rental", etc. Preferably, the failed target can be: door lock target and roller shutter lock target.

[0052] In the embodiments of this specification, the second location information may be the location information of the detection box corresponding to the failed target in the point of interest image.

[0053] In practical applications, applicants observe relevant data on historically failed points of interest (POIs) to identify various initial failure targets that can serve as criteria for determining whether a POI is failed. Specifically, these initial failure targets may include, but are not limited to, door lock targets (such as...). Figure 5a As shown), the target of the roller shutter lock (such as...) Figure 5b As shown), posted objects containing the text "renovation / transfer / relocation / demolition / rental" (e.g.) Figure 5c As shown), housing vacancy targets (such as...) Figure 5d (as shown in the image) etc.

[0054] See Figure 6 , Figure 6 This is a flowchart illustrating a method for screening failed targets according to an embodiment of this application. Specifically, the method for screening failed targets may include:

[0055] S601, acquire the second sample interest point images corresponding to various initial failed targets.

[0056] Specifically, the second sample interest point image can be a sample interest point image containing the corresponding initial failed target.

[0057] S602, based on the second sample interest point image corresponding to each initial failure target, perform failure feature detection training on the preset failure feature detection network to obtain the network training result corresponding to each initial failure target.

[0058] Specifically, the network training results can include the detection accuracy, recall, and F1 score for each initial failed target. Detection accuracy represents the proportion of samples that actually contain the corresponding initial failed target among those that are detected. Recall represents the proportion of positive samples that are detected containing the corresponding initial failed target out of all positive samples. The F1 score can comprehensively represent the detection accuracy and recall, where F1 score = (2 × recall × detection accuracy) / (recall + detection accuracy).

[0059] S603, based on the network training results corresponding to multiple initial failure targets, filters multiple initial failure targets to determine the failure target.

[0060] Specifically, the above-mentioned screening of multiple initial failure targets based on the network training results corresponding to multiple initial failure targets to determine failure targets may include: selecting the initial failure targets whose corresponding network training results meet the preset screening conditions as failure targets.

[0061] In practical applications, preset filtering conditions can be pre-set based on the requirements for detecting point of interest failure. In a specific embodiment, when the network training results include detection accuracy, recall, and F1 score, the preset filtering conditions can be the highest detection accuracy and / or the highest recall and / or the highest F1 score.

[0062] In one specific embodiment, images of 1870 failed points of interest were obtained from historical point of interest images. Through mining and analysis of these 1870 failed point of interest images, the frequency of occurrence of four types of initial failed targets was found to be as follows: door lock targets 652 times, roller shutter lock targets 657 times, posted targets 208 times, and vacant house targets 21 times. Correspondingly, door lock targets accounted for 42% of the total number of targets, roller shutter lock targets accounted for 43%, posted targets accounted for 14%, and vacant house targets accounted for 1%.

[0063] After labeling door lock targets, roller shutter lock targets, posted objects targets, and vacant house targets in 1870 failed point of interest images, a pre-designed failure feature detection network was trained based on the labeled failed point of interest images. To approximate the actual situation as closely as possible, the ratio of the training set, validation set, and test set in the 1870 failed point of interest images was 6:1:3. The final accuracy, recall, and F1 score of the pre-designed failure feature detection network on the test set for the four initial failed targets are shown in the table below:

[0064]

[0065] A comparative analysis of the relevant data for the four initial failed targets revealed that the data volume for posted objects and vacant houses was too small, and their frequency of occurrence was low. Secondly, from the model training results, the detection accuracy of posted objects was less than 50%, and the recall rate was less than 55%, indicating that the model could not learn the features of posted objects well. The model training results for vacant houses were similar. On the other hand, door locks and roller shutter locks were two initial failed targets with high differences and significance, and their image numbers accounted for a relatively large proportion. Therefore, door locks and roller shutter locks were identified as failed targets.

[0066] As can be seen from the above embodiments, by utilizing the network training results corresponding to multiple initial failed targets, multiple initial failed targets are screened to determine the failed targets, so as to ensure the accuracy of failed target detection and thus improve the accuracy of interest point failure detection.

[0067] In the embodiments described in this specification, such as Figure 4 As shown, the above failure feature detection network can be obtained by training the network in the following way:

[0068] S401, Obtain a first sample interest point image containing annotation location information corresponding to at least one annotation failure target.

