Abnormal population detection method, device, equipment, storage medium and product

Abstract

The application discloses an abnormal crowd detection method, device, equipment, storage medium and product; the method comprises the following steps: screening a target monitoring image containing a target person from a monitoring image set; determining a first international mobile subscriber identity (IMSI) code set associated with the target person according to a first collection position and a first collection time of the target monitoring image; determining a crowd anomaly index of a target area according to the first IMSI code set and a second IMSI code set, wherein the second IMSI code set comprises IMSI codes collected in the target area, and the crowd anomaly index is used for representing the possibility of abnormal gathering behavior in the target area; and if the crowd anomaly index of the target area exceeds a preset threshold, generating an alarm information. The application can realize efficient positioning analysis on abnormal personnel or crowds without pre-acquiring the IMSI code of the abnormal personnel, and greatly improves the detection efficiency of the abnormal personnel or crowds.

Description

Technical Field

[0001] This invention relates to the field of video surveillance and security technology, and in particular to methods, devices, equipment, storage media, and products for detecting abnormal crowds. Background Technology

[0002] With the widespread deployment of video surveillance equipment in public areas and the rapid development of artificial intelligence technology, intelligent detection of abnormal groups based on video surveillance data has become a key focus of current research and practice.

[0003] Currently, existing technologies typically employ facial recognition algorithms to identify specific individuals in a camera, collect the International Mobile Subscriber Identity (IMSI) of these individuals using a code-collecting device, and then track their movements based on the IMSI code to achieve efficient location tracking.

[0004] However, the above method requires obtaining the IMSI code of a specific person in advance. If the specific person changes their mobile phone number, it will be impossible to locate the specific person or group. Therefore, there is an urgent need for a more efficient method for detecting abnormal groups. Summary of the Invention

[0005] This invention provides a method, apparatus, device, storage medium, and product for detecting abnormal populations, enabling the automatic and efficient identification of International Mobile Subscriber Identity (IMSI) codes associated with specific individuals, and the monitoring and location of their suspected companions, thereby improving the detection efficiency of abnormal individuals and / or populations.

[0006] According to one aspect of the present invention, a method for detecting abnormal populations is provided, comprising: Filter the surveillance images from the surveillance image set to find target surveillance images containing the target person; Based on the first acquisition location and the first acquisition time of the target surveillance image, determine the first International Mobile Subscriber Identity (IMSI) code set associated with the target person; Based on the first IMSI code set and the second IMSI code set, a population anomaly index for the target area is determined. The second IMSI code set includes IMSI codes collected within the target area. The population anomaly index is used to characterize the possibility of abnormal clustering behavior in the target area. If the abnormal population index in the target area exceeds a preset threshold, an alarm message will be generated.

[0007] According to another aspect of the present invention, an abnormal population detection device is provided, comprising: The surveillance image filtering module is used to filter out target surveillance images containing the target personnel from the surveillance image set; The first IMSI code set determination module is used to determine the first International Mobile Subscriber Identity (IMSI) code set associated with the target person based on the first acquisition location and the first acquisition time of the target surveillance image. The crowd anomaly index determination module is used to determine the crowd anomaly index of a target area based on a first IMSI code set and a second IMSI code set, wherein the second IMSI code set includes IMSI codes collected in the target area, and the crowd anomaly index is used to characterize the possibility of abnormal clustering behavior in the target area. The alarm information generation module is used to generate alarm information if the population anomaly index in the target area exceeds a preset threshold.

[0008] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the abnormal population detection method according to any embodiment of the present invention.

[0009] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the abnormal population detection method according to any embodiment of the present invention.

[0010] According to another aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements the abnormal population detection method according to any embodiment of the present invention.

[0011] The technical solution of this invention identifies target personnel by filtering target surveillance images containing target personnel from a set of surveillance images. Based on the first acquisition location and first acquisition time of the target surveillance image, it determines the first International Mobile Subscriber Identity (IMSI) code set associated with the target personnel, enabling dynamic acquisition of the target personnel's IMSI code and solving the problem of being unable to identify and locate them due to changes in their mobile phone numbers. By determining the crowd anomaly index of the target area based on the first IMSI code set and the second IMSI code set, it achieves efficient monitoring of whether abnormal gathering behavior exists in the target area. The second IMSI code set includes IMSI codes acquired within the target area, and the crowd anomaly index characterizes the probability of abnormal gathering behavior in the target area. Furthermore, when the crowd anomaly index of the target area exceeds a preset threshold, it generates alarm information, achieving real-time detection of abnormal crowds in the target area. This invention can effectively improve the detection efficiency of abnormal personnel and / or crowds.

[0012] 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

[0013] 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.

[0014] Figure 1 This is a flowchart of an abnormal population detection method provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of an abnormal population detection method provided in Embodiment 2 of the present invention; Figure 3 This is a flowchart of an abnormal population detection method provided in Embodiment 3 of the present invention; Figure 4 This is a schematic diagram of the structure of an abnormal population detection device according to Embodiment 4 of the present invention; Figure 5 This is a schematic diagram of the structure of an electronic device that implements the abnormal population detection method of Embodiment 5 of the present invention. Detailed Implementation

[0015] 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. 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.

[0016] It should be noted that the terms "first," "second," and "target," 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 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 a non-exclusive inclusion; for example, a process, method, system, product, or apparatus 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 apparatus.

[0017] Example 1 Figure 1 This is a flowchart illustrating an abnormal population detection method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations involving the detection of abnormal individuals or populations. The method can be executed by an abnormal population detection device, which can be implemented in hardware and / or software and can be configured in an electronic device. Figure 1 As shown, the method includes: S101. Filter out target surveillance images containing the target personnel from the surveillance image set.

