Method and system for finding and locating personnel based on multi-camera linkage in a park

Abstract

The present invention relates to the technical field of intelligent management of park personnel, and discloses a method and system for finding and locating personnel based on the linkage of multiple cameras in the park. The method comprises the following steps: performing a first feature matching and personnel identification and positioning of a first feature information group consisting of a small number of feature information within a large area on the monitored personnel in the video frame of the retrieved park monitoring video stream; optimizing the first feature information group into a second feature information group according to the candidate search and positioning personnel in the positioning result; updating the initial search and positioning area into a real-time search and positioning area based on the regional positioning monitoring comparison relationship table; thereby performing a second feature matching and personnel identification and positioning of a large number of feature information within a small area on the monitored personnel; utilizing a distributed processing mechanism of preliminary screening and precise positioning to avoid resource consumption of a large number of useless features in the process of personnel identification, reducing the pressure of data processing load in the park, and improving the efficiency of finding and locating personnel based on the linkage of multiple cameras in the park.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent management of park personnel, and in particular to a method and system for finding and locating personnel based on the linkage of multiple cameras in a park. Background Art

[0002] Intelligent park personnel management uses advanced information technologies, such as the Internet of Things, big data, and artificial intelligence, to comprehensively perceive, analyze, and manage the information, behaviors, and activities of people within the park, in order to achieve efficient, safe, and convenient park operations and improve the park's service quality and management level. The application of this technology can improve the park's emergency response efficiency, optimize operational management, and ensure park safety.

[0003] In actual applications, smart management of park personnel is usually based on multiple cameras installed in the park. The information collection network constructed by multiple cameras has the characteristics of strong intuitiveness, rich information, non-contact monitoring and good traceability. It has good application effects in the smart management of park personnel in most scenarios. However, some scenarios still have application limitations. First, when there are many people in the park, identifying people in the surveillance images captured by cameras is subject to a large data processing load. Especially in high-precision recognition scenarios, each frame of surveillance footage captured by each camera requires a feature group consisting of a large number of features for feature matching and identification (for example, facial features and facial contour features under the facial feature type; height features, body features, limb movement features, and posture features under the action feature type; color features, style features, accessories features, and carried equipment features under the clothing feature type). Such recognition requirements will reduce the efficiency and real-time positioning of park personnel, affecting the efficiency of park emergency response and safety management. Second, in large-scale parks with high confidentiality requirements, if a production failure or accident occurs in a certain area, relevant processing personnel need to be notified and dispatched to the area to perform maintenance and processing. However, due to the restrictions on the carrying of communication equipment within the park, relevant processing personnel will perform tasks in different production areas of the large park, which increases the difficulty of locating park personnel and communicating information.

[0004] Therefore, how to improve the efficiency of finding and locating personnel through the linkage of multiple cameras in the park, reduce the data processing load pressure of intelligent management of park personnel, improve the accuracy of positioning park personnel and reduce the difficulty of conveying information to the positioned personnel is a technical problem that needs to be solved urgently. Summary of the Invention

[0005] The main purpose of the present invention is to provide a method and system for finding and locating personnel based on the linkage of multiple cameras in a park, aiming to solve at least one of the above-mentioned technical problems.

[0006] To achieve the above object, the present invention provides a method for finding and locating personnel based on the linkage of multiple cameras in a park, the method comprising the following steps:

[0007] According to the layout information of several key monitoring points in the target park and the positioning area and path trajectory in the target park, a regional positioning monitoring comparison relationship table is established;

[0008] Upon receiving a request for personnel location search, the system retrieves the park surveillance video information; wherein the park surveillance video information includes the park surveillance video stream collected at each key monitoring point;

[0009] Parsing the person identifier in the person search and positioning request, determining a feature information set of the target person to be located and an initial search and positioning area, and performing a first person positioning action using a first feature information group in the feature information set and an identification feature information group of each monitored person in each park surveillance video stream within the initial search and positioning area;

[0010] Based on the positioning result of the first personnel positioning action, the first feature information group is optimized into a second feature information group, and based on the regional positioning monitoring comparison relationship table, the initial search positioning area is updated to a real-time search positioning area;

[0011] The second feature information group in the feature information set and the identification feature information group of each monitored person in the real-time search and positioning area are used to perform a second personnel positioning action to generate the location information of the target search and positioning person in the target park.