[0069] In practical applications, training data can be determined before network training. Specifically, in this embodiment, a first sample interest point image containing annotation location information corresponding to at least one labeled invalid target can be obtained as training data.

[0070] Specifically, the labeled failed targets can be the failed targets pre-labeled in the first sample point of interest image, and the labeling location information can be the location information of the corresponding label box in the first sample point of interest image.

[0071] S402, the first sample interest point image is input into a preset failure feature detection network to perform failure feature detection, and the predicted location information corresponding to at least one sample failure target in the first sample interest point image is obtained.

[0072] S403, based on the labeled location information and the predicted location information, determines the target loss information.

[0073] S404. Based on the target loss information, a preset failure feature detection network is trained to obtain the failure feature detection network.

[0074] In an optional embodiment, the target loss information may include the failed target location loss; correspondingly, determining the target loss information based on the labeled location information and the predicted location information may include: determining the failed target location loss based on the labeled location information and the predicted location information.

[0075] In one specific embodiment, determining the failure target location loss based on the labeled location information and the predicted location information may include determining the failure target location loss between the labeled location information and the predicted location information based on a preset loss function.

[0076] In one specific embodiment, the failure target location loss can characterize the difference between the labeled location information and the predicted location information.

[0077] In a specific embodiment, the preset loss function may include, but is not limited to, the cross-entropy loss function, the logistic loss function, the exponential loss function, etc.

[0078] In an optional embodiment, a preset failure feature detection network is trained based on the target loss information. The resulting failure feature detection network may include:

[0079] S4041, based on the target loss information, update the network parameters of the failure feature detection network.

[0080] S4042, Based on the updated failure feature detection network, repeat the failure feature detection training iteration operation of steps S402, S403 and S4041 until the detection convergence condition is met; the preset failure feature detection network obtained when the detection convergence condition is met is used as the failure feature detection network.

[0081] In an optional embodiment, the aforementioned convergence detection condition can be that the number of training iterations reaches a preset number of training iterations. Optionally, the convergence detection condition can also be that the target loss information is less than a specified threshold. In the embodiments of this specification, the preset number of training iterations and the specified threshold can be preset in conjunction with the training speed and accuracy of the network in practical applications.

[0082] As can be seen from the above embodiments, training the preset failure feature detection network with sample interest point images that include the labeled location information corresponding to the failed target can improve the generalization ability of the failure feature detection network, thereby improving the accuracy of interest point failure detection.

[0083] Furthermore, it should be noted that the failure feature detection network described in this application embodiment is not limited to the aforementioned preset failure feature detection network. In practical applications, it may also include other machine learning networks, such as R-CNN, Faster R-CNN, etc. This application embodiment is not limited to the aforementioned machine learning networks.

[0084] In a specific implementation, the failure feature detection network may include: an image feature extraction layer, an image feature aggregation layer, and a detection result output layer. Specifically, the image feature extraction layer can be used to scale the point of interest image at different scales, and then extract image features from the point of interest images at different scales; the image feature aggregation layer can be used to aggregate image features at different scales to generate a feature pyramid, enabling the network to have scale invariance; the detection result output layer can output the detection of the failure target based on the feature pyramid, that is, to detect the failure target on the point of interest images at different scales through anchor boxes, thereby generating a detection result with detection category, detection category confidence, and detection box position information.

[0085] In one specific embodiment, when the type of the failed target is a door lock target or a roller shutter lock target, the detection category may include: door lock target, roller shutter lock target and non-failed target.

[0086] As can be seen from the above embodiments, the failure feature detection network performs scaling processing on the point of interest image at different scales, then extracts image features from the point of interest images at different scales, and aggregates the image features at different scales to generate a feature pyramid. Based on the feature pyramid, the network outputs the detection of failure targets, which can fully detect failure targets in point of interest images at different scales and improve the accuracy of failure target detection.

[0087] S203, based on the first location information and the second location information, perform failure identification on the interest point to be identified, and obtain failure identification information of the interest point to be identified.

[0088] In the embodiments of this specification, failure identification information can be used to identify whether a point of interest to be identified is invalid. Specifically, failure identification information may include: point of interest invalid, point of interest valid, and point of interest invalid and unidentifiable.

[0089] In a specific embodiment, such as Figure 7 As shown, the failure identification of the interest point to be identified based on the first location information and the second location information can include the following:

[0090] S701, based on the first location information and the second location information, perform location analysis on at least one point of interest identifier and at least one failed target to obtain the location relationship information between at least one point of interest identifier and at least one failed target.