[0018] In this embodiment, the surveillance image set can be acquired by surveillance equipment deployed within the target surveillance area. Each image in the surveillance image set can carry the image's acquisition location information and acquisition time information. The target surveillance area may include one or more surveillance areas, the specifics of which can be determined based on the actual application scenario. The target personnel can be understood as specific individuals to be detected. The target surveillance images can be understood as images containing the target personnel selected from the surveillance image set.

[0019] Optionally, the monitoring image set includes multiple monitoring images containing image information of the person to be tested; wherein, the step of filtering target monitoring images containing the target person from the monitoring image set includes: extracting the human feature information of the person to be tested from the target monitoring images using a preset human feature recognition algorithm; determining the similarity score between the human feature information of the person to be tested and the human feature information of the target person using a preset matching function; and determining the monitoring images corresponding to the human feature information of the person to be tested with similarity scores greater than a similarity threshold as target monitoring images containing the target person. The advantage of this setup is that it allows for real-time monitoring of whether the target person appears, facilitating subsequent location and tracking; furthermore, compared to conventional techniques that use facial information for filtering, this technique can effectively improve the accuracy of filtering based on human feature information. For example, techniques that use facial information for filtering are limited by the shooting angle, making it difficult to obtain facial information and thus difficult to accurately identify the target person; while identifying the target person based on human feature information has lower requirements for the shooting angle, thus possessing higher reliability.

[0020] In this embodiment, the person to be tested can be understood as a person who can be effectively identified in the surveillance image. For example, a surveillance image in a set of surveillance images may contain three people, but due to the shooting angle of the surveillance equipment or environmental reasons, only two people can be effectively identified. These two people are the people to be tested, and the other person is not effectively identified and will not be processed subsequently. The preset human feature recognition algorithm can be a general human feature recognition algorithm, such as the Person Re-identification (ReID) algorithm. Human feature information may include human feature vectors extracted using the preset human feature recognition algorithm. The preset matching function can be used to calculate the similarity score between the human feature vector of the person to be tested and the human feature vector of the target person. The similarity score can be represented by the cosine similarity between the human feature vector of the person to be tested and the human feature vector of the target person. The human feature vector of the target person can be obtained in advance through the following steps: obtaining the image information of the target person using the surveillance equipment in the preset area and / or based on historical video; inputting the image information of the target person into the preset human feature recognition algorithm (e.g., the ReID algorithm) to generate the human feature vector of the target person. In addition, the generated human feature vectors of the target personnel can be stored for subsequent identification of the target personnel.

[0021] It should be noted that the acquisition, storage, use, and processing of data in the technical solution of this invention all comply with the relevant provisions of national laws and regulations.

[0022] For example, real-time surveillance images captured by cameras deployed at traffic intersections are acquired. The ReID algorithm is used to extract the human feature vectors of pedestrians contained in the surveillance images as the human feature information of the target person. A preset matching function is then used to calculate the similarity score between the current pedestrian's human feature vector and the target person's human feature vector. If the similarity score is greater than a similarity threshold, the current surveillance image is determined as the target surveillance image containing the target person. The preset matching function is expressed by the following expression: ; in, Represents the similarity score. The human feature vector representing the person being tested. A human feature vector representing the target person. The Euclidean norm, or L2 norm, represents the human feature vector of the person being tested. The Euclidean norm of the human feature vector representing the target person. If If so, the current person to be tested is identified as the target person, the current monitoring image is identified as the target monitoring image including the target person, and the acquisition location information and acquisition time information of the current monitoring image are recorded.

[0023] S102. Based on the first acquisition location and the first acquisition time of the target monitoring image, determine the set of first International Mobile Subscriber Identity (IMSI) codes associated with the target personnel.

[0024] In this embodiment, the first acquisition location can be understood as the acquisition location of the target surveillance image (which can be represented by latitude and longitude coordinates), and the first acquisition time can be understood as the acquisition time of the target surveillance image. Since the target surveillance image contains the target person, the first acquisition location and the first acquisition time can be determined as the location and time of the target person's appearance. The first IMSI code set includes the IMSI codes associated with the target person; for example, it can be the IMSI code set of the target person and their companions. The IMSI code is the unique identifier of the Subscriber Identity Module (SIM) in a mobile phone, and the IMSI code can be used to identify the target person.

[0025] For example, cameras are installed within the target monitoring area to obtain a set of monitoring images. IMSI code information is collected in real time near the cameras using mobile IMSI base station equipment. Each IMSI code also includes collection location and collection time information. After filtering out target monitoring images containing the target person, a first set of IMSI codes associated with the target person is determined based on the IMSI codes collected by the mobile IMSI base station equipment near the first collection location within a preset time period before and after the first collection time. The preset time period can be pre-set according to the collection range of the mobile IMSI base station equipment and the movement speed of the person. For example, in a traffic intersection scenario where the collection range of the mobile IMSI base station equipment is small and the movement speed of people is fast, a shorter preset time period (e.g., 10 seconds) can be set; in an open park scenario where the collection range of the mobile IMSI base station equipment is large and the movement speed of people is slow, a longer preset time period (e.g., 30 seconds) can be set. For example, the IMSI codes collected by the mobile detection base station equipment near the first collection location within a preset time before and after the first collection time are used as the IMSI codes of the people traveling with the target person; then, a first IMSI code set is constructed based on the IMSI codes of the people traveling with the target person.

[0026] S103. Determine the population anomaly index of the target area based on the first IMSI code set and the second IMSI code set, wherein the second IMSI code set includes the IMSI codes collected in the target area, and the population anomaly index is used to characterize the possibility of abnormal clustering behavior in the target area.

