A method, apparatus, electronic device, and storage medium for adding points of interest.

Abstract

This invention discloses a method, apparatus, electronic device, and storage medium for adding points of interest (POIs). The method includes: determining the location of the POI based on an image captured by at least one imaging device; determining the target object trajectory type corresponding to the POI based on an image of the POI obtained by capturing the POI at its location; determining the name of the POI based on the target object trajectory type; and adding the POI to a map based on its location and name. This invention enables automatic and dynamic addition of POIs, improving the adaptability of POI data addition to flexibly changing scenarios.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a method, apparatus, electronic device, and storage medium for adding points of interest. Background Technology

[0002] Points of Interest (POIs) are crucial data for location services. A POI can be a building, a shop, a mailbox, or a bus stop, among other things. POI data can include the POI name, address information, location coordinates, and category, allowing users to quickly and accurately find the location information they need.

[0003] In current technologies, adding Points of Interest (POIs) to electronic maps typically relies on manual intervention. When temporary events or emergencies occur, it's impossible to promptly locate these events or incidents using the electronic map. Therefore, this method of manually adding POI data is poorly adaptable to dynamically changing scenarios. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and storage medium for adding points of interest (POIs), enabling automatic and dynamic addition of POIs and improving the adaptability of POI data addition to flexibly changing scenarios.

[0005] In a first aspect, embodiments of the present invention provide a method for adding points of interest, the method comprising:

[0006] Determine the location of the point of interest based on images captured by at least one imaging device;

[0007] Based on the image of the point of interest obtained by taking pictures of the location of the point of interest, the trajectory type of the target object corresponding to the point of interest is determined;

[0008] The name of the point of interest is determined based on the trajectory type of the target object corresponding to the point of interest;

[0009] Add points of interest to the map based on their location and name.

[0010] Secondly, embodiments of the present invention also provide a device for adding points of interest, the device comprising:

[0011] The point of interest location determination module is used to determine the location of points of interest based on images captured by at least one imaging device;

[0012] The target object trajectory type determination module is used to determine the target object trajectory type corresponding to the point of interest based on the point of interest image obtained by taking pictures of the location of the point of interest;

[0013] The point of interest name determination module is used to determine the name of the point of interest based on the target object trajectory type corresponding to the point of interest;

[0014] The Point of Interest (POI) adding module is used to add POIs to the map based on their location and name.

[0015] Thirdly, embodiments of the present invention also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for adding points of interest as described in any of the embodiments of the present invention.

[0016] Fourthly, embodiments of the present invention also provide a storage medium for storing computer-executable instructions, which, when executed by a computer processor, are used to perform the point-of-interest addition method as described in any of the embodiments of the present invention.

[0017] The technical solution of this invention intelligently analyzes images captured by a camera to obtain the location of points of interest (POIs). It then identifies the trajectory of the target object in the image captured at the POI location to determine the trajectory type corresponding to the POI, and determines the name of the POI based on the trajectory type. Finally, based on the location and name of the POI, it adds it to the map. This solves the problem of poor adaptability to flexible and changing scenes in existing manual POI addition methods, achieving automatic and dynamic addition of POIs and improving the adaptability of POI data addition to flexible and changing scenes.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of a method for adding points of interest provided in Embodiment 1 of the present invention;

[0021] Figure 2 This is a schematic diagram of the mapping relationship between a pixel coordinate system and a spatial coordinate system provided in Embodiment 1 of the present invention;

[0022] Figure 3 This is a schematic diagram of a candidate target object trajectory type provided in Embodiment 1 of the present invention;

[0023] Figure 4 This is a flowchart of a method for adding points of interest provided in Embodiment 2 of the present invention;

[0024] Figure 5 This is a schematic diagram of the intersection of the focus directions of a target object provided in Embodiment 2 of the present invention;

[0025] Figure 6 This is a schematic diagram of the structure of a point-of-interest (POI) adding device provided in Embodiment 3 of the present invention;

[0026] Figure 7 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices. In the embodiments of this application, certain software, components, models, and other existing industry solutions may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solutions of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0029] The acquisition, transmission, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.

[0030] Example 1

[0031] Figure 1 The flowchart of a method for adding points of interest (POIs) is provided in Embodiment 1 of the present invention. This embodiment is applicable to the situation of automatically and dynamically adding POIs to a map. The method can be executed by a POI adding device, which can be implemented in hardware and / or software. The POI adding device can be configured in a server and used in conjunction with a shooting device.

[0032] like Figure 1 As shown, the method includes:

[0033] S110. Determine the location of the point of interest based on the image captured by at least one imaging device.

[0034] This process can involve dividing the map into regions and adding all cameras in a specific region to a camera set, or defining a detection area and adding all cameras within that area to the camera set. Points of interest (POIs) are then detected based on images captured by each camera in the camera set.

[0035] The location of a point of interest refers to its spatial location, which can be represented by spatial coordinates or latitude and longitude. This embodiment does not impose any restrictions on this.

[0036] In an optional embodiment, the location of a point of interest (POI) is determined based on images captured by at least one imaging device. This can be achieved through image processing of the captured images, intelligently analyzing target objects within the imaging device's field of view (the area that the imaging device can capture). The target object can be a person, a vehicle, or other similar object. When a target object is determined to exist within the imaging device's field of view, its behavior is analyzed, including but not limited to whether the target object is in motion or stationary, whether its movement trajectory shares commonalities, whether multiple target objects cluster together in a dispersed or concentrated manner, and its orientation. By analyzing the behavior of multiple target objects over a given time period, the presence and location of the POI within the imaging device's field of view are determined.