[0091] S702, based on location relationship information, performs failure identification on the points of interest to be identified, and obtains failure identification information of the points of interest to be identified.

[0092] In the embodiments of this specification, the positional relationship information can characterize the positional information of the failed target in the point of interest image relative to the point of interest marker. Specifically, the positional relationship information can include: horizontal region overlap information, vertical orientation relationship information, and distance information. The horizontal region overlap information can characterize the horizontal region overlap ratio between the failed target and the point of interest marker, and the vertical orientation relationship information can characterize the vertical orientation relationship between the failed target and the point of interest marker.

[0093] Specifically, when there is at least one point of interest (POI) identifier and at least one failed target in the point of interest image, position analysis is performed on each POI identifier and the at least one failed target to determine whether there is a matching failed target for each POI identifier. If there is a matching failed target for each POI identifier, failure identification is performed on the POI corresponding to each POI identifier based on the matching failed target.

[0094] See Figure 11 , Figure 11This is a schematic diagram of a point-of-interest image provided in an embodiment of this application. Specifically, Figure 11 The point of interest (POI) image contains three POI identifiers and three failed targets. After performing positional analysis on the three POI identifiers and the three failed targets, it is determined that POI identifier 1 matches failed target 1, POI identifier 2 matches failed target 2, and POI identifier 3 matches failed target 3. Therefore, POI identifier 1 is detected as a failed target based on failed target 1, POI identifier 2 is detected as a failed target based on failed target 2, and POI identifier 3 is detected as a failed target based on failed target 3.

[0095] As can be seen from the above embodiments, based on the first location information and the second location information, position analysis is performed on at least one point of interest identifier and at least one failed target, and based on the position relationship information, failure identification is performed on the point of interest to be identified. By determining the failed target corresponding to each point of interest identifier, the accuracy of failure identification of the point of interest corresponding to each point of interest identifier is improved.

[0096] In a specific embodiment, such as Figure 8 As shown, the above-mentioned location analysis of at least one point of interest identifier and at least one failed target based on the first location information and the second location information, to obtain the location relationship information between at least one point of interest identifier and at least one failed target, may include:

[0097] S801, determine the identifier to be analyzed from at least one point of interest identifier, wherein the identifier to be analyzed is any point of interest identifier that has not undergone location analysis from at least one point of interest identifier.

[0098] S802, determine the target to be analyzed from at least one failed target, wherein the target to be analyzed is any unmatched failed target among at least one failed target.

[0099] S803, based on the first location information of the identifier to be analyzed and the second location information of the target to be analyzed, perform position analysis on the identifier to be analyzed and the target to be analyzed to obtain the positional relationship information between the identifier to be analyzed and the target to be analyzed.

[0100] Specifically, each of the at least one point of interest identifiers is taken as the identifier to be analyzed, and the current identifier to be analyzed is compared with the current unmatched failed target to obtain the positional relationship information between the two.

[0101] As can be seen from the above embodiments, by determining the identifier to be analyzed from at least one point of interest identifier, wherein the identifier to be analyzed is any point of interest identifier that has not undergone location analysis, and by determining the target to be analyzed from at least one failed target, wherein the target to be analyzed is any failed target that has not been matched, it is possible to avoid re-matching failed targets that have already been successfully matched with a certain point of interest identifier. This improves the efficiency of location analysis of at least one point of interest identifier and at least one failed target, while also improving the accuracy of matching determination between point of interest identifier and failed target.

[0102] In a specific embodiment, such as Figure 9 As shown, the points of interest to be identified include: the target points of interest corresponding to the identifier to be analyzed. Based on the location relationship information, the above-mentioned failure identification of the points of interest to be identified can be performed to obtain the failure identification information of the points of interest to be identified, which may include:

[0103] S901, if the positional relationship information between the identifier to be analyzed and the target to be analyzed meets the preset positional conditions, the target to be analyzed is used as the matching target of the identifier to be analyzed.

[0104] S902, based on the matching target, perform failure identification on the target interest point to obtain failure identification information of the target interest point.

[0105] Specifically, the preset location conditions can be pre-set based on the type of interest point to be identified and the accuracy requirements of matching failed targets in actual applications.