[0027] In this embodiment, the second IMSI code set can be understood as the IMSI code set collected by mobile detection base station equipment deployed in the target area. The target area can be understood as the area where abnormal crowd behavior needs to be detected; it can be the aforementioned target monitoring area or an area different from the aforementioned target monitoring area, set according to actual detection needs. Monitoring equipment and mobile detection base station equipment are deployed in the target area. The crowd anomaly index is used to characterize the probability of abnormal gathering behavior in the target area. Specifically, it can reflect the crowd density in the target area, the frequency of appearance of target personnel, and the proportion of abnormal personnel (which can be understood as target personnel and their companions) in the crowd gathered in the target area.

[0028] For example, the number of target individuals and their companions can be determined based on the number of IMSI codes in the first IMSI code set. Specifically, the number of companions of the target individual is determined based on the number of IMSI codes in the first IMSI code set, and then the number of target individuals (e.g., 1) is summed with the number of companions to obtain the total number of target individuals and their companions. Furthermore, when determining the number of companions of the target individual based on the number of IMSI codes in the first IMSI code set, a deduplication operation can be performed. Specifically, for two IMSI codes in the first IMSI code set, it can be determined whether the collection location information and collection time information attached to these two IMSI codes are consistent. If they are consistent, it means that these two IMSI codes belong to the same person, and the count only needs to be recorded once. The number of people in the target area can be determined based on the number of IMSI codes in the second IMSI code set, and then the population anomaly index of the target area can be calculated using the number of target individuals and their companions, the number of people in the target area, and the area of ​​the target area.

[0029] S104. If the abnormal population index in the target area exceeds the preset threshold, an alarm message will be generated.

[0030] In this embodiment, the preset threshold can be dynamically set according to the characteristics of the target area, thereby improving the accuracy of detecting crowd gathering behavior in the target area. For example, a smaller threshold can be set for areas with high pedestrian traffic (e.g., shopping malls or traffic intersections), while a larger threshold can be set for areas with low pedestrian traffic (e.g., school gates or library corners). In addition, the preset threshold can also be dynamically adjusted according to time periods. For example, a smaller threshold can be set for traffic intersection areas during commuting hours, and a larger threshold can be set for traffic intersection areas during non-commuting hours, thereby ensuring that the generated alarm information has a higher confidence level.

[0031] This invention provides a method for detecting abnormal crowds, which can automatically and efficiently identify IMSI codes associated with specific individuals and monitor and locate suspected companions, thereby improving the detection efficiency of abnormal individuals and / or crowds. By filtering target surveillance images containing the target individuals from a set of surveillance images, the method achieves the identification of the target individuals. Based on the first acquisition location and first acquisition time of the target surveillance image, it determines the first set of International Mobile Subscriber Identity (IMSI) codes associated with the target individuals, enabling dynamic acquisition of the target individuals' IMSI codes and solving the problem of being unable to identify and locate them due to changes in their mobile phone numbers. By determining the crowd anomaly index of the target area based on the first IMSI code set and the second IMSI code set, it achieves efficient monitoring of whether abnormal gathering behavior exists in the target area. The second IMSI code set includes IMSI codes acquired within the target area, and the crowd anomaly index characterizes the probability of abnormal gathering behavior in the target area. Furthermore, when the crowd anomaly index of the target area exceeds a preset threshold, it can generate alarm information, achieving real-time detection of abnormal crowds in the target area.

[0032] Example 2 Figure 2 This is a flowchart of an abnormal population detection method provided in Embodiment 2 of the present invention. This embodiment is a refinement based on the above embodiments. Figure 2 As shown, the method includes: S201. Filter out target surveillance images containing the target personnel from the surveillance image set.

[0033] Optionally, the set of monitoring images is acquired by a preset monitoring device, and a preset mobile detection base station device is set within a preset location range of the preset monitoring device.

[0034] In this embodiment, a preset monitoring device can be deployed within the target monitoring area. This device can be used to collect and store monitoring images in real time. A preset mobile code detection base station device can be set within a preset location range of the preset monitoring device (e.g., within a 2-meter interval between the preset monitoring devices). This device is used to collect IMSI codes within the target monitoring area. The collected IMSI codes include a second collection location (which can be represented by latitude and longitude coordinates) and a second collection time (the timestamp of the IMSI code collected by the mobile code detection base station device). By jointly deploying the monitoring device and the mobile code detection base station device within the target monitoring area, it is possible to quickly obtain the IMSI code information associated with a target person when their presence is detected, providing a foundation for subsequent monitoring of the target person.

[0035] S202. Based on the first acquisition location and the first acquisition time of the target monitoring image, select candidate IMSI codes that meet the preset conditions from the preset IMSI database.

[0036] The preset IMSI database includes IMSI codes reported by the preset mobile detection base station equipment. The preset conditions include that the deviation between the second acquisition position and the first acquisition position of the acquired IMSI code is within a preset position deviation range, and / or that the deviation between the second acquisition time and the first acquisition time of the acquired IMSI code is within a preset time deviation range. The advantage of this setting is that when filtering candidate IMSI codes in the preset IMSI database using the preset conditions, it can combine the positional correlation between the second acquisition position of the acquired IMSI code and the first acquisition position of the target monitoring image (which can be understood as the location where the target person appears), and / or the temporal correlation between the second acquisition time of the acquired IMSI code and the first acquisition time of the target monitoring image (which can be understood as the time when the target person appears), to quickly filter out IMSI codes near the location where the target person appears, and / or near the time when the target person appears, and use these IMSI codes as candidate IMSI codes. The preset position deviation range can be pre-set according to the acquisition range of the IMSI code; for example, for a small traffic intersection, a smaller preset position deviation range can be set (e.g., 0-5 meters); for a large park, a larger preset position deviation range can be set (e.g., 0-20 meters). Correspondingly, the preset time deviation range can be pre-set according to the IMSI code collection range and the speed of personnel flow; for example, for traffic intersections with a small range and a fast personnel flow speed, a smaller preset time deviation range can be set (e.g., 0-10 seconds); for parks with a large range and a fast personnel flow speed, a larger preset time deviation range can be set (e.g., 0-30 seconds).