[0012] Optionally, the step of establishing a regional positioning monitoring comparison relationship table based on the layout information of several key monitoring points in the target park and the positioning areas and path trajectories in the target park specifically includes:

[0013] Obtaining the layout information of several key monitoring points in the target park; wherein the layout information includes the monitoring camera layout position and monitoring camera identification of each key monitoring point;

[0014] Calling the regional electronic map of the target park, querying the regional locations of several positioning areas recorded in the regional electronic map and the path track connecting each positioning area;

[0015] A regional positioning monitoring comparison relationship table is established based on the monitoring camera layout position and monitoring camera identification of each key monitoring point, the regional position of several positioning areas, and the path trajectory connecting each positioning area.

[0016] Optionally, the step of establishing a regional positioning monitoring comparison relationship table based on the monitoring camera layout position and monitoring camera identification of each key monitoring point and the regional positions of several positioning areas and the path trajectory connecting each positioning area specifically includes:

[0017] According to the monitoring camera identification of each key monitoring point, query the monitoring camera's layout angle, layout height and focal length parameters, and determine the monitoring area of each key monitoring point based on the monitoring camera's layout position, layout angle, layout height and focal length parameters of each key monitoring point;

[0018] According to the regional position of each positioning area and the path trajectory connecting each positioning area, and in accordance with the relationship between the monitoring area of each key monitoring point and the position of each trajectory point in the path trajectory, a number of associated key monitoring points are allocated to each positioning area;

[0019] The association relationship between each positioning area and several key monitoring points is established as a regional positioning monitoring comparison relationship table.

[0020] Optionally, the personnel search and location request includes conditional trigger generation and active trigger generation;

[0021] The conditional trigger generation is configured to generate a personnel location request corresponding to the personnel to be located at a preset time interval, and the duration of the preset time interval is associated with the type of the personnel to be located;

[0022] The active trigger generation is configured to generate a person search and location request corresponding to the person search and location according to the person search and location in the user request information.

[0023] Optionally, parsing the person identifier in the person search and positioning request to determine the feature information set of the target person to be located and the initial search and positioning area step specifically includes:

[0024] Parsing the personnel location search request and extracting the personnel identifier of the target person to be located;

[0025] Using the personnel identification, matching the feature information set of the target person to be located in the target park personnel registration database; wherein the feature information set is configured as a feature set consisting of a plurality of facial features, a plurality of action features, and a plurality of clothing features of the target person to be located;

[0026] Using the personnel identification, the target park personnel positioning database is searched for the area union of the target positioning personnel's search and positioning areas determined when the personnel search and positioning requests were triggered a preset number of times in the past, and the area union is used as the initial search and positioning area for the target positioning personnel.

[0027] Optionally, performing a first personnel positioning step using the first characteristic information group in the characteristic information set and the identification characteristic information group of each monitored person in each park monitoring video stream within the initial search and positioning area specifically includes:

[0028] Extracting a first feature information group from the feature information set; wherein the first feature information group includes feature information of a first target type quantity selected from the feature recognition priority list;

[0029] The first personnel positioning action is performed using the first characteristic information group and the identification characteristic information group of the characteristic information type corresponding to the first characteristic information group for each monitored personnel in each park monitoring video stream within the initial search and positioning area.

[0030] Optionally, the feature recognition priority list is configured to store a number of feature information of each target park personnel arranged according to feature recognition priority, and the feature recognition priority of each feature information is negatively correlated to the proportion of the feature information type corresponding to the feature information among all target park personnel.

[0031] Optionally, based on the positioning result of the first personnel positioning action, the first feature information group is optimized into a second feature information group, and based on the regional positioning monitoring comparison relationship table, the initial search for the positioning area is updated to a real-time search for the positioning area step, specifically including:

[0032] Obtaining a positioning result of the first personnel positioning action; wherein the positioning result includes a plurality of candidate positioning personnel obtained by comparing and identifying the feature information of the first target type quantity with the identification feature information group of each monitored personnel in each park monitoring video stream;

[0033] Counting the actual number of candidates for location search, and optimizing the feature information of the first target category in the first feature information group to the feature information of the second target category in accordance with the ratio of the actual number to the preset number of candidates, to construct a second feature information group;

[0034] Acquire several candidate key monitoring points collected for candidate search and positioning personnel, and update the sum of the positioning areas associated with the candidate key monitoring points to the real-time search and positioning area based on the association between each positioning area and the key monitoring points recorded in the regional positioning and monitoring comparison relationship table.