[0037] For example, when the target object is a person, the number of target objects exceeds a preset threshold, and the target objects are clustered together in a convergent manner, a point of interest (POI) can be considered to exist. The spatial coordinates of the center point of the clustered area in the image are then transformed to obtain the spatial location of the POI. Similarly, when the target object is a vehicle, the number of target objects exceeds a preset threshold, and the movement trajectory of the target objects is a detour, a POI can be considered to exist. The center points of the areas detoured by multiple movement trajectories in the image are then transformed from pixel coordinates to spatial coordinates to obtain the spatial location of the POI.

[0038] In another optional embodiment, the location of the point of interest (POI) is determined based on images captured by at least one imaging device. This can be achieved by detecting target objects in the images captured by the imaging device; when the target object is a person, the gaze direction of the target object is detected; when the target object is a vehicle, the orientation of the target object is detected. Intelligent analysis is performed on the gaze direction or orientation of the target object corresponding to at least one imaging device to determine whether there are commonalities or intersections in the gaze direction or orientation of the target object. By analyzing the gaze directions or orientations of multiple target objects within a given time period, it is determined whether a POI exists within the field of view of the imaging device, and the location of the POI is determined.

[0039] For example, when the target object is a person, the positions and viewing directions of each target object captured by each shooting device within a preset time period are statistically analyzed in the images. The positions and viewing directions of each target object are then converted to a spatial coordinate system and marked on a map. Based on the positions and viewing directions of each target object on the map, it is determined whether the viewing directions of each target object intersect in space. If they intersect, it is further determined whether the intersection points of the viewing directions are concentrated within a preset range. If so, a point of interest (POI) is identified, and the centroid of each intersection point is taken as the location of the POI. When the target object is a vehicle, similarly, the orientations of each target object captured by each shooting device within a preset time period are statistically analyzed in the images. The orientations of each target object are then converted to a spatial coordinate system, and the positions and orientations of each target object are marked on a map. If the orientations of each target object are relatively uniform, and each target object is stationary, a POI is identified, and the center point of each target object's position is taken as the location of the POI.

[0040] It should be noted that in this embodiment, subsequent operations are only performed on newly added points of interest (POIs). After obtaining the location of the POI through intelligent analysis of the image captured by the camera, it is necessary to compare the POI with existing POIs on the map. If it is determined to be a new POI, the subsequent steps are performed to add the POI. If it is determined that the distance between the location of the POI and the location of an existing POI on the map is less than or equal to a distance threshold, then the POI is determined to be an existing POI, and no further operations are performed on it.

[0041] In this embodiment, the location of the point of interest is obtained by intelligently analyzing the images captured by the imaging device. In the event of temporary activities or sudden accidents, the specific location of the point of interest is determined, thereby achieving rapid positioning of temporarily added points of interest.

[0042] S120. Based on the image of the point of interest obtained by taking pictures of the location of the point of interest, determine the trajectory type of the target object corresponding to the point of interest.

[0043] Specifically, an image of the point of interest can be obtained by capturing the location of the point of interest using a camera whose field of view can cover it. Furthermore, if no camera has a field of view that covers the point of interest, it can be determined whether a gimbal camera is nearby, and the gimbal camera can be rotated until its field of view covers the point of interest.

[0044] The target object can be a person or a vehicle, etc., and this embodiment does not limit the specific type of the target object. The target object trajectory type is used to reflect the overall pattern of the trajectory of each target object at the point of interest location. In this embodiment, trajectory recognition and trajectory trend analysis are performed on each target object in the image of the point of interest within a preset time period. Based on the pattern of the trajectory of the target object that exceeds a certain proportion threshold, the target object trajectory type corresponding to the point of interest is determined.

[0045] Understandably, different trajectory types can be set for different types of target objects, and the types of points of interest obtained from trajectory analysis may also differ. For example, if the target object is a person, the target object's trajectory might be a stationary convergence type, an entrance movement type, an exit movement type, etc., and correspondingly, the types of points of interest determined by trajectory analysis might be temporary activities, stage performances, or safety incidents, etc. If the target object is a vehicle, the target object's trajectory might be a moving detour type, a stationary convergence type, etc., and correspondingly, the types of points of interest determined by trajectory analysis might be temporary road repairs, safety incidents, or vehicle exhibitions, etc.

[0046] In this embodiment, based on the point of interest image obtained by capturing the location of the point of interest, the target object trajectory type corresponding to the point of interest is determined. Specifically, the target object trajectory of each target object in the point of interest image within a preset time period is determined, and the target object trajectory in the obtained point of interest image is converted to a spatial coordinate system to obtain the target object trajectory in the spatial coordinate system.

[0047] The transformation relationship between the pixel coordinate system and the spatial coordinate system can be determined by the lens parameters, installation height, and installation angle of the shooting device. Figure 2 A schematic diagram illustrating the mapping relationship between pixel coordinates and spatial coordinates is provided, such as... Figure 2 As shown, a and b are the lens parameters of the CAM imaging device. Vectors in images captured by CAM. for The three-dimensional spatial vectors within the CAM's visible field of view ABCD can be used to determine the transformation relationship between the pixel coordinate system and the spatial coordinate system based on the CAM's installation height, installation angle, and a and b. The transformation of the target object's viewing direction / orientation in the image to the spatial coordinate system mentioned in S110, and the transformation of the target object's trajectory in the image to the spatial coordinate system here, can both be achieved through this transformation relationship between the pixel coordinate system and the spatial coordinate system.

[0048] After obtaining the trajectory of the target object in the spatial coordinate system, the trajectory of each target object is marked on the map, and the trajectory of each target is compared with the pre-set candidate target object trajectory types. The target object trajectory type that matches the point of interest is selected from the candidate target object trajectory types.

[0049] For example, Figure 3 A schematic diagram of the trajectory type of a candidate target object is provided, such as... Figure 3 As shown, the trajectory types of candidate target objects are divided into convergence type, entrance type, exit type, wandering type, stationary type, and stage type. Figure 3 This embodiment is merely an illustrative example of candidate target object trajectory types and does not limit the number or specific form of candidate target object trajectory types.