[0037] In this embodiment, the target monitoring area may include one or more, and the preset mobile code detection base station devices within a target monitoring area may also include one or more. The IMSI codes collected by these preset mobile code detection base station devices can be reported to the data center to form a preset IMSI database.

[0038] For example, the preset conditions are that the deviation between the second acquisition position and the first acquisition position of the IMSI code is within a preset position deviation range, and the deviation between the second acquisition time and the first acquisition time of the IMSI code is within a preset time deviation range. Then, after filtering out target surveillance images containing the target person, IMSI codes that meet the preset conditions are selected as candidate IMSI codes from the preset IMSI database. The advantage of this setting is that the preset conditions fully consider the positional correlation between the second acquisition position of the IMSI code and the first acquisition position of the target surveillance image, as well as the temporal correlation between the second acquisition time of the IMSI code and the first acquisition time of the target surveillance image. This allows for the rapid acquisition of IMSI codes near the location and time of the target person's appearance when filtering candidate IMSI codes from the preset IMSI database, further improving the speed and accuracy of candidate IMSI code selection.

[0039] S203. Determine the first set of IMSI codes associated with the target person based on the candidate IMSI codes.

[0040] Optionally, the cumulative number of times each candidate IMSI code meets the preset conditions is determined; based on the top preset number of candidate IMSI codes with the highest cumulative counts, a first set of IMSI codes associated with the target person is determined. The advantage of this setup is that using the cumulative count as a filtering criterion can effectively filter out IMSI codes that occasionally appear in the target monitoring area, improving filtering accuracy; furthermore, using the top preset number of candidate IMSI codes with the highest cumulative counts as the first set of IMSI codes associated with the target person yields multiple IMSI codes related to the target person. These IMSI codes have a high correlation with the target person, providing a data foundation for further analysis of the target person's behavior.

[0041] For example, a preset voting mechanism can be used to filter out the first set of IMSI codes associated with the target person from the candidate IMSI codes. The specific steps are as follows: Whenever a target surveillance image containing the target person is selected, IMSI codes that meet the preset conditions are selected from the preset IMSI database as candidate IMSI codes, and the cumulative count of each IMSI code is incremented by 1; the initial value of the cumulative count is 0.

[0042] After multiple screenings, multiple candidate IMSI codes can be obtained. At this point, the candidate IMSI codes can be sorted according to the cumulative number of times each candidate IMSI code meets the preset conditions. The candidate IMSI codes with the highest cumulative number of times (e.g., the top 3) are determined as the first set of IMSI codes associated with the target person.

[0043] S204. Determine the population anomaly index of the target area based on the first IMSI code set and the second IMSI code set, wherein the second IMSI code set includes the IMSI codes collected in the target area, and the population anomaly index is used to characterize the possibility of abnormal clustering behavior in the target area.

[0044] S205. If the abnormal population index in the target area exceeds the preset threshold, an alarm message will be generated.

[0045] In this embodiment of the invention, a set of monitoring images is obtained based on monitoring equipment. Target monitoring images containing target personnel are selected from the monitoring image set. Based on the first acquisition location and first acquisition time of the target monitoring images, candidate IMSI codes that meet preset conditions are selected from a preset IMSI database. Then, the first set of IMSI codes associated with the target personnel is determined based on the candidate IMSI codes. This achieves an effective combination of image feature information and IMSI feature information, providing a foundation for subsequent monitoring of the target personnel and their companions, and effectively improving the detection efficiency and accuracy of abnormal personnel or groups.

[0046] Example 3 Figure 3 This is a flowchart of an abnormal population detection method provided in Embodiment 3 of the present invention. This embodiment is a refinement based on the above embodiments. Figure 3 As shown, the method includes: S301. Filter out target surveillance images containing the target personnel from the surveillance image set.

[0047] S302. Based on the first acquisition location and the first acquisition time of the target monitoring image, determine the set of first International Mobile Subscriber Identity (IMSI) codes associated with the target personnel.

[0048] S303. Determine the density clustering index of the target area based on the number of IMSI codes in the second IMSI code set.

[0049] In this embodiment, the target area may include multiple sub-areas. For example, an urban public safety monitoring system can be deployed in multiple important urban locations (e.g., schools, shopping malls, etc., each location corresponding to a sub-area) to monitor and identify potential abnormal crowds. Each sub-area has a fixed area. Density clustering index can be used to represent the population density of the target area.

[0050] Optionally, based on the number of IMSI codes in the second IMSI code set, the density clustering index of the target region is determined, including the following A1 to A3: A1. Based on the third acquisition position of each IMSI code in the second IMSI code set, determine the number of IMSI codes acquired in each sub-region of the target area.

[0051] In this embodiment, the second IMSI code set can be understood as the set of IMSI codes collected by the preset mobile detection base station device in the target area. Each IMSI code carries the collection time information and location information. Therefore, the sub-area where each IMSI code is located can be determined according to the third collection location of each IMSI code, and then the number of IMSI codes collected in each sub-area can be obtained.

[0052] A2. For each sub-region, determine the density clustering index of the current sub-region based on the ratio of the number of IMSI codes collected in the current sub-region to the area of ​​the current sub-region.

[0053] In this embodiment, under normal circumstances, each person has a mobile phone, and one mobile phone number corresponds to one IMSI code. Therefore, the number of IMSI codes collected in the current sub-area can be determined as the number of people in the current sub-area. The ratio of the number of IMSI codes collected in the current sub-area to the area of ​​the current sub-area can be used as the density clustering index of the current sub-area, which can effectively reflect the density of the population in the current sub-area.