[0035] Optionally, performing a second personnel positioning action using the second characteristic information group in the characteristic information set and the identification characteristic information group of each monitored person in the surveillance video stream of each park in the real-time positioning area to generate position information of the target positioning person in the target park specifically includes:

[0036] Using the newly added features in the second feature information group compared to the first feature information group in the feature information set and the identification feature information corresponding to the newly added feature types in the identification feature information group of each monitoring personnel in the monitoring video stream of each park in the real-time positioning area, feature comparison and identification are performed;

[0037] The monitored person with the highest feature matching rate is used as the person matching result for performing the second personnel positioning action, and the positioning area where the monitored person is located is used as the target to find the location information of the positioned person in the target park.

[0038] In addition, in order to achieve the above-mentioned purpose, the present invention also provides a system for finding and locating personnel based on the linkage of multiple cameras in a park, comprising:

[0039] Establish a module for establishing a regional positioning monitoring comparison relationship table based on the layout information of several key monitoring points in the target park and the positioning area and path trajectory in the target park;

[0040] A retrieval module is used to retrieve park surveillance video information when receiving a personnel location request; wherein the park surveillance video information includes the park surveillance video stream collected at each key monitoring point;

[0041] A parsing module is used to parse the personnel identification in the personnel search and positioning request, determine the feature information set of the target personnel to be located and the initial search and positioning area, and perform a first personnel positioning action using the first feature information group in the feature information set and the identification feature information group of each monitored personnel in each park monitoring video stream within the initial search and positioning area;

[0042] an optimization module, configured to optimize the first feature information group into a second feature information group based on the positioning result of the first personnel positioning action, and update the initial search and positioning area into a real-time search and positioning area based on the regional positioning monitoring comparison relationship table;

[0043] A generation module is used to perform a second personnel positioning action using the second feature information group in the feature information set and the identification feature information group of each monitored person in the monitoring video stream of each park in the real-time positioning area to generate the location information of the target positioning person in the target park.

[0044] The beneficial effects of the present invention are: a method and system for finding and locating personnel based on the linkage of multiple cameras in a park are proposed, by performing the first feature matching and personnel identification and positioning of a first feature information group composed of a small number of feature information within a large area on the monitored personnel in the video frame of the retrieved park monitoring video stream, optimizing the first feature information group into a second feature information group according to the candidate search and positioning personnel in the positioning result, updating the initial search and positioning area to a real-time search and positioning area based on the regional positioning monitoring comparison relationship table, thereby performing the second feature matching and personnel identification and positioning of a large number of feature information within a small area on the monitored personnel, utilizing the distributed processing mechanism of preliminary screening and precise positioning to avoid the resource consumption of a large number of useless features in the process of personnel identification, reducing the data processing load pressure of the park, and improving the efficiency of finding and positioning personnel in the linkage of multiple cameras in the park. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of the method for finding and locating personnel based on the linkage of multiple cameras in a park according to the present invention;

[0046] Figure 2 This is a structural diagram of the system for finding and locating personnel based on the linkage of multiple cameras in a park. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0048] The embodiment of the present invention provides a method for finding and locating personnel based on the linkage of multiple cameras in a park, referring to Figure 1 , Figure 1 The present invention is a flowchart of an embodiment of a method for finding and locating personnel based on the linkage of multiple cameras in a park.