[0050] Specifically, if the ratio of the number of target object trajectories meeting the stationary condition within a preset time period to the total number of target object trajectories is greater than or equal to a ratio threshold, then the target object trajectory type is determined to be stationary. Meeting the stationary condition means that the target object's position remains unchanged within the preset time period, or its position change is less than or equal to a preset distance.

[0051] In an optional embodiment, the similarity between the trajectory of each target object in the map and the trajectory diagram corresponding to the candidate target object trajectory type can be calculated. The similarity is then sorted from high to low, and the target object trajectory type matching the point of interest is selected from the candidate target object trajectory types. Specifically, the candidate target object trajectory type with the highest similarity can be selected; alternatively, the candidate target object trajectory type with the highest similarity and a similarity greater than or equal to a similarity threshold can be selected. If there are multiple candidate target object trajectory types with a similarity greater than or equal to the similarity threshold, a preset number or a preset proportion of the candidate target object trajectory types with the highest similarity can also be selected. This embodiment does not impose any limitations on this.

[0052] In another optional embodiment, for each target object's trajectory in the map, the number of target object trajectories matching the trajectory diagram corresponding to the candidate target object trajectory type can be counted, and the ratio of the number of matching target object trajectories to the total number of target object trajectories can be calculated. If the ratio is determined to be greater than or equal to a preset ratio threshold, then the candidate target object trajectory type is determined to be a target object trajectory type matching the point of interest.

[0053] In this embodiment, by capturing images of points of interest at their locations and performing intelligent analysis on these images, the trajectory type of the target object at the point of interest is determined, thereby identifying the descriptive data of the point of interest.

[0054] S130. Determine the name of the point of interest based on the target object trajectory type corresponding to the point of interest.

[0055] The name of a point of interest can be a general label, such as "emergency" or a more specific name, such as "performance at booth xx in shopping mall A". This embodiment does not restrict the specific form of the name of the point of interest.

[0056] In an optional embodiment, the name of the point of interest (POI) is determined based on the trajectory type of the target object corresponding to the POI. This can be achieved by combining the target object trajectory type with road network data after determining the POI's trajectory type. Specifically, based on road network data, existing POIs that are closest to the POI's location, or within a certain distance range, are identified. The name of the POI is then determined based on the description data of the existing POIs and the target object's trajectory type. For example, if the closest existing POI to the POI's location is "xx Square," and the target object is a person with a trajectory type of "entrance," the POI's name can be determined as "Event Entrance." If the closest existing POI to the POI's location is "xx Shopping Mall xx Gate," and the target object is a person with a trajectory type of "convergence," the POI's name can be determined as "xx Shopping Mall xx Gate Booth Performance."

[0057] In another optional embodiment, the name of the point of interest (POI) is determined based on the trajectory type of the target object corresponding to the POI. Alternatively, after determining the trajectory type of the target object, image processing, such as text extraction or face recognition, is performed on the POI image. The trajectory type of the target object is then combined with the image processing results to determine the name of the POI. For example, taking text extraction as an example, text is extracted from the POI image, and semantic recognition is performed on the extracted text. If it is determined that the POI image contains advertising slogans and the trajectory type of the target object is static, then the name of the POI can be determined as an advertising identifier. Taking face recognition as an example, the recognized face is compared with a pre-set face database. If it is determined that the recognized face tag is a celebrity or star and the trajectory type of the target object is static, then the name of the POI can be determined as an advertising identifier.

[0058] In this embodiment, by performing intelligent analysis of the target object trajectory at the location of the point of interest, the scene classification at the location of the point of interest can be determined, thereby determining the name of the point of interest.

[0059] Furthermore, S130 may include:

[0060] A1. Determine the interest point label that matches the interest point based on the target object trajectory type corresponding to the interest point;

[0061] A2. Determine the name of the point of interest based on the point of interest label and at least one piece of scene information of the point of interest.

[0062] Points of interest (POIs) are used to indicate the scene type corresponding to a POI. For example, POIs may include event entrance, event exit, booth performance, emergency, etc. The correspondence between different target object trajectory types and POIs can be pre-set. For example, when the target object trajectory type is an entrance, the corresponding POI can be an event entrance; when the target object trajectory type is an exit, the corresponding POI can be an event exit; when the target object trajectory type is a convergence, the corresponding POI can be an emergency, exhibition, etc. This embodiment does not limit the correspondence between target object trajectory types and POIs.

[0063] Scene information may include the results of text extraction or face recognition of images of points of interest, information related to the location of points of interest extracted from big data, descriptive data of existing points of interest within a certain distance from the location of the point of interest, and field information obtained by voice extraction within a certain distance from the location of the point of interest. This embodiment does not limit the specific type of scene information.

[0064] In an optional embodiment, A1 may include: determining the point of interest label corresponding to the point of interest based on the target object trajectory type and a pre-set correspondence between target object trajectory types and point of interest labels. When there are multiple point of interest labels corresponding to the target object trajectory type in the correspondence between target object trajectory types and point of interest labels, the point of interest label corresponding to the point of interest can be determined by combining the point of interest location, scene information, etc.

[0065] For example, when the target object's trajectory type is convergence, the corresponding point-of-interest (POI) tags include three types: emergency, exhibition, and performance. The POI tag matching the POI can be determined by combining the POI's location. For instance, if the POI's location is at an intersection, the matching POI tag is "emergency." If the POI's location is at a shopping mall, the matching POI tag can be further determined by combining scene information. For example, if text extraction is performed on the POI image and the field "pet adoption" is identified, the POI tag can be determined as "exhibition," and can be further refined to "pet adoption exhibition."