[0054] A3. The average density clustering index of all sub-regions in the target region is taken as the density clustering index of the target region.

[0055] In this embodiment, if the target region includes multiple sub-regions, the average density clustering index of all sub-regions can be used as the density clustering index of the target region; if the target region contains only one region, the density clustering index of that region can be calculated directly.

[0056] Optionally, the density clustering index of the target region is represented by the following expression: ; in, This represents the density aggregation index of the target area R within a preset time period T; This represents the total number of subregions in the target region R; Represents the sub-region number, with a value range of 100. ; Represents the sub-region within a preset duration T. The number of IMSI codes collected within; The area of ​​the sub-region.

[0057] For example, if the area in front of a school is 100 square meters, and the city's public safety monitoring system records that 50 people appeared in this area in the past 24 hours, then the density clustering index of this area is: ; The density clustering index for this area is 0.5, meaning that in the past 24-hour statistical period, an average of 0.5 people appeared per square meter in the area near the school gate. Through steps A1 to A3 above, the density clustering index of each sub-area of ​​the target area, as well as the average density clustering index of the target area, can be calculated separately. The advantage of this setup is that it allows for monitoring of both the population density of sub-areas and the population density of the target area. When the density clustering index of a certain area suddenly increases, it indicates abnormal population gathering behavior in that area, alerting relevant staff to focus on monitoring that area and improving monitoring efficiency.

[0058] S304. Determine the frequency clustering index of the target area based on the frequency at which each IMSI code in the second IMSI code set is collected.

[0059] In this embodiment, the frequency at which each IMSI code in the second IMSI code set is collected can be represented by the number of times each IMSI code appears in the target area within a preset time period. The frequency clustering index can be used to represent the frequency at which certain individuals or groups of people repeatedly appear within the target area within a preset time period, and can be used to detect abnormal repetitive clustering behavior.

[0060] Optionally, this step may specifically include: determining the frequency at which each IMSI code in the second IMSI code set is collected in the target area according to a preset indication function, and determining the frequency as the frequency aggregation index of the target area; wherein, the preset indication function is used to indicate whether the IMSI code in the second IMSI code set is collected in the target area at a specified time.

[0061] Optionally, taking a sub-region of the target region as an example, the frequency clustering index of this sub-region is represented by the following expression: ; in, Represents a sub-region within a preset duration T. The frequency clustering index; Represents a sub-region within a preset duration T. The number of IMSI codes collected internally; It represents a specific point in time within a preset duration T, i.e., a designated moment; This represents an indicator function used to indicate a point in time. sub-region Does it exist inside? When the time point sub-region Memory exists hour, When the time point sub-region It does not exist inside. hour, ; This represents the i-th IMSI code in the second IMSI code set.

[0062] For example, if there are three unusual individuals who frequently appeared at the school gate in the past 24 hours, appearing 10, 9, and 8 times respectively, then the frequency clustering index at the school gate in the past 24 hours is: ; This means the frequency clustering index for this area is 0.9, indicating that on average, each anomalous individual appeared in the school gate area 9 times within the past 24-hour statistical period. The advantage of this setup is that by using the frequency with which each IMSI code in the second IMSI code set is collected in the target area, the frequency clustering index of the target area can be determined. This facilitates the screening of individuals who frequently appear in the target area within a preset time period, helping to identify potential anomalous individuals. It also allows for focused monitoring of target individuals to determine their frequency of appearance in the target area, thereby improving the efficiency of monitoring anomalous individuals.

[0063] S305. Determine the specific IMSI clustering index of the target area based on the intersection of the first IMSI code set and the second IMSI code set.

[0064] In this embodiment, the intersection of the first IMSI code set and the second IMSI code set can represent the number of people in the target area along with the target person and their companions; a specific IMSI clustering index can be used to represent the number of IMSIs in the target area that the first IMSI code set (which can represent the number of people in the target area along with their companions) appears in, and can be used to measure the proportion of abnormal people in the target area.

[0065] Optionally, this step may specifically include: determining a specific IMSI aggregation index for the target area based on the ratio of the number of IMSI codes in the intersection of the first IMSI code set and the second IMSI code set to the number of people detected appearing in the target area.

[0066] In this embodiment, since the second IMSI code set is the set of IMSI codes collected by the mobile detection base station equipment in the target area, and generally each person carries a mobile phone, the number of IMSI codes in the second IMSI code set can be used as the number of people detected in the target area. Alternatively, the number of people detected in the target area can also be determined by monitoring equipment (e.g., cameras) deployed in the target area.

[0067] Optionally, the specific IMSI clustering index for the target area is represented by the following expression: ; in, The specific IMSI clustering index represents the target area R within a preset time period T; The second IMSI code set representing the target area within a preset time period T; Represents the first IMSI code set; This represents the number of people detected in the target area within a preset time period T.

[0068] For example, the first IMSI code set is the IMSI code set of the defined target population, including 30 IMSI codes. In the past 24 hours, 20 IMSI codes were detected appearing at the school gate, and the number of people detected appearing at the school gate in the past 24 hours was 50. Then the specific IMSI clustering index at the school gate is: ; The advantage of this setup is that it enables dynamic detection of the target population. When a specific IMSI clustering index increases, it indicates that the target population is exhibiting abnormal clustering behavior, which can further improve the monitoring efficiency of abnormal populations.

[0069] S306. Determine the population anomaly index of the target area based on the density clustering index, frequency clustering index, specific IMSI clustering index of the target area, and preset weighting coefficients.