[0049] In this embodiment, a method for finding and locating personnel based on the linkage of multiple cameras in a park is provided, and the method includes the following steps:

[0050] S100: establishing a regional positioning monitoring comparison relationship table based on the layout information of several key monitoring points in the target park and the positioning areas and path trajectories in the target park;

[0051] S200: Upon receiving a personnel location request, retrieve park surveillance video information; wherein the park surveillance video information includes park surveillance video streams collected at each key monitoring point;

[0052] S300: Parse the person identifier in the person search and positioning request, determine the feature information set of the target person to be located and the initial search and positioning area, and perform a first person positioning operation using the first feature information group in the feature information set and the identification feature information group of each monitored person in each park surveillance video stream within the initial search and positioning area;

[0053] S400: Based on the positioning result of the first personnel positioning action, the first feature information group is optimized into a second feature information group, and based on the regional positioning monitoring comparison relationship table, the initial search positioning area is updated to a real-time search positioning area;

[0054] S500: Utilize the second feature information group in the feature information set and the identification feature information group of each monitored person in the real-time search and positioning area in the monitoring video stream of each park to perform a second personnel positioning action, and generate the location information of the target search and positioning person in the target park.

[0055] It should be noted that the intelligent management of park personnel is usually based on multiple cameras installed in the park. The information collection network constructed by multiple cameras has the characteristics of strong intuitiveness, rich information, non-contact monitoring and good traceability. It has good application effects in the intelligent management of park personnel in most scenarios. However, some scenarios still have application limitations. First, when there are many people in the park, identifying people in the surveillance images captured by cameras is subject to a large data processing load. Especially in high-precision recognition scenarios, each frame of surveillance footage captured by each camera requires a feature group consisting of a large number of features for feature matching and identification (for example, facial features and facial contour features under the facial feature type; height features, body features, limb movement features, and posture features under the action feature type; color features, style features, accessories features, and carried equipment features under the clothing feature type). Such recognition requirements will reduce the efficiency and real-time positioning of park personnel, affecting the efficiency of park emergency response and safety management. Second, in large-scale parks with high confidentiality requirements, if a production failure or accident occurs in a certain area, relevant processing personnel need to be notified and dispatched to the area to perform maintenance and processing. However, due to the restrictions on the carrying of communication equipment within the park, relevant processing personnel will perform tasks in different production areas of the large park, which increases the difficulty of locating park personnel and communicating information.

[0056] In order to solve the above problems, this embodiment performs the first feature matching and personnel identification and positioning of the first feature information group composed of a small number of feature information within a large area on the monitored personnel in the video frame of the retrieved park monitoring video stream, optimizes the first feature information group into a second feature information group according to the candidate search and positioning personnel in the positioning result, and updates the initial search and positioning area to the real-time search and positioning area based on the regional positioning monitoring comparison relationship table, so as to perform the second feature matching and personnel identification and positioning of a large number of feature information within a small area on the monitored personnel, and utilizes the distributed processing mechanism of preliminary screening and precise positioning to avoid the resource consumption of a large number of useless features in the personnel identification process, reduce the data processing load pressure of the park, improve the personnel search and positioning efficiency of the multi-camera linkage in the park, improve the accuracy of the positioning of park personnel and reduce the difficulty of conveying information to the positioned personnel.

[0057] In a preferred embodiment, the step of establishing a regional positioning monitoring comparison relationship table based on the layout information of several key monitoring points in the target park and the positioning area and path trajectory in the target park specifically includes:

[0058] S110: Acquire layout information of several key monitoring points in the target park; wherein the layout information includes the monitoring camera layout position and monitoring camera identification of each key monitoring point;

[0059] S120: calling the regional electronic map of the target park, querying the regional locations of several positioning areas recorded in the regional electronic map and the path track connecting each positioning area;

[0060] S130: Establishing a regional positioning monitoring comparison relationship table based on the monitoring camera layout position and monitoring camera identification of each key monitoring point, the regional positions of several positioning areas, and the path trajectory connecting each positioning area.

[0061] Furthermore, according to the monitoring camera layout position and monitoring camera identification of each key monitoring point and the regional position of several positioning areas and the path trajectory connecting each positioning area, the regional positioning monitoring comparison relationship table is established, which specifically includes:

[0062] S131: According to the monitoring camera identification of each key monitoring point, query the deployment angle, deployment height and focal length parameters of the monitoring camera, and determine the monitoring area of each key monitoring point based on the deployment position, deployment angle, deployment height and focal length parameters of the monitoring camera at each key monitoring point;

[0063] S132: assigning a number of associated key monitoring points to each positioning area according to the regional position of each positioning area and the path track connecting each positioning area, and according to the relationship between the monitoring area of each key monitoring point and the position of each track point in the path track;

[0064] S133: Establish the association relationship between each positioning area and several key monitoring points as a regional positioning monitoring comparison relationship table.