[0066] In another optional embodiment, A1 may further include: selecting an interest point label that matches the interest point from at least two pre-set candidate interest point labels, based on the target object trajectory type corresponding to the interest point, the location of the interest point, and at least one target object feature corresponding to the interest point.

[0067] This embodiment uses the example of determining the point of interest label that matches the point of interest based on the target object's trajectory type, the location of the point of interest, and the characteristics of the target object.

[0068] When the target object is a person, its characteristics can include facial expression, age group, and clothing type; when the target object is a vehicle, its characteristics can include speed and brand. Understandably, target object characteristics can take different forms under different point-of-interest (POI) tags, while under the same POI tag, different target objects may exhibit similar characteristics. For example, for a booth performance POI tag, it could correspond to facial expression or clothing characteristics; the target object might generally display a happy expression, or their clothing might be a performance dress. For a vehicle exhibition POI tag, it could correspond to brand characteristics; the target objects might be distributed across different brands by area, with vehicles within the same area corresponding to the same brand, or even all vehicles within the entire POI location area corresponding to the same brand.

[0069] For different candidate point of interest (POI) tags, matching target object trajectory types, POI locations, and target object features can be set, and their corresponding feature values ​​can be determined. The same candidate POI tag can be assigned different corresponding target object trajectory types, POI locations, and target object features as needed; this embodiment does not impose any restrictions on this. For example, Table 1 provides a table of candidate POI tag feature values, taking the target object trajectory type, POI location, target object expression, and feature values ​​corresponding to different candidate POI tags as examples. Table 1 is shown below:

[0070] Table 1

[0071]

[0072]

[0073] In this embodiment, after determining the location of the point of interest (POI) and the trajectory type of the target object corresponding to the POI, face recognition is performed on the POI image to determine the expression of the target object. Based on the current POI location, the trajectory type of the target object corresponding to the POI, and the target object's expression, the current feature value is calculated. Feature comparison is performed between the current feature value and the feature values ​​corresponding to each candidate POI label. Based on the feature comparison results, the POI label matching the POI is determined from among the candidate POI labels.

[0074] Specifically, candidate interest point labels with feature comparison similarity greater than or equal to a preset similarity threshold can all be used as interest point labels matching the interest point. Alternatively, the feature comparison similarity can be sorted, and a preset number of candidate interest point labels with the highest similarity can be selected. Then, combined with scene information, interest point labels matching the scene information are further determined as the final interest point labels. Alternatively, candidate interest point labels with feature comparison similarity greater than or equal to a preset similarity threshold and the highest similarity can be used as interest point labels matching the interest point. This embodiment does not restrict the specific method for determining the interest point labels matching the interest point from among the candidate interest point labels based on the feature comparison results.

[0075] It should be noted that if the feature similarity of each candidate point of interest (POI) label is lower than the preset similarity threshold, a prompt to add POI labels can be made through the user interface. If the user interface receives a confirmation instruction to add POI labels, the corresponding candidate POI labels will be used as the POI labels that match the POI. Simultaneously, feature learning can be performed on the candidate POI labels based on the current target object trajectory type, POI location, and target object characteristics to improve the accuracy of subsequent POI label determination.

[0076] Furthermore, A2 can include:

[0077] A21. Determine scene information, wherein the scene information includes at least one of the following: text information obtained by character recognition and extraction of the point of interest image, text information obtained by recognizing the voice data that matches the current point of interest location, description data of existing points of interest that meet the preset distance range condition with the current point of interest location, and association fields that match the current point of interest location.

[0078] A22. If the number of scene information items is determined to be one item, and the scene information matches the point of interest tag, then the name of the point of interest is determined based on the point of interest tag and the scene information;

[0079] A23. If it is determined that the number of scene information items is at least two, the scene information items match each other, and the scene information items match the point of interest tag, then the name of the point of interest is determined based on the point of interest tag and the scene information.

[0080] In this embodiment, scene information can be obtained in the following ways: performing text recognition and extraction on the point of interest image, and using the extracted text information as scene information; capturing voice data matching the current point of interest location from big data, and performing text recognition on the voice data, and using the recognized text information as scene information; identifying existing points of interest within a certain distance range from the current point of interest location on the map, and determining the description data of the existing points of interest as scene information; capturing the associated fields of the current point of interest location from big data, and filtering the associated fields whose frequency of occurrence within a preset time period reaches a preset threshold proportion of the total frequency of occurrence of all associated fields of the current point of interest location, and using these as associated fields matching the current point of interest location.

[0081] In one optional embodiment, if the scene information includes only one item, after extracting the scene information, it is determined whether the scene information matches the point of interest (POI) label. If they match, the name of the POI is determined by combining the scene information and the POI label. For example, if text recognition is performed on the POI image and the field "xx shopping mall" is extracted, and the POI label is "exhibition", then it can be determined that the scene information matches the POI label, and the POI name is added as "xx shopping mall exhibition".

[0082] In another optional embodiment, when the scene information includes multiple items, it is determined whether the various scene information items match. If the various scene information items match, it is further determined whether the scene information matches the point of interest tag. If they match, the name of the point of interest is determined by combining the various scene information items and the point of interest tag.

[0083] In a specific example, if text recognition and extraction of the image of a point of interest yields the field "xx shopping mall", and existing points of interest within a certain distance of the current point of interest location on the map include Shop A, whose address in its description data is "xx shopping mall, ground floor shop xx", and the extracted associated field is "xx organization pet adoption day", with the point of interest tag being "exhibition", then it can be determined that the three pieces of scene information match, and that the three pieces of scene information also match the point of interest tag. Based on the above three pieces of scene information and the point of interest tag, the point of interest name can be determined as "xx shopping mall xx organization pet adoption exhibition".