[0106] In a specific embodiment, when the positional relationship information includes horizontal region overlap information and vertical orientation relationship information, the preset positional conditions may include: the vertical orientation relationship information indicating that the target to be analyzed is located below the identifier to be analyzed, and the horizontal region overlap information indicating that the horizontal region overlap ratio between the target to be analyzed and the identifier to be analyzed is greater than a preset ratio threshold. This preset ratio threshold can be pre-set based on the accuracy requirements of matching the identifier to be analyzed and the target to be analyzed in practical applications.

[0107] In an optional embodiment, the failure identification of target interest points based on the matching target, to obtain failure identification information of target interest points, may include:

[0108] When the identifier to be analyzed has a matching target, the failure of the point of interest is used as the failure identification information of the target point of interest.

[0109] In another optional embodiment, the above-described failure identification of target interest points based on matching targets to obtain failure identification information of target interest points may include:

[0110] If no matching target exists for the identifier to be analyzed, the valid points of interest are used as the failure identification information of the target points of interest.

[0111] As can be seen from the above embodiments, when the positional relationship information between the identifier to be analyzed and the target to be analyzed meets the preset positional conditions, the target to be analyzed is used as the matching target of the identifier to be analyzed, and based on the matching target, the failure identification of the target interest point corresponding to the identifier to be analyzed can be performed. This can improve the accuracy of the matching determination between the identifier to be analyzed and the target to be analyzed, while also improving the accuracy of the failure identification of the target interest point.

[0112] In the embodiments of this specification, the first location information may include first coordinate information, and the second location information may include second coordinate information. Specifically, the first coordinate information may include the coordinate information of the annotation box corresponding to the point of interest identifier, and the second coordinate information may include the coordinate information of the detection box corresponding to the failed target. Generally, the coordinate information here is the coordinate in a Cartesian coordinate system.

[0113] In a specific embodiment, such as Figure 10 As shown, when the positional relationship information includes horizontal region overlap information and vertical orientation relationship information, the positional analysis performed on the identifier to be analyzed and the target to be analyzed based on the first positional information of the identifier to be analyzed and the second positional information of the target to be analyzed, to obtain the positional relationship information between the identifier to be analyzed and the target to be analyzed, may include:

[0114] S1001, based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, perform longitudinal orientation analysis on the identifier to be analyzed and the target to be analyzed, and generate longitudinal orientation relationship information between the identifier to be analyzed and the target to be analyzed.

[0115] S1002, based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, perform lateral region overlap analysis on the identifier to be analyzed and the target to be analyzed to generate lateral region overlap information between the identifier to be analyzed and the target to be analyzed.

[0116] Assume that the minimum and maximum x-coordinates of the marker to be analyzed are Xb_min and Xb_max, respectively, and the coordinates of the marker center point are (Xb_center, Yb_center); the minimum and maximum x-coordinates of the target to be analyzed are Xe_min and Xe_max, respectively, and the coordinates of the target center point are (Xe_center, Ye_center).

[0117] Specifically, based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, a longitudinal orientation analysis is performed on the identifier and the target to be analyzed to generate longitudinal orientation relationship information between them. This can include: generating longitudinal orientation relationship information between the identifier and the target based on a comparison of the ordinate of the identifier's center point and the target's center point. Generally, when the ordinate of the identifier's center point is greater than the ordinate of the target's center point, the longitudinal orientation relationship information indicates that the target is below the identifier; when the ordinate of the identifier's center point is less than the ordinate of the target's center point, the longitudinal orientation relationship information indicates that the target is above the identifier.

[0118] Specifically, based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, a lateral region overlap analysis is performed on the identifier and the target to be analyzed to generate lateral region overlap information between them, which may include:

[0119] Obtain the intersection and union intervals of the horizontal coordinates between the identifier and the target to be analyzed; based on the intersection and union intervals, obtain the horizontal region overlap ratio between the identifier and the target to be analyzed; use the horizontal region overlap ratio as the horizontal region overlap information.

[0120] Specifically, the intersection interval of the horizontal coordinates, Xin_area, is:

[0121] max(0, min(Xb_max , Xe_max)- max(Xb_min , Xe_min));

[0122] The union of the x-coordinates, the interval Xun_area, is:

[0123] max(0, max(Xb_max , Xe_max)- min(Xb_min , Xe_min));

[0124] The horizontal region overlap ratio is obtained as Xin_area / Xun_area.

[0125] As can be seen from the above embodiments, by performing longitudinal orientation analysis and lateral region overlap analysis on the identifier and the target to be analyzed, the accuracy of matching determination between the identifier and the target to be analyzed can be improved.