[0065] In this embodiment, by obtaining the layout information of each key monitoring point and calling the regional electronic map of the target park, the path trajectory of each positioning area is queried, and then the layout angle, layout height and focal length parameters of the monitoring camera are queried according to the layout position and monitoring camera identification of the monitoring camera, so as to determine the monitoring area of each key monitoring point. Finally, by considering the relationship between the monitoring area of each key monitoring point and the position of each trajectory point in the path trajectory, a regional positioning monitoring comparison relationship table is established. Therefore, by establishing the association relationship between the key monitoring point and each positioning area in the park, after the candidate search and positioning personnel are subsequently determined, the initial search and positioning area of each candidate search and positioning personnel can be updated to the real-time search and positioning area through the regional positioning monitoring comparison relationship table, thereby reducing the data processing load pressure of the intelligent management of park personnel.

[0066] In a preferred embodiment, the personnel search and positioning request includes conditional trigger generation and active trigger generation; wherein, the conditional trigger generation is configured to generate a personnel search and positioning request corresponding to the person to be located at a preset time interval, and the duration of the preset time is associated with the type of person to be located; wherein, the active trigger generation is configured to generate a personnel search and positioning request corresponding to the person to be located according to the person to be located in the user request information.

[0067] In this embodiment, the triggering of the personnel search and positioning request can include passive triggering of the interval time and active triggering of the user request. By passively triggering the generation of the personnel search and positioning request, the location information of each person being searched and positioned can be updated in real time. The location information is used to determine the real-time search and positioning area when the person being searched and positioned is determined to be a candidate for the person being searched and positioned. By actively triggering the generation of the personnel search and positioning request, the user's search and positioning needs for the target person can be responded to in a timely manner. It should be noted that, taking into account the importance and urgency of the personnel in different parks, a corresponding interval preset time can be assigned to each person being searched and positioned according to the type of personnel. For personnel with high importance and high demand, a shorter interval preset time can be assigned, thereby obtaining better real-time location information, making the generated real-time search and positioning area more accurate. For personnel with low importance and low demand, a longer interval preset time can be assigned to reduce the system computing load.

[0068] In a preferred embodiment, parsing the person identifier in the person search and positioning request, determining the feature information set of the target person to be located and the initial search and positioning area step specifically includes:

[0069] S310: Parse the personnel location search request and extract the personnel identifier of the target person to be located;

[0070] S320: Using the personnel identifier, matching a feature information set of the target person to be located in a personnel registration database of the target park; wherein the feature information set is configured as a feature set consisting of a plurality of facial features, a plurality of action features, and a plurality of clothing features of the target person to be located;

[0071] S330: Using the personnel identification, search the target park personnel positioning database for the area union of the search and positioning areas determined when the personnel search and positioning requests of the target person were triggered a preset number of times in the past, and use the area union as the initial search and positioning area of the target person.

[0072] On this basis, the first personnel positioning step is performed using the first characteristic information group in the characteristic information set and the identification characteristic information group of each monitored person in each park monitoring video stream in the initial search and positioning area, specifically including:

[0073] S340: extracting a first feature information group from the feature information set; wherein the first feature information group includes feature information of a first target type quantity selected from the feature recognition priority list;

[0074] S350 performs a first personnel positioning action by using the first feature information group and the identification feature information group of the feature information type corresponding to the first feature information group for each monitored personnel in each park monitoring video stream within the initial search and positioning area.

[0075] In actual application, the feature recognition priority list is configured to store a number of feature information of each target park personnel arranged according to feature recognition priority, and the feature recognition priority of each feature information is negatively correlated to the proportion of the feature information type corresponding to the feature information among all target park personnel.