[0084] It should be noted that if scene information cannot be extracted, or if multiple pieces of scene information do not match, or if the scene information does not match the point of interest (POI) tag, the POI tag can be directly used as the POI name. Additionally, the POI location can be added to the POI tag to differentiate the POIs. For example, if the POI tag is "Sudden Accident" and the POI location is "Intersection of xx Road and xx Street," then the POI name can be determined as "Sudden Accident at the Intersection of xx Road and xx Street."

[0085] S140. Add points of interest to the map based on their location and name.

[0086] In this embodiment, after determining the location and name of the point of interest (POI), it can be added to the map. Furthermore, the server can record specific information about the target imaging device capturing the POI image and data such as the trajectory type of the target object at that POI, facilitating subsequent POI differentiation, anomaly handling, and POI failure detection.

[0087] Furthermore, this embodiment can continuously monitor the field of view of the target capturing device to detect the failure of the points of interest. Specifically, if it is determined that the number of target objects in the image captured by the target capturing device is less than or equal to a preset threshold, or the target objects no longer exhibit a clustered pattern, or the trajectory type of the target objects changes, then the point of interest is determined to be invalid. After determining that the point of interest is invalid, it is deleted from the map.

[0088] The technical solution of this invention intelligently analyzes images captured by a camera to obtain the location of points of interest (POIs). It then identifies the trajectory of the target object in the images captured at the POI locations to determine the trajectory type corresponding to the POI, and determines the name of the POI based on the trajectory type. Finally, based on the location and name of the POI, it adds it to the map. This solves the problem of poor adaptability to flexible and changing scenarios in existing manual POI addition methods. It enables the automatic and dynamic addition and deletion of temporarily occurring popular POIs, improving the adaptability of POI data addition to flexible and changing scenarios.

[0089] Example 2

[0090] Figure 2 This is a flowchart of a method for adding points of interest according to Embodiment 2 of the present invention. Based on the above embodiments, the present invention further specifies the process of determining the location of the point of interest, the process of determining the trajectory type of the target object of the point of interest, and the process of determining the name of the point of interest.

[0091] like Figure 2 As shown, the method includes:

[0092] S210. Detect target objects in images captured by at least one imaging device and determine the target object interest direction corresponding to the imaging device.

[0093] The "direction of focus" refers to the orientation of the target object as captured by the camera over a sustained period of time. Understandably, when a target object continues to face a certain direction for an extended period, it can be assumed that the target object is paying close attention to that direction, and a hotspot event may be occurring in that direction.

[0094] When the target object is a person, the direction of attention can be determined by the body's orientation, facial orientation, or gaze direction. Specifically, body orientation can be determined by performing human detection on the image to locate the body contour; extracting features from the body contour; calculating the body's tilt angle based on the extracted features to analyze the orientation of the body contour. Facial orientation can be determined by performing face detection on the image to locate the facial region; extracting feature points from the facial region, such as eyes, nose, and mouth; and inferring the facial orientation by calculating the relative positions of the feature points in the facial coordinate system. Gazing direction can be determined by performing eye region detection on the image to extract the eye region; extracting key points from the eye region, such as the iris and pupil; constructing an eye geometric model based on the eye region and its key points; locating the center position of the pupil; and determining the optical axis direction based on the eye geometric model and the pupil center position; and determining the gaze direction based on the optical axis direction. The detection of body orientation, facial orientation, and gaze direction can all be achieved by training machine learning or deep learning models with a large number of sample images to obtain the corresponding detection models. However, this embodiment does not limit the form of expression and specific determination process of the target object's focus direction when the target object is a person.

[0095] When the target object is a vehicle, the direction of interest can be determined by the vehicle's heading and pose. Specifically, the vehicle's heading can be determined by detecting the vehicle in the image, extracting the vehicle region, identifying front feature points (such as headlights and logos), and analyzing the heading based on these feature points. The vehicle's pose can be determined by detecting the vehicle in the image, extracting features (such as contours, edges, and wheel positions), and converting these features into 3D space through 3D reconstruction or projection transformation to determine the vehicle's position and orientation. Alternatively, the steering angle of the wheels can be inferred by analyzing their shape and relative position, thus determining the vehicle's pose. Similarly, the aforementioned heading and pose can be determined by training a machine learning or deep learning model with a large number of sample images to obtain the corresponding detection model. However, this embodiment does not limit the representation or specific determination process of the target object's direction of interest when the target object is a vehicle.

[0096] It should be noted that the target object focus direction in this embodiment refers to the focus direction of the target object in three-dimensional space. After obtaining the focus direction of the target object in the image through image detection, the focus direction is transformed into the spatial coordinate system through the transformation relationship between the pixel coordinate system and the spatial coordinate system in the above embodiment.

[0097] In this embodiment, the direction of attention of the target object is determined by the image captured by the imaging device, thereby determining whether a hotspot event has occurred. If a hotspot event is determined to have occurred, it is considered that there may be a point of interest.

[0098] Furthermore, S210 may include:

[0099] B1. Detect target objects in the images captured by the target imaging device, obtain the target objects, and determine the orientation of the target objects;

[0100] B2. If the duration for which the orientation of the target object satisfies the orientation consistency condition is greater than or equal to a preset time threshold, then the focus direction of the target object is determined based on the orientation of the target object.

[0101] B3. If the ratio of the number of target objects whose direction of interest meets the direction consistency condition to the total number of target objects in the image captured by the target imaging device is greater than or equal to a preset first ratio threshold, then the target object direction of interest corresponding to the target imaging device is determined according to each direction of interest that meets the direction consistency condition.

[0102] Here, "target imaging device" refers to the single imaging device currently being processed. For each target object detected by this device, the direction of interest is determined individually. The specific manifestations of the target object's orientation under different target objects, as well as the specific process for determining the target object's orientation, have been explained above and will not be repeated here.