[0126] In an optional embodiment, before performing failure identification on the point of interest to be identified based on the first location information and the second location information to obtain failure identification information of the point of interest to be identified, the above method may further include:

[0127] 1) Obtain the acquisition time of the point of interest image;

[0128] Accordingly, the failure identification of the interest point to be identified based on the first location information and the second location information, and the resulting failure identification information of the interest point to be identified, may include:

[0129] 2) When the collection time falls within the preset valid time period, based on the first location information and the second location information, the point of interest to be identified is invalidated to obtain invalidation identification information.

[0130] Specifically, the preset effective time period can be a pre-defined time period during which the failed target can effectively characterize the failure features of the point of interest. In practical applications, the preset effective time period can be preset in combination with the region where the point of interest to be identified is located and the type of point of interest to be identified. Preferably, the preset effective time period can be the normal business hours of the point of interest to be identified recorded in the electronic map platform, for example, the preset effective time period can be 9:00-21:00.

[0131] As can be seen from the above embodiments, when the collection time is within a preset valid time period, failure identification of the point of interest to be identified is performed based on the first location information and the second location information to obtain failure identification information. This can avoid misjudgment of point of interest failure identification caused by the collection time being within a time period when the failure target cannot characterize the failure features of the point of interest, thereby improving the accuracy of point of interest failure identification.

[0132] This application also provides a point of interest failure identification device, such as... Figure 12 As shown, the point of interest failure identification device may include:

[0133] Information acquisition module 1210 is used to acquire the interest point image corresponding to the interest point to be identified and the first location information corresponding to at least one interest point identifier in the interest point image;

[0134] The failure feature detection module 1220 is used to input the point of interest image into the failure feature detection network to detect failure features and obtain the second location information corresponding to at least one failed target in the point of interest image.

[0135] The failure identification module 1230 is used to identify the failure of the point of interest to be identified based on the first location information and the second location information, and obtain the failure identification information of the point of interest to be identified.

[0136] In one specific embodiment, the above-described apparatus may further include:

[0137] The first sample acquisition module is used to acquire second sample interest point images corresponding to various initial failed targets;

[0138] The network training result module is used to train the preset failure feature detection network based on the second sample interest point image corresponding to each initial failure target, and obtain the network training result corresponding to each initial failure target.

[0139] The failure target determination module is used to filter multiple initial failure targets based on the network training results corresponding to multiple initial failure targets, and determine the failure target.

[0140] In the embodiments of this specification, the above-mentioned failure feature detection network can be obtained by training the network using the following device:

[0141] The second sample acquisition module is used to acquire a first sample interest point image containing annotation location information corresponding to at least one labeled invalid target.

[0142] The sample failure feature detection module is used to input the first sample interest point image into a preset failure feature detection network to detect failure features and obtain the predicted location information of at least one sample failure target in the first sample interest point image.

[0143] The target loss information determination module is used to determine the target loss information based on the labeled location information and the predicted location information;

[0144] The network training module is used to train a preset failure feature detection network based on the target loss information, thereby obtaining the failure feature detection network.

[0145] In one specific embodiment, the failure identification module 1230 described above may include:

[0146] The location analysis unit is used to perform location analysis on at least one point of interest identifier and at least one failed target based on first location information and second location information, so as to obtain the location relationship information between at least one point of interest identifier and at least one failed target.

[0147] The first failure identification unit is used to identify the failure of the point of interest to be identified based on the location relationship information, and obtain the failure identification information of the point of interest to be identified.

[0148] In one specific embodiment, the above-mentioned position analysis unit may include:

[0149] The identifier to be analyzed is used to determine the identifier to be analyzed from at least one point of interest identifier, wherein the identifier to be analyzed is any point of interest identifier that has not undergone location analysis among at least one point of interest identifier;

[0150] The target unit to be analyzed is used to determine the target to be analyzed from at least one failed target, wherein the target to be analyzed is any unmatched failed target among at least one failed target;

[0151] The location relationship information unit is used to perform location analysis on the identifier to be analyzed and the target to be analyzed based on the first location information of the identifier to be analyzed and the second location information of the target to be analyzed, so as to obtain the location relationship information between the identifier to be analyzed and the target to be analyzed.