[0070] In this embodiment, the density clustering index, frequency clustering index, and specific IMSI clustering index of the target area are each assigned a different preset weight coefficient, reflecting their importance. Each preset weight coefficient can be set according to the application scenario. For example, if the target area is a transportation hub, where crowd gathering is the primary monitoring target, the preset weight coefficient corresponding to the density clustering index can be increased. If the target area is a commercial area, focusing on the frequency of the same person entering and exiting a certain area, the preset weight coefficient corresponding to the frequency clustering index can be increased. If the target area is a specific control area, focusing on the activities of specific personnel, the preset weight coefficient corresponding to the specific IMSI clustering index can be increased. The initial values ​​of each preset weight coefficient can be equal, and the specific values ​​are not limited. The advantage of this setting is that it allows for dynamic adjustment of the importance of different indicators, enabling multi-level monitoring of the target area.

[0071] Optionally, the anomaly index of the target area's population is represented by the following expression: ; in, This represents the abnormal population index in the target area within a preset time period T. The preset weighting coefficients representing the density clustering index of the target area. The preset weighting coefficients representing the frequency clustering index of the target region. Preset weighting coefficients for specific IMSI clustering indices representing the target region.

[0072] S307. If the abnormal population index in the target area exceeds the preset threshold, an alarm message will be generated.

[0073] Optionally, if the abnormal population index in the target area exceeds the preset abnormal population index threshold, a first-level alarm message will be generated.

[0074] In this embodiment, if the crowd anomaly index in the target area exceeds a preset crowd anomaly index threshold, it indicates that a crowd gathering has occurred in the target area, and the target person and their companions constitute a high proportion of the gathered crowd. In this case, a first-level alarm (which can be considered the highest level alarm) needs to be triggered to prompt staff to take appropriate safety precautions. The preset crowd anomaly index threshold can be set based on empirical values ​​or pre-calibrated based on historical data. For example, the calibration process can be as follows: collect historical data for a certain period in the target area, calculate the crowd anomaly index corresponding to that period, and take its high quantile (e.g., the 95th percentile) as the initial threshold; then, substitute historical normal scenarios and historical abnormal scenarios into the verification, perform regression optimization based on the detection rate and false alarm rate, and determine the final preset crowd anomaly index threshold. Alternatively, the calibration process can be as follows: calculate the corresponding crowd anomaly index in different historical crowd gathering scenarios multiple times, and use the average or median of multiple crowd anomaly indices as the preset crowd anomaly index threshold.

[0075] The method further includes the following B1 to B2: B1. If the density clustering index of the target area does not exceed the preset density clustering index threshold, the frequency clustering index of the target area does not exceed the preset frequency clustering index threshold, and the specific IMSI clustering index of the target area exceeds the preset IMSI clustering index threshold, then a second-level alarm message is generated. The preset density clustering index threshold, preset frequency clustering index threshold, and preset IMSI clustering index threshold can all be set based on empirical values ​​or pre-calibrated based on historical data to obtain specific values; the calibration process can adopt the aforementioned process for obtaining the preset abnormal population index threshold, which will not be elaborated further.

[0076] In this embodiment, if the density clustering index and the frequency clustering index of the target area do not exceed the corresponding set thresholds, but the specific IMSI clustering index of the target area exceeds the preset IMSI clustering index threshold, it indicates that the target person and their companions account for a high proportion of the crowd in the target area. At this time, a second-level alarm message (which can be considered as a medium-level alarm message) is triggered to prompt staff to pay close attention to the behavior of the target person and their companions.

[0077] B2. If the density clustering index of the target area exceeds the preset density clustering index threshold or the frequency clustering index of the target area exceeds the preset frequency clustering index threshold, and the specific IMSI clustering index of the target area does not exceed the preset IMSI clustering index threshold, then a third-level alarm message is generated.

[0078] In this embodiment, if the density clustering index of the target area exceeds the preset density clustering index threshold or the frequency clustering index of the target area exceeds the preset frequency clustering index threshold, but the specific IMSI clustering index of the target area does not exceed the preset IMSI clustering index threshold, it indicates that there may be personnel gathering behavior in the target area or the target personnel frequently appear in the target area, but the target personnel and their companions (the target personnel and their companions can be understood as abnormal personnel) have not yet gathered in the target area. At this time, a third-level alarm message (which can be considered a low-level alarm message) is triggered to prompt staff to strengthen the monitoring of the target area.

[0079] Based on the crowd gathering characteristics reflected by the above three indicators, different alarm thresholds are set to form three levels of alarm information, realizing multi-level and efficient dynamic alarms for the target area. This further improves the detection efficiency of abnormal personnel or groups and promptly prompts staff to take corresponding safety precautions or management measures, which is conducive to improving the safety of the target area.

[0080] Example 4 Figure 4 This is a schematic diagram of an abnormal population detection device provided in Embodiment 4 of the present invention. Figure 4 As shown, the device includes: a monitoring image filtering module 401, a first IMSI code set determination module 402, a crowd anomaly index determination module 403, and an alarm information generation module 404.

[0081] The surveillance image filtering module is used to filter out target surveillance images containing the target personnel from the surveillance image set; The first IMSI code set determination module is used to determine the first International Mobile Subscriber Identity (IMSI) code set associated with the target person based on the first acquisition location and the first acquisition time of the target surveillance image. The crowd anomaly index determination module is used to determine the crowd anomaly index of a target area based on a first IMSI code set and a second IMSI code set, wherein the second IMSI code set includes IMSI codes collected in the target area, and the crowd anomaly index is used to characterize the possibility of abnormal clustering behavior in the target area. The alarm information generation module is used to generate alarm information if the population anomaly index in the target area exceeds a preset threshold.