[0103] In an optional embodiment, the target object orientation in B2 satisfies the direction consistency condition, meaning the angular difference between target object orientations is less than or equal to a preset angular threshold. When the target object orientation is represented by a vector, it means the vector distance between target object orientations is less than a preset vector distance threshold. Taking a person as an example, it can be understood that for a target object captured by a target imaging device, if it consistently looks in the same direction for a period of time, it can be considered that it is paying more attention to that direction. In this case, the average value of the target object orientation over the duration can be directly calculated as the target object's direction of attention.

[0104] In another optional embodiment, the target object orientation in B2 satisfies the direction consistency condition, which can also mean that the intersection of the target object orientations is within a preset distance range. Taking a person as an example, it can be understood that for a target object captured by a target imaging device, when it is in motion within the target imaging device's field of view, if the target object is moving while continuously looking at a certain hotspot event for a period of time, then the target object orientation during that continuous time period should converge within the area where the hotspot event occurred. In this case, the average value of the target object orientation over the duration can also be calculated as the target object's direction of focus; alternatively, the target object orientation corresponding to the midpoint of the duration can be directly used as the target object's direction of focus. This embodiment does not impose any restrictions on this.

[0105] It should be noted that when determining the target object's direction of interest, the target object's position must also be recorded. Specifically, when calculating the target object's direction of interest by averaging its orientation over a duration, the average position of the target object over the duration should be recorded along with the direction of interest. When using the target object's orientation at the midpoint of the duration as the direction of interest, the position of the target object at that midpoint should be recorded along with the direction of interest. Similarly, the target object's position refers to its location in the spatial coordinate system.

[0106] In this embodiment, the target object focus direction corresponding to the target shooting device is determined based on the trend or pattern of the target object focus direction captured by the target shooting device within a certain time period.

[0107] In an optional embodiment, the focus direction in B3 satisfies the direction consistency condition, which can mean that the angle difference between the focus directions of the target objects is less than or equal to a preset angle threshold. When the focus direction of the target objects is represented by a vector, it means that the vector distance between the focus directions of the target objects is less than a preset vector distance threshold. Taking a person as an example, it can be understood that for a target imaging device, if, among all target objects captured within a certain time period, more than a first proportion threshold are all looking in the same direction, then a hotspot event has occurred in that direction, and there may be a point of interest in that direction. In this case, the average value of the focus directions of each target object that satisfy the direction consistency condition can be calculated as the focus direction of the target object corresponding to the target imaging device.

[0108] In another optional embodiment, the focus direction in B3 satisfying the direction consistency condition can also refer to the intersection of the focus directions of the target objects being within a preset distance range. Taking a person as the target object as an example, it is understood that there may be situations where the target object is in motion within the field of view of the target shooting device, or different target objects are scattered throughout the field of view of the target shooting device. In this case, for the target shooting device, if, within a certain time period captured by the device, the focus directions of target objects exceeding a first proportion threshold all converge within a small area, it can be considered that a hotspot event has occurred within that area, and points of interest may exist within that area. In this case, the average value of the focus directions of each target object satisfying the direction consistency condition can be calculated as the focus direction of the target object corresponding to the target shooting device; alternatively, the focus direction of the target object located at the center position can be used as the focus direction of the target object corresponding to the target shooting device; or all focus directions of target objects satisfying the direction consistency condition can be used as the focus direction of the target object corresponding to the target shooting device. This embodiment does not impose any restrictions on this.

[0109] S220. Determine the location of the point of interest based on the direction of interest of the target object corresponding to at least one shooting device.

[0110] In this embodiment, the location of a point of interest (POI) is determined by marking the direction of interest and the location of the target object corresponding to at least one imaging device on a map. It should be noted that after determining the location of the POI, it is necessary to compare it with the locations of existing POIs on the map. Only after confirming that the POI is a newly added POI can subsequent operations be performed.

[0111] In this embodiment, when more than a certain proportion of target objects are interested in a certain direction of interest, the ray of interest is recorded by the position of the target object and the direction of interest. When multiple shooting devices are present, the intersection of the rays of interest is determined. When the ray intersection points to the same area, the location of the point of interest is considered to be in that area.

[0112] Furthermore, S220 may include:

[0113] C1. If the number of shooting devices is determined to be one, the location of the point of interest is determined based on the image obtained by shooting the target object of the shooting device in the direction of interest.

[0114] C2. If the number of shooting devices is determined to be at least two, then determine at least one intersection point of the target object's attention direction corresponding to each shooting device, and determine the position of the point of interest based on the intersection point.

[0115] Specifically, when only one camera can be identified as having the target object's direction of interest, the camera's field of view is determined to cover that direction of interest, or the camera's pan / tilt head is rotated until the field of view covers that direction of interest. The image captured by this camera, along with the target object's direction of interest, is then used for image recognition to determine the location of points of interest. For example, target object clustering region identification can be performed on the image, and the center point of the clustered region can be used as the location of the point of interest.

[0116] When multiple shooting devices can be identified to determine the direction of interest of the target object, the intersection point of the direction of interest of the target object corresponding to each shooting device is determined. If there is only one intersection point, the position of the intersection point is taken as the position of the point of interest. If there are at least two intersection points, it is determined whether the intersection points are concentrated in a small area. For example, it can be determined whether the maximum distance between the intersection points is less than or equal to a distance threshold. A minimum outer circle can be determined for each intersection point, and it can be determined whether the radius of the minimum outer circle is less than or equal to a preset radius threshold. This embodiment does not impose any restrictions on this. If it is determined that the intersection points are concentrated in a small area, the centroid of each intersection point can be taken as the position of the point of interest. If it is determined that the intersection points are not concentrated in a small area, that is, the intersection points are relatively scattered, the images obtained by shooting each intersection point can be determined separately, and the images can be used to help determine whether there is a point of interest in the distribution area of ​​each intersection point, and the position of the point of interest can be determined.