[0152] In a specific embodiment, the first location information may include first coordinate information, the second location information may include second coordinate information, and the location relationship information may include: horizontal region overlap information and vertical orientation relationship information. The aforementioned location relationship information unit may include:

[0153] The longitudinal orientation analysis unit is used to perform longitudinal orientation analysis on the identifier to be analyzed and the target to be analyzed based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, and to generate longitudinal orientation relationship information between the identifier to be analyzed and the target to be analyzed.

[0154] The horizontal region overlap analysis unit is used to perform horizontal region overlap analysis on the identifier to be analyzed and the target to be analyzed based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, and generate horizontal region overlap information between the identifier to be analyzed and the target to be analyzed.

[0155] In one specific embodiment, the interest point to be identified includes: the target interest point corresponding to the identifier to be analyzed, and the first failure identification unit may include:

[0156] The target matching unit is used to select the target to be analyzed as the matching target of the target to be analyzed when the positional relationship information between the identifier to be analyzed and the target to be analyzed meets the preset positional conditions.

[0157] The failure identification information unit is used to identify the failure of the target interest point based on the matching target, and obtain the failure identification information of the target interest point.

[0158] In an optional embodiment, the above-described apparatus may further include:

[0159] The acquisition time acquisition unit is used to acquire the acquisition time of the point of interest image;

[0160] Accordingly, the failure identification module 1230 mentioned above may include:

[0161] The second failure identification unit is used to identify the point of interest to be identified based on the first location information and the second location information when the collection time is within a preset valid time period, and to obtain failure identification information.

[0162] It should be noted that the apparatus and method embodiments described above are based on the same inventive concept.

[0163] This application provides a point of interest failure identification device, which includes a processor and a memory. The memory stores at least one instruction or at least one program segment. The at least one instruction or at least one program segment is loaded and executed by the processor to implement the point of interest failure identification method provided in the above method embodiments.

[0164] Furthermore, Figure 13 A schematic diagram of the hardware structure of an interest point failure identification device for implementing the interest point failure identification method provided in the embodiments of this application is shown. The interest point failure identification device can participate in or include the interest point failure identification apparatus provided in the embodiments of this application. Figure 13 As shown, the point-of-interest failure identification device 130 may include one or more processors 1302 (shown as 1302a, 1302b, ..., 1302n in the figure) 1302 (processor 1302 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 1304 for storing data, and a transmission device 1306 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 13 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, the point of interest failure identification device 130 may also include a... Figure 13 The more or fewer components shown, or having the same Figure 13 The different configurations shown.

[0165] It should be noted that the aforementioned one or more processors 1302 and / or other data processing circuitry are generally referred to herein as "data processing circuitry". This data processing circuitry may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be wholly or partially integrated into any other element within the point of interest failure identification device 130 (or mobile device). As described in the embodiments of this application, this data processing circuitry serves as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0166] The memory 1304 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the point of interest failure identification method described in this embodiment. The processor 1302 executes various functional applications and data processing by running the software programs and modules stored in the memory 1304, thereby implementing the aforementioned point of interest failure identification method. The memory 1304 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 1304 may further include memory remotely located relative to the processor 1302, and these remote memories can be connected to the point of interest failure identification device 130 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0167] The transmission device 1306 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the point of interest failure identification device 130. In one example, the transmission device 1306 includes a network interface controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In one embodiment, the transmission device 1306 may be a radio frequency (RF) module for wireless communication with the Internet.

[0168] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the point of interest failure identification device 130 (or mobile device).

[0169] Embodiments of this application also provide a computer-readable storage medium, which can be disposed in an interest point failure identification device to store at least one instruction or at least one program related to implementing the interest point failure identification method in the method embodiment. The at least one instruction or the at least one program is loaded and executed by the processor to implement the interest point failure identification method provided in the above method embodiment.

[0170] Optionally, in this embodiment, the storage medium may be located in at least one of the multiple network servers in a computer network. Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0171] Embodiments of this application also provide a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the point-of-interest failure identification method as provided in the method embodiments.

[0172] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are also possible or may be advantageous.

[0173] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and apparatus embodiments are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0174] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0175] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for identifying point-of-interest (POI) failures, characterized in that, The method includes: Obtain the interest point image corresponding to the interest point to be identified and the first location information corresponding to at least one interest point identifier in the interest point image; The point of interest image is input into a failure feature detection network for failure feature detection to obtain second location information corresponding to at least one failure target in the point of interest image that belongs to a preset detection category. Based on the first location information and the second location information, the point of interest to be identified is invalidated to obtain the invalidation information of the point of interest to be identified. The preset detection category is determined in the following way: Obtain second sample interest point images corresponding to failed targets of various initial detection categories; Based on the second sample interest point image corresponding to the failed target for each initial detection category, the preset failure feature detection network is trained to obtain the network training result corresponding to the failed target for each initial detection category. Based on the network training results corresponding to the failed targets of the various initial detection categories, the various initial detection categories are filtered to determine the preset detection category.