[0082] This invention provides an abnormal crowd detection device that can automatically and efficiently identify IMSI codes associated with specific individuals and monitor and locate suspected companions, thereby improving the detection efficiency of abnormal individuals and / or crowds. By filtering target surveillance images containing the target individuals from a set of surveillance images, the device achieves the identification of the target individuals. Based on the first acquisition location and first acquisition time of the target surveillance image, it determines the first set of International Mobile Subscriber Identity (IMSI) codes associated with the target individuals, enabling dynamic acquisition of the target individuals' IMSI codes and solving the problem of being unable to identify and locate individuals due to changes in their mobile phone numbers. By determining the crowd anomaly index of the target area based on the first IMSI code set and the second IMSI code set, it achieves efficient monitoring of whether abnormal gathering behavior exists in the target area. The second IMSI code set includes IMSI codes acquired within the target area, and the crowd anomaly index characterizes the probability of abnormal gathering behavior in the target area. Furthermore, when the crowd anomaly index of the target area exceeds a preset threshold, it can generate alarm information, achieving real-time detection of abnormal crowds in the target area.

[0083] Optionally, the set of monitoring images is acquired by a preset monitoring device, and a preset mobile detection base station device is set within a preset location range of the preset monitoring device.

[0084] Optionally, the first IMSI code set determination module includes: A candidate IMSI code filtering unit is used to filter candidate IMSI codes that meet preset conditions from a preset IMSI database based on the first acquisition location and the first acquisition time of the target monitoring image; wherein, the preset IMSI database includes the acquired IMSI codes reported by the preset mobile detection base station equipment; the preset conditions include that the deviation between the second acquisition location and the first acquisition location of the acquired IMSI code is within a preset location deviation range, and / or that the deviation between the second acquisition time and the first acquisition time of the acquired IMSI code is within a preset time deviation range; The first IMSI code set determination unit is used to determine the first IMSI code set associated with the target person based on the candidate IMSI codes.

[0085] Optionally, the first IMSI code set determination unit includes: The cumulative count determination subunit is used to determine the cumulative count of each candidate IMSI code that meets the preset conditions; The associated IMSI code determination subunit is used to determine the first set of IMSI codes associated with the target person based on the candidate IMSI codes with the highest cumulative count and the previous preset number.

[0086] Optional, the population anomaly index determination module includes: The density clustering index determination unit is used to determine the density clustering index of the target area based on the number of IMSI codes in the second IMSI code set. The frequency clustering index determination unit is used to determine the frequency clustering index of the target area based on the frequency at which each IMSI code in the second IMSI code set is collected. A specific IMSI clustering index determination unit is used to determine the specific IMSI clustering index of a target area based on the intersection of the first IMSI code set and the second IMSI code set. The index fusion unit is used to determine the population anomaly index of the target area based on the density clustering index of the target area, the frequency clustering index of the target area, the specific IMSI clustering index of the target area, and a preset weighting coefficient.

[0087] Optionally, the target region includes multiple sub-regions; the density clustering index determination unit is specifically used for: Based on the third acquisition position of each IMSI code in the second IMSI code set, determine the number of IMSI codes acquired in each sub-region of the target region; For each sub-region, the density clustering index of the current sub-region is determined based on the ratio of the number of IMSI codes collected in the current sub-region to the area of ​​the current sub-region. The average density clustering index of all sub-regions in the target region is taken as the density clustering index of the target region.

[0088] Optionally, a frequency aggregation index determination unit is used specifically for: The frequency at which each IMSI code in the second IMSI code set is collected in the target area is determined according to a preset indication function, and the frequency is determined as the frequency aggregation index of the target area; wherein, the preset indication function is used to indicate whether the IMSI code in the second IMSI code set is collected in the target area at a specified time.

[0089] Optionally, the specific IMSI clustering index determination unit is used for: The specific IMSI clustering index of the target area is determined by the ratio of the number of IMSI codes in the intersection of the first IMSI code set and the second IMSI code set to the number of people detected in the target area.

[0090] Optionally, the alarm information generation module is specifically used for: If the abnormal population index in the target area exceeds the preset abnormal population index threshold, a first-level alarm message will be generated. The device further includes: The second alarm information generation module is used to generate a second-level alarm information if the density clustering index of the target area does not exceed a preset density clustering index threshold, the frequency clustering index of the target area does not exceed a preset frequency clustering index threshold, and the specific IMSI clustering index of the target area exceeds a preset IMSI clustering index threshold. The third alarm information generation module is used to generate a third-level alarm information if the density clustering index of the target area exceeds a preset density clustering index threshold or the frequency clustering index of the target area exceeds a preset frequency clustering index threshold, and the specific IMSI clustering index of the target area does not exceed a preset IMSI clustering index threshold.

[0091] The abnormal population detection device provided in the embodiments of the present invention can execute the abnormal population detection method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0092] Example 5 Figure 5 A schematic diagram of an electronic device 500 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 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.

[0093] like Figure 5As shown, the electronic device 500 includes at least one processor 501 and a memory, such as a read-only memory (ROM) 502 or a random access memory (RAM) 503, communicatively connected to the at least one processor 501. The memory stores computer programs executable by the at least one processor. The processor 501 can perform various appropriate actions and processes based on the computer program stored in the ROM 502 or loaded into the RAM 503 from storage unit 508. The RAM 503 can also store various programs and data required for the operation of the electronic device 500. The processor 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0094] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, such as keyboard, mouse, etc.; output unit 507, such as various types of monitors, speakers, etc.; storage unit 508, such as disk, optical disk, etc.; and communication unit 509, such as network card, modem, wireless transceiver, etc. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0095] Processor 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 501 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 501 performs the various methods and processes described above, such as abnormal crowd detection methods.