[0117] Figure 5 A schematic diagram illustrating the intersection of the attention directions of target objects is provided, such as... Figure 5 As shown, the ray of interest for the target object corresponding to imaging device B can be obtained through... To represent this, the ray of interest for the target object corresponding to the imaging device D can be obtained through... To express. and If the points converge at point E, then the location of point E is the location of the point of interest.

[0118] S230. Based on the image of the point of interest obtained by taking pictures of the location of the point of interest, determine the trajectory type of the target object corresponding to the point of interest.

[0119] S240. Determine the interest point label that matches the interest point based on the target object trajectory type corresponding to the interest point.

[0120] S250. Determine the name of the point of interest based on the point of interest label and at least one piece of scene information of the point of interest.

[0121] The processes of determining the target object trajectory type corresponding to the point of interest, determining the point of interest label matching the point of interest based on the target object trajectory type, and determining the point of interest name by combining the point of interest label and scene information have been described in the above embodiments and will not be repeated here.

[0122] S260. Add points of interest to the map based on their location and name.

[0123] Furthermore, after S260, the method further includes: if the ratio of the number of target objects whose direction of interest meets the direction consistency condition to the total number of target objects in the image captured by the target imaging device is less than or equal to a preset second ratio threshold, or if the trajectory type of the target object corresponding to the point of interest changes, then the point of interest is deleted in the map.

[0124] The second proportional threshold can be the same as or less than the first proportional threshold; this embodiment does not impose any restrictions on this.

[0125] In this embodiment, when the proportion of the target object focusing on the direction of the point of interest decreases to a certain threshold, the point of interest can be considered invalid. Alternatively, if the trajectory type of the target object corresponding to the point of interest changes based on the point of interest image, the point of interest can be considered invalid. When a point of interest becomes invalid, it is deleted from the map.

[0126] The technical solution of this embodiment detects the orientation of a target object. When the orientation of the target object remains consistent for a period of time, the direction of interest for the target object is determined. When more than a certain proportion of target objects show interest in this direction of interest, the ray of interest is recorded, indicating a hotspot event on that ray. Combining the ray of interest from multiple shooting devices, the intersection of the rays is determined. When the intersection points point to the same area, a point of interest (POI) is identified, and its location is determined. By capturing images of the POI locations, the trajectory type of the target object is identified. The POI is then tagged and classified based on the trajectory type, POI location, and target object features. Finally, the POI name is determined by combining the scene information corresponding to the POI location. This enables the automatic and rapid addition of POIs when temporary hotspot events occur. Simultaneously, the status of POIs is continuously monitored, and POIs are automatically deleted after the temporary hotspot event ends, achieving automatic addition and deletion of POIs.

[0127] Example 3

[0128] Figure 6 This is a schematic diagram of a point-of-interest (POI) adding device provided in Embodiment 3 of the present invention.

[0129] like Figure 6 As shown, the device includes:

[0130] The point of interest location determination module 310 is used to determine the location of the point of interest based on an image captured by at least one imaging device;

[0131] The target object trajectory type determination module 320 is used to determine the target object trajectory type corresponding to the point of interest based on the point of interest image obtained by taking pictures of the location of the point of interest;

[0132] The point of interest name determination module 330 is used to determine the name of the point of interest based on the target object trajectory type corresponding to the point of interest;

[0133] The point of interest (POI) adding module 340 is used to add POIs to the map based on their location and name.

[0134] The technical solution of this invention intelligently analyzes images captured by a camera to obtain the location of points of interest (POIs). It then identifies the trajectory of the target object in the image captured at the POI location to determine the trajectory type corresponding to the POI, and determines the name of the POI based on the trajectory type. Finally, based on the location and name of the POI, it adds it to the map. This solves the problem of poor adaptability to flexible and changing scenes in existing manual POI addition methods, achieving automatic and dynamic addition of POIs and improving the adaptability of POI data addition to flexible and changing scenes.

[0135] Optionally, the point of interest location determination module 310 includes:

[0136] The target object attention direction determination unit is used to detect target objects in images captured by at least one imaging device and determine the target object attention direction corresponding to the imaging device.

[0137] The point of interest location determination unit is used to determine the location of the point of interest based on the focus direction of the target object corresponding to at least one shooting device.

[0138] Optionally, the target object focus direction determination unit is specifically used for:

[0139] Target object detection is performed on the images captured by the target imaging device to obtain the target object and determine its orientation;

[0140] If the duration for which the orientation of the target object satisfies the orientation consistency condition is greater than or equal to a preset time threshold, then the focus direction of the target object is determined based on the orientation of the target object.

[0141] If the ratio of the number of target objects whose direction of interest meets the direction consistency condition to the total number of target objects in the image captured by the target imaging device is greater than or equal to a preset first ratio threshold, then the target object direction of interest corresponding to the target imaging device is determined according to each direction of interest that meets the direction consistency condition.

[0142] Optionally, the point of interest name determination module 330 includes:

[0143] The point of interest label determination unit is used to determine the point of interest label that matches the point of interest based on the target object trajectory type corresponding to the point of interest.

[0144] The point of interest name determination unit is used to determine the name of the point of interest based on the point of interest label and at least one piece of scene information of the point of interest.

[0145] Optional, the point of interest label determination unit is specifically used for:

[0146] Based on the target object trajectory type corresponding to the point of interest, the location of the point of interest, and at least one target object feature corresponding to the point of interest, select the point of interest label that matches the point of interest from at least two pre-set candidate point of interest labels.

[0147] Optional, the point of interest name determination unit, specifically used for:

[0148] Determine scene information, which includes at least one of the following: text information obtained by character recognition and extraction of the image of the point of interest, text information obtained by recognizing the voice data that matches the current location of the point of interest, description data of existing points of interest that meet the preset distance range condition of the current location of the point of interest, and association fields that match the current location of the point of interest.