[0076] In this embodiment, by parsing the person identifier in the personnel search and location request, matching the feature information set of the target person to be located with the target campus personnel registration database, and searching the target campus personnel location database for the initial search and location area of the target person to be located. Subsequently, a first feature information group is constructed by selecting the feature information of the first number of target categories from the feature recognition priority list. This first feature information group is then used to perform feature matching and personnel identification and location using the identification feature information group of each monitored person in each campus surveillance video stream within the initial search and location area. It should be noted that the number of feature information of each target campus person is determined based on the proportion of each feature information present in all target campus persons. In this embodiment, a negative correlation is set, i.e., a higher the proportion, the fewer people to be located with this feature information during identification and screening, and the feature information is positioned further back in the feature recognition priority list. Conversely, a lower the proportion, the higher the position. This negative correlation can screen out more non-targeted people to be located during the initial screening process of the personnel location operation, thereby reducing the data processing load of the intelligent management of campus personnel.

[0077] In a preferred embodiment, based on the positioning result of the first personnel positioning action, the first feature information group is optimized into the second feature information group, and based on the regional positioning monitoring comparison relationship table, the initial search for the positioning area is updated to the real-time search for the positioning area step, which specifically includes:

[0078] S410: Obtaining a positioning result of the first personnel positioning action; wherein the positioning result includes a plurality of candidate positioning personnel obtained by comparing and identifying feature information based on the number of feature information of the first target type and the identification feature information group of each monitored person in each park monitoring video stream;

[0079] S420: Counting the actual number of candidates for location search, and optimizing the feature information of the first target category in the first feature information group to feature information of the second target category in accordance with the ratio of the actual number to the preset number of candidates, thereby constructing a second feature information group;

[0080] S430: Acquire several candidate key monitoring points collected for candidate search and positioning personnel, and update the sum of the positioning areas associated with the candidate key monitoring points into the real-time search and positioning area according to the association between each positioning area and the key monitoring points recorded in the regional positioning and monitoring comparison relationship table.

[0081] On this basis, the second feature information group in the feature information set and the identification feature information group of each monitored person in the surveillance video stream of each park in the real-time search and positioning area are used to perform a second personnel positioning action to generate the position information of the target search and positioning person in the target park, specifically including:

[0082] S510: performing feature comparison and identification using the newly added features in the second feature information group compared to the first feature information group in the feature information set and the identification feature information corresponding to the newly added feature types in the identification feature information group of each monitored person in the surveillance video stream of each park in the real-time location area;

[0083] S520: The monitored person with the highest feature matching rate is used as the person matching result for performing the second person positioning action, and the positioning area where the monitored person is located is used as the target to search for the position information of the positioned person in the target park.

[0084] In this embodiment, a person identification is used to determine a feature information set to initially search for a positioning area, thereby performing a first feature matching and personnel identification and positioning of a first feature information group consisting of a small number of feature information within a large area for the monitoring personnel in the video frame of the retrieved park monitoring video stream. The first feature information group is optimized to a second feature information group based on the candidate search and positioning personnel in the positioning result. The initial search and positioning area is updated to a real-time search and positioning area based on the regional positioning monitoring comparison relationship table. Thus, a second feature matching and personnel identification and positioning of a large number of feature information within a small area is performed for the monitoring personnel in the video frame of the retrieved park monitoring video stream. By performing feature matching and personnel identification twice, a small number of feature matching is used to perform a preliminary screening of the personnel identification and positioning area, narrowing the regional positioning range and reducing the amount of data processed by the personnel identification feature matching. Then, a large number of features are used to perform the precise positioning of the target person. Thus, by performing feature matching and personnel identification twice, the present invention utilizes a distributed processing mechanism of preliminary screening and precise positioning to avoid the resource consumption of a large number of useless features in the personnel identification process, reduce the data processing load pressure of the park personnel intelligent management, improve the efficiency of personnel search and positioning in the park multi-camera linkage, improve the accuracy of park personnel positioning and reduce the difficulty of transmitting information to the positioning personnel.

[0085] Reference Figure 2 , Figure 2 This is a structural block diagram of an embodiment of the system for finding and locating personnel based on the linkage of multiple cameras in a park according to the present invention.