[0096] In some embodiments, the abnormal crowd detection method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by processor 501, one or more steps of the abnormal crowd detection method described above may be performed. Alternatively, in other embodiments, processor 501 may be configured to perform the abnormal crowd detection method by any other suitable means (e.g., by means of firmware).

[0097] 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.

[0098] 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.

[0099] 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.

[0100] 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).

[0101] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or 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.

[0102] 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.

[0103] This disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the abnormal population detection method provided in the above embodiments.

[0104] 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.

[0105] 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 detecting abnormal populations, characterized in that, include: Filter the surveillance images from the surveillance image set to find target surveillance images containing the target person; Based on the first acquisition location and the first acquisition time of the target surveillance image, determine the first International Mobile Subscriber Identity (IMSI) code set associated with the target person; Based on the first IMSI code set and the second IMSI code set, a population anomaly index for the target area is determined. The second IMSI code set includes IMSI codes collected within the target area. The population anomaly index is used to characterize the possibility of abnormal clustering behavior in the target area. If the abnormal population index in the target area exceeds a preset threshold, an alarm message will be generated.

2. The abnormal population detection method according to claim 1, characterized in that, The monitoring image set is acquired by a preset monitoring device, and a preset mobile detection base station device is set within a preset location range of the preset monitoring device; The step of determining the first International Mobile Subscriber Identity (IMSI) code set associated with the target person based on the first acquisition location and the first acquisition time of the target surveillance image includes: Based on the first acquisition location and the first acquisition time of the target monitoring image, candidate IMSI codes that meet preset conditions are filtered out from the preset IMSI database; wherein, the preset IMSI database includes the acquired IMSI codes reported by the preset mobile code detection base station equipment; the preset conditions include that the deviation between the second acquisition location and the first acquisition location of the acquired IMSI code is within a preset location deviation range, and / or, the deviation between the second acquisition time and the first acquisition time of the acquired IMSI code is within a preset time deviation range; The first set of IMSI codes associated with the target person is determined based on the candidate IMSI codes.

3. The method according to claim 2, characterized in that, The step of determining the first set of IMSI codes associated with the target person based on the candidate IMSI codes includes: Determine the cumulative number of times each candidate IMSI code meets the preset conditions; The first set of IMSI codes associated with the target person is determined based on the candidate IMSI codes with the highest cumulative count and a pre-set number of counts.

4. The abnormal population detection method according to claim 1, characterized in that, Based on the first IMSI code set and the second IMSI code set, the anomaly index of the population in the target area is determined, including: The density clustering index of the target region is determined based on the number of IMSI codes in the second IMSI code set. The frequency clustering index of the target area is determined based on the frequency at which each IMSI code in the second IMSI code set is collected. Based on the intersection of the first IMSI code set and the second IMSI code set, a specific IMSI clustering index for the target area is determined; The population anomaly index of the target area is determined based on the density clustering index of the target area, the frequency clustering index of the target area, the specific IMSI clustering index of the target area, and a preset weighting coefficient.

5. The abnormal population detection method according to claim 4, characterized in that, The target area includes multiple sub-regions; The step of determining the density clustering index of the target region based on the number of IMSI codes in the second IMSI code set includes: Based on the third acquisition position of each IMSI code in the second IMSI code set, determine the number of IMSI codes acquired in each sub-region of the target region; For each sub-region, the density clustering index of the current sub-region is determined based on the ratio of the number of IMSI codes collected in the current sub-region to the area of ​​the current sub-region. The average density clustering index of all sub-regions in the target region is taken as the density clustering index of the target region. The step of determining the frequency clustering index of the target area based on the frequency at which each IMSI code in the second IMSI code set is collected includes: The frequency at which each IMSI code in the second IMSI code set is collected in the target area is determined according to a preset indication function, and the frequency is determined as the frequency aggregation index of the target area; wherein, the preset indication function is used to indicate whether the IMSI code in the second IMSI code set is collected in the target area at a specified time; The step of determining a specific IMSI clustering index for a target region based on the intersection of the first IMSI code set and the second IMSI code set includes: The specific IMSI clustering index of the target area is determined by the ratio of the number of IMSI codes in the intersection of the first IMSI code set and the second IMSI code set to the number of people detected in the target area.

6. The abnormal population detection method according to claim 5, characterized in that, If the abnormal population index in the target area exceeds a preset threshold, an alarm message is generated, including: If the abnormal population index in the target area exceeds the preset abnormal population index threshold, a first-level alarm message will be generated. The method further includes: If the density clustering index of the target area does not exceed the preset density clustering index threshold, the frequency clustering index of the target area does not exceed the preset frequency clustering index threshold, and the specific IMSI clustering index of the target area exceeds the preset IMSI clustering index threshold, then a second-level alarm message is generated. If the density clustering index of the target area exceeds a preset density clustering index threshold or the frequency clustering index of the target area exceeds a preset frequency clustering index threshold, and the specific IMSI clustering index of the target area does not exceed a preset IMSI clustering index threshold, then a third-level alarm message is generated.

7. An abnormal population detection device, characterized in that, include: The surveillance image filtering module is used to filter out target surveillance images containing the target personnel from the surveillance image set; The first IMSI code set determination module is used to determine the first International Mobile Subscriber Identity (IMSI) code set associated with the target person based on the first acquisition location and the first acquisition time of the target surveillance image. The crowd anomaly index determination module is used to determine the crowd anomaly index of a target area based on a first IMSI code set and a second IMSI code set, wherein the second IMSI code set includes IMSI codes collected in the target area, and the crowd anomaly index is used to characterize the possibility of abnormal clustering behavior in the target area. The alarm information generation module is used to generate alarm information if the population anomaly index in the target area exceeds a preset threshold.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the abnormal population detection method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the abnormal population detection method according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the abnormal population detection method according to any one of claims 1-6.

Images