2. The method according to claim 1, characterized in that, The step of performing failure identification on the point of interest to be identified based on the first location information and the second location information to obtain failure identification information of the point of interest to be identified includes: Based on the first location information and the second location information, position analysis is performed on the at least one point of interest identifier and the at least one failed target to obtain positional relationship information between the at least one point of interest identifier and the at least one failed target; Based on the location relationship information, failure identification is performed on the point of interest to be identified to obtain failure identification information of the point of interest to be identified.

3. The method according to claim 2, characterized in that, The step of performing location analysis on the at least one point of interest identifier and the at least one failed target based on the first location information and the second location information to obtain the location relationship information between the at least one point of interest identifier and the at least one failed target includes: The identifier to be analyzed is determined from the at least one point of interest identifier, wherein the identifier to be analyzed is any point of interest identifier among the at least one point of interest identifier that has not undergone location analysis; The target to be analyzed is determined from the at least one failed target, wherein the target to be analyzed is any unmatched failed target among the at least one failed target; Based on the first location information of the identifier to be analyzed and the second location information of the target to be analyzed, position analysis is performed on the identifier to be analyzed and the target to be analyzed to obtain the positional relationship information between the identifier to be analyzed and the target to be analyzed.

4. The method according to claim 3, characterized in that, The points of interest to be identified include: the target points of interest corresponding to the identifiers to be analyzed; the failure identification of the points of interest to be identified based on the location relationship information, to obtain the failure identification information of the points of interest to be identified, includes: If the positional relationship information between the identifier to be analyzed and the target to be analyzed meets the preset positional conditions, the target to be analyzed is used as the matching target of the identifier to be analyzed. Based on the matching target, failure identification is performed on the target interest point to obtain failure identification information of the target interest point.

5. The method according to claim 3, characterized in that, The first location information includes first coordinate information, the second location information includes second coordinate information, and the location relationship information includes: horizontal region overlap information and vertical orientation relationship information. The step of performing location analysis on the identifier to be analyzed and the target to be analyzed based on the first location information of the identifier to be analyzed and the second location information of the target to be analyzed, to obtain the location relationship information between the identifier to be analyzed and the target to be analyzed, includes: Based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, longitudinal orientation analysis is performed on the identifier to be analyzed and the target to be analyzed to generate longitudinal orientation relationship information between the identifier to be analyzed and the target to be analyzed. Based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, a horizontal region overlap analysis is performed on the identifier to be analyzed and the target to be analyzed to generate horizontal region overlap information between the identifier to be analyzed and the target to be analyzed.

6. The method according to claim 1, characterized in that, Before performing failure identification on the point of interest to be identified based on the first location information and the second location information to obtain failure identification information of the point of interest to be identified, the method further includes: Obtain the acquisition time of the image of the point of interest; The step of performing failure identification on the point of interest to be identified based on the first location information and the second location information to obtain failure identification information of the point of interest to be identified includes: If the collection time falls within a preset valid time period, the point of interest to be identified is identified as faulty based on the first location information and the second location information, and the fault identification information is obtained.

7. The method according to any one of claims 1 to 6, characterized in that, The failure feature detection network is trained in the following manner: Obtain a first sample interest point image containing annotation location information corresponding to at least one labeled invalid target; The first sample point of interest image is input into a preset failure feature detection network to detect failure features, and the predicted location information of at least one sample failure target in the first sample point of interest image is obtained. Based on the labeled location information and the predicted location information, the target loss information is determined; Based on the target loss information, the preset failure feature detection network is trained to obtain the failure feature detection network.