[0086] like Figure 2 As shown, the system for finding and locating personnel based on multi-camera linkage in a park proposed in an embodiment of the present invention includes:

[0087] Establishing module 10, for establishing a regional positioning monitoring comparison relationship table based on the layout information of several key monitoring points in the target park and the positioning area and path trajectory in the target park;

[0088] The retrieval module 20 is used to retrieve the park monitoring video information when receiving a personnel location request; wherein the park monitoring video information includes the park monitoring video stream collected at each key monitoring point;

[0089] A parsing module 30 is configured to parse the person identifier in the person location request, determine a feature information set of the target person to be located and an initial location search area, and perform a first location operation using a first feature information group in the feature information set and an identification feature information group of each monitored person in each park surveillance video stream within the initial location search area;

[0090] an optimization module 40 for optimizing the first feature information group into a second feature information group based on the positioning result of the first personnel positioning action, and updating the initial search and positioning area into a real-time search and positioning area based on the regional positioning monitoring comparison relationship table;

[0091] Generation module 50 is used to perform a second personnel positioning action using the second characteristic information group in the characteristic information set and the identification characteristic information group of each monitored personnel in the monitoring video stream of each park in the real-time positioning area to generate the location information of the target positioning personnel in the target park.

[0092] Other embodiments or specific implementation methods of the system for finding and locating personnel based on the linkage of multiple cameras in a park of the present invention can refer to the above-mentioned method embodiments and will not be repeated here.

[0093] It should be understood that, in the description of this specification, reference to terms such as "one embodiment," "another embodiment," "other embodiments," or "first to Nth embodiments" means that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples.

[0094] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0095] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for finding and locating personnel based on the linkage of multiple cameras in a park, characterized in that: The method comprises the following steps: Based on the layout information of several key monitoring points in the target park and the positioning areas and path trajectories in the target park, a regional positioning monitoring comparison relationship table is established; specifically including: Obtaining the layout information of several key monitoring points in the target park; wherein the layout information includes the monitoring camera layout position and monitoring camera identification of each key monitoring point; Calling the regional electronic map of the target park, querying the regional locations of several positioning areas recorded in the regional electronic map and the path track connecting each positioning area; According to the monitoring camera identification of each key monitoring point, query the monitoring camera's layout angle, layout height and focal length parameters, and determine the monitoring area of each key monitoring point based on the monitoring camera's layout position, layout angle, layout height and focal length parameters of each key monitoring point; According to the regional position of each positioning area and the path trajectory connecting each positioning area, and in accordance with the relationship between the monitoring area of each key monitoring point and the position of each trajectory point in the path trajectory, a number of associated key monitoring points are allocated to each positioning area; Establish the association relationship between each positioning area and several key monitoring points as a regional positioning monitoring comparison relationship table; Upon receiving a request for personnel location search, the system retrieves the park surveillance video information; wherein the park surveillance video information includes the park surveillance video stream collected at each key monitoring point; Parsing the person identifier in the person search and positioning request, determining a feature information set of the target person to be located and an initial search and positioning area, and performing a first person positioning action using a first feature information group in the feature information set and an identification feature information group of each monitored person in each park surveillance video stream within the initial search and positioning area; Based on the positioning result of the first personnel positioning action, the first feature information group is optimized into a second feature information group, and based on the regional positioning monitoring comparison relationship table, the initial search positioning area is updated to a real-time search positioning area; The second feature information group in the feature information set and the identification feature information group of each monitored person in the real-time search and positioning area are used to perform a second personnel positioning action to generate the location information of the target search and positioning person in the target park.

2. The method for finding and locating personnel based on multi-camera linkage in a park as claimed in claim 1, characterized in that: The personnel search and positioning request includes conditional trigger generation and active trigger generation; The conditional trigger generation is configured to generate a personnel location request corresponding to the personnel to be located at a preset time interval, and the duration of the preset time interval is associated with the type of the personnel to be located; The active trigger generation is configured to generate a person search and location request corresponding to the person search and location according to the person search and location in the user request information.

3. The method for finding and locating personnel based on multi-camera linkage in a park as claimed in claim 2, characterized in that: The steps of parsing the person identifier in the person search and positioning request, determining the feature information set of the target person to be located and the initial search and positioning area, specifically include: Parsing the personnel location search request and extracting the personnel identifier of the target person to be located; Using the personnel identification, matching the feature information set of the target person to be located in the target park personnel registration database; wherein the feature information set is configured as a feature set consisting of a plurality of facial features, a plurality of action features, and a plurality of clothing features of the target person to be located; Using the personnel identification, the target park personnel positioning database is searched for the area union of the target positioning personnel's search and positioning areas determined when the personnel search and positioning requests were triggered a preset number of times in the past, and the area union is used as the initial search and positioning area for the target positioning personnel.