[0149] If the number of scene information items is determined to be one, and the scene information matches the point of interest tag, then the name of the point of interest is determined based on the point of interest tag and the scene information;

[0150] If it is determined that the number of scene information items is at least two, the scene information items match each other, and the scene information items match the point of interest tag, then the name of the point of interest is determined based on the point of interest tag and the scene information.

[0151] Optionally, the device further includes:

[0152] The point of interest deletion module is used to delete points of interest in the map if the ratio of the number of target objects that meet the direction consistency condition in the direction of interest to the total number of target objects in the image captured by the target imaging device is less than or equal to a preset second ratio threshold, or if the trajectory type of the target object corresponding to the point of interest changes.

[0153] The point-of-interest (POI) adding device provided in this embodiment of the invention can execute the point-of-interest adding method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0154] Example 4

[0155] Figure 7 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0156] like Figure 7 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0157] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0158] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method for adding points of interest.

[0159] In some embodiments, the method for adding points of interest may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for adding points of interest described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the method for adding points of interest by any other suitable means (e.g., by means of firmware).

[0160] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0161] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0162] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0163] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device 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 pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; 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 sound input, voice input, or tactile input).

[0164] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0165] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0166] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0167] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. 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 substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for adding points of interest, characterized in that, include: Determine the location of the point of interest based on images captured by at least one imaging device; Based on the image of the point of interest obtained by taking pictures of the location of the point of interest, the trajectory type of the target object corresponding to the point of interest is determined; The name of the point of interest is determined based on the target object trajectory type corresponding to the point of interest, including: Based on the target object trajectory type corresponding to the point of interest, determine the point of interest tag that matches the point of interest; wherein, the point of interest tag is used to represent the scene type corresponding to the point of interest. The correspondence between different target object trajectory types and point of interest tags is pre-set. When there are multiple point of interest tags corresponding to the target object trajectory type, the point of interest tag corresponding to the point of interest is determined by combining the point of interest location and scene information. The name of the point of interest is determined based on the point of interest label and at least one piece of scene information of the point of interest; Add points of interest to the map based on their location and name.

2. The method according to claim 1, characterized in that, Determine the location of points of interest based on images captured by at least one imaging device, including: Target object detection is performed on images captured by at least one imaging device, and the focus direction of the target object corresponding to the imaging device is determined; The location of the point of interest is determined based on the direction of attention of the target object corresponding to at least one shooting device.

3. The method according to claim 2, characterized in that, Target object detection is performed on images captured by at least one imaging device, and the focus direction of the target object corresponding to the imaging device is determined, including: Target object detection is performed on the images captured by the target imaging device to obtain the target object and determine its orientation; If the duration for which the orientation of the target object satisfies the orientation consistency condition is greater than or equal to a preset time threshold, then the focus direction of the target object is determined based on the orientation of the target object. If the ratio of the number of target objects whose direction of interest meets the direction consistency condition to the total number of target objects in the image captured by the target imaging device is greater than or equal to a preset first ratio threshold, then the target object direction of interest corresponding to the target imaging device is determined according to each direction of interest that meets the direction consistency condition.

4. The method according to claim 1, characterized in that, Based on the target object trajectory type corresponding to the point of interest, determine the point of interest label that matches the point of interest, including: Based on the target object trajectory type corresponding to the point of interest, the location of the point of interest, and at least one target object feature corresponding to the point of interest, select the point of interest label that matches the point of interest from at least two pre-set candidate point of interest labels.

5. The method according to claim 1, characterized in that, The name of the point of interest is determined based on the point of interest tag and at least one piece of scene information of the point of interest, including: Determine scene information, which includes at least one of the following: text information obtained by character recognition and extraction of the image of the point of interest, text information obtained by recognizing the voice data that matches the current location of the point of interest, description data of existing points of interest that meet the preset distance range condition of the current location of the point of interest, and association fields that match the current location of the point of interest. If the number of scene information items is determined to be one, and the scene information matches the point of interest tag, then the name of the point of interest is determined based on the point of interest tag and the scene information; If it is determined that the number of scene information items is at least two, the scene information items match each other, and the scene information items match the point of interest tag, then the name of the point of interest is determined based on the point of interest tag and the scene information.

6. The method according to claim 3, characterized in that, After adding points of interest to the map, the following is also included: If the ratio of the number of target objects that meet the direction consistency condition in the direction of interest to the total number of target objects in the image captured by the target imaging device is less than or equal to a preset second ratio threshold, or if the trajectory type of the target object corresponding to the point of interest changes, then the point of interest is deleted in the map.

7. A device for adding points of interest, characterized in that, include: The point of interest location determination module is used to determine the location of points of interest based on images captured by at least one imaging device; The target object trajectory type determination module is used to determine the target object trajectory type corresponding to the point of interest based on the point of interest image obtained by taking pictures of the location of the point of interest; The point of interest name determination module is used to determine the name of the point of interest based on the target object trajectory type corresponding to the point of interest; The Point of Interest (POI) name determination module includes: The point of interest (POI) tag determination unit is used to determine the POI tag matching the POI based on the target object trajectory type corresponding to the POI. The POI tag is used to represent the scene type corresponding to the POI. The correspondence between different target object trajectory types and POI tags is preset. When there are multiple POI tags corresponding to the target object trajectory type, the POI tag corresponding to the POI is determined by combining the POI location and scene information. The point of interest name determination unit is used to determine the name of the point of interest based on the point of interest label and at least one piece of scene information of the point of interest; The Point of Interest (POI) adding module is used to add POIs to the map based on their location and name.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for adding points of interest as described in any one of claims 1-6.

9. A storage medium for storing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the method for adding points of interest as described in any one of claims 1-6.

Images