8. A device for identifying point-of-interest (POI) failures, characterized in that, The device includes: The information acquisition module is used to acquire the interest point image corresponding to the interest point to be identified and the first location information corresponding to at least one interest point identifier in the interest point image; The failure feature detection module is used to input the point of interest image into the failure feature detection network for failure feature detection, and obtain the second location information corresponding to at least one failure target in the point of interest image that belongs to a preset detection category; The failure identification module is used to identify the failure of the point of interest to be identified based on the first location information and the second location information, and to obtain the failure identification information of the point of interest to be identified. The device further includes: The first sample acquisition module is used to acquire second sample interest point images corresponding to failed targets of multiple initial detection categories; The network training result module is used to train the preset failure feature detection network based on the second sample interest point image corresponding to the failure target of each initial detection category, and obtain the network training result corresponding to the failure target of each initial detection category. The preset detection category determination module is used to filter the multiple initial detection categories based on the network training results corresponding to the failed targets of the multiple initial detection categories, and determine the preset detection category.

9. The apparatus according to claim 8, characterized in that, The failure identification module includes: A location analysis unit is configured to perform location analysis on the at least one point of interest identifier and the at least one failed target based on the first location information and the second location information, and obtain location relationship information between the at least one point of interest identifier and the at least one failed target; The first failure identification unit is used to identify the failure of the point of interest to be identified based on the location relationship information, and obtain the failure identification information of the point of interest to be identified.

10. The apparatus according to claim 9, characterized in that, The location analysis unit includes: An identifier to be analyzed is used to determine an identifier to be analyzed from the at least one point of interest identifiers, wherein the identifier to be analyzed is any point of interest identifier among the at least one point of interest identifiers that has not undergone location analysis; The target unit to be analyzed is used to determine the target to be analyzed from the at least one failed target, wherein the target to be analyzed is any unmatched failed target among the at least one failed target; The location relationship information unit is used to perform location analysis on the identifier to be analyzed and the target to be analyzed based on the first location information of the identifier to be analyzed and the second location information of the target to be analyzed, so as to obtain the location relationship information between the identifier to be analyzed and the target to be analyzed.

11. The apparatus according to claim 10, characterized in that, The interest points to be identified include: the target interest points corresponding to the identifier to be analyzed, and the first failure identification unit includes: A target matching unit is used to use the target to be analyzed as the matching target of the target to be analyzed when the positional relationship information between the identifier to be analyzed and the target to be analyzed meets a preset positional condition. The failure identification information unit is used to identify the failure of the target interest point based on the matching target, and obtain the failure identification information of the target interest point.

12. The apparatus according to claim 10, characterized in that, The first location information includes first coordinate information, the second location information includes second coordinate information, and the location relationship information includes: horizontal region overlap information and vertical orientation relationship information. The location relationship information unit includes: The longitudinal orientation analysis unit is used to perform longitudinal orientation analysis on the identifier to be analyzed and the target to be analyzed based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, and generate longitudinal orientation relationship information between the identifier to be analyzed and the target to be analyzed. The horizontal region overlap analysis unit is used to perform horizontal region overlap analysis on the identifier to be analyzed and the target to be analyzed based on the first coordinate information of the identifier to be analyzed and the second coordinate information of the target to be analyzed, and generate horizontal region overlap information between the identifier to be analyzed and the target to be analyzed.

13. The apparatus according to claim 8, characterized in that, The device further includes: The acquisition time acquisition unit is used to acquire the acquisition time of the point of interest image; The failure identification module includes: The second failure identification unit is used to identify the point of interest to be identified based on the first location information and the second location information when the collection time is within a preset valid time period, and to obtain the failure identification information.

14. The apparatus according to any one of claims 8 to 13, characterized in that, The failure feature detection network is trained using the following device: The second sample acquisition module is used to acquire a first sample interest point image containing annotation location information corresponding to at least one labeled invalid target. The sample failure feature detection module is used to input the first sample interest point image into a preset failure feature detection network to detect failure features and obtain the predicted location information of at least one sample failure target in the first sample interest point image. The target loss information determination module is used to determine target loss information based on the labeled location information and the predicted location information; The network training module is used to train the preset failure feature detection network based on the target loss information to obtain the failure feature detection network.

15. A point-of-interest (POI) failure identification device, characterized in that, The device includes a processor and a memory, the memory storing at least one instruction or at least one program, the at least one instruction or the at least one program being loaded and executed by the processor to implement the point of interest failure identification method as described in any one of claims 1 to 7.

16. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction or at least one program segment, which is loaded and executed by a processor to implement the point of interest failure identification method as described in any one of claims 1 to 7.

17. A computer program product, characterized in that, The computer program product includes at least one instruction or at least one program segment, which is loaded and executed by a processor to implement the point of interest failure identification method as described in any one of claims 1 to 7.

Images