4. The method for finding and locating personnel based on multi-camera linkage in a park as claimed in claim 1, characterized in that: The first step of performing personnel positioning is performed using the first characteristic information group in the characteristic information set and the identification characteristic information group of each monitored person in each park monitoring video stream within the initial search and positioning area, specifically including: Extracting a first feature information group from the feature information set; wherein the first feature information group includes feature information of a first target type quantity selected from the feature recognition priority list; The first personnel positioning action is performed using the first characteristic information group and the identification characteristic information group of the characteristic information type corresponding to the first characteristic information group for each monitored personnel in each park monitoring video stream within the initial search and positioning area.

5. The method for finding and locating personnel based on multi-camera linkage in a park as claimed in claim 4, characterized in that: The feature recognition priority list is configured to store a plurality of feature information of each target park personnel arranged according to feature recognition priority, and the feature recognition priority of each feature information is negatively correlated to the proportion of the feature information type corresponding to the feature information among all target park personnel.

6. The method for finding and locating personnel based on multi-camera linkage in a park as claimed in claim 4, characterized in that: Based on the positioning result of the first personnel positioning action, the first feature information group is optimized into a second feature information group, and based on the regional positioning monitoring comparison relationship table, the initial search for positioning area is updated to a real-time search for positioning area step, specifically including: Obtaining a positioning result of the first personnel positioning action; wherein the positioning result includes a plurality of candidate positioning personnel obtained by comparing and identifying the feature information of the first target type quantity with the identification feature information group of each monitored personnel in each park monitoring video stream; Counting the actual number of candidates for location search, and optimizing the feature information of the first target category in the first feature information group to the feature information of the second target category in accordance with the ratio of the actual number to the preset number of candidates, to construct a second feature information group; Acquire several candidate key monitoring points collected for candidate search and positioning personnel, and update the sum of the positioning areas associated with the candidate key monitoring points to the real-time search and positioning area based on the association between each positioning area and the key monitoring points recorded in the regional positioning and monitoring comparison relationship table.

7. The method for finding and locating personnel based on multi-camera linkage in a park as claimed in claim 6, characterized in that: The steps of performing a second personnel positioning action by using the second characteristic information group in the characteristic information set and the identification characteristic information group of each monitored person in the monitoring video stream of each park in the real-time search and positioning area to generate the position information of the target search and positioning person in the target park specifically include: Using the newly added features in the second feature information group compared to the first feature information group in the feature information set and the identification feature information corresponding to the newly added feature types in the identification feature information group of each monitoring personnel in the monitoring video stream of each park in the real-time positioning area, feature comparison and identification are performed; The monitored person with the highest feature matching rate is used as the person matching result for performing the second personnel positioning action, and the positioning area where the monitored person is located is used as the target to find the location information of the positioned person in the target park.

8. A system for finding and locating people based on the linkage of multiple cameras in a park, characterized by: A method for finding and locating personnel based on multi-camera linkage in a park as described in any one of claims 1 to 7, the system comprising: Establish a module for establishing a regional positioning monitoring comparison relationship table based on the layout information of several key monitoring points in the target park and the positioning area and path trajectory in the target park; A retrieval module is used to retrieve park surveillance video information when receiving a personnel location request; wherein the park surveillance video information includes the park surveillance video stream collected at each key monitoring point; A parsing module is used to parse the personnel identification in the personnel search and positioning request, determine the feature information set of the target personnel to be located and the initial search and positioning area, and perform a first personnel positioning action using the first feature information group in the feature information set and the identification feature information group of each monitored personnel in each park monitoring video stream within the initial search and positioning area; an optimization module, configured to optimize the first feature information group into a second feature information group based on the positioning result of the first personnel positioning action, and update the initial search and positioning area into a real-time search and positioning area based on the regional positioning monitoring comparison relationship table; A generation module is used to perform a second personnel positioning action using the second feature information group in the feature information set and the identification feature information group of each monitored person in the monitoring video stream of each park in the real-time positioning area to generate the location information of the target positioning person in the target park.

Images