Intelligent monitoring analysis method and device, electronic equipment and storage medium

Abstract

The application relates to a kind of monitoring intelligent analysis method, device, electronic equipment and storage medium.The monitoring video collected by the monitoring system arranged in the target area is managed, and the method comprises the following steps: obtaining the retrieval condition of the target to be queried;According to the retrieval condition, the monitoring video in the monitoring system is selected to obtain the video to be determined;Based on the video to be determined, all the target to be determined meeting the retrieval condition is selected and the picture of the target to be determined is obtained;The picture of the target to be queried is determined from the picture of the target to be determined;According to the picture of the target to be queried, the video segment of the target to be queried in the video to be determined is determined;According to the video segment, the behavior characteristics of the target to be queried are analyzed;The distribution of the monitoring camera in the monitoring system is combined with the behavior characteristics to determine all the segments of the target to be queried in the monitoring video meeting the retrieval condition;All the segments are spliced according to time sequence to generate target video;Based on the target video, the route map of the action track of the target to be queried is drawn.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of video processing, in particular to a monitoring intelligent analysis method and device, electronic equipment and storage medium. BACKGROUND

[0002] With the continuous development of modernization, urban construction is also more and more complete, among them, a large number of monitoring systems are set up in the city, which can effectively monitor the road vehicle flow and shoot illegal driving.

[0003] However, since each monitoring system is relatively independent, when calling multiple monitoring systems to jointly investigate the action track of a target, a large amount of manpower and multiple screens are needed to simultaneously call video data in multiple monitoring systems, and then the video data is watched bit by bit to determine the video segment where the target is located, and finally the action track of the target is drawn by manual induction. The whole investigation process needs to waste a lot of time and labor cost. SUMMARY

[0004] The present application provides a monitoring intelligent analysis method, device, electronic equipment and storage medium. In order to solve the problem that a lot of manpower cost is wasted to read all monitoring videos when searching for a target from a large number of monitoring videos, and the problem that part of the action track of the target may be missed when drawing the action track of the target.

[0005] In a first aspect, the present application provides a monitoring intelligent analysis method for managing the monitoring video collected by the monitoring system arranged in the target area, the method comprising:

[0006] Obtaining the retrieval condition of the target to be queried; the retrieval condition includes: time condition, place condition, feature condition, retrieval time span;

[0007] According to the time condition and the place condition, the monitoring video in the monitoring system is selected to obtain the to-be-determined video;

[0008] Based on the to-be-determined video, all to-be-determined targets meeting the feature condition are selected, and the picture of the to-be-determined target is obtained;

[0009] The picture of the target to be queried is determined from the picture of the to-be-determined target;

[0010] According to the picture of the target to be queried, the video segment of the target to be queried in the to-be-determined video is determined;

[0011] According to the video segment, the behavior characteristics of the target to be queried are analyzed;

[0012] Determine all segments of the target video that meet the search time span of the target to be queried according to the behavior characteristics and the distribution of the monitoring camera in the monitoring system;

[0013] Splice all segments according to time sequence to generate a target video;

[0014] Based on the target video, draw a route map of the action track of the target to be queried, and display the route map and the target video.

[0015] Through the scheme, the search condition of the target to be queried is input by the user, the picture of the target to be determined that meets the search condition is generated, the picture of the target to be queried is determined from the picture of the target to be determined, the video segment of the target to be queried in the monitoring is confirmed through the picture of the target to be queried, the search range of the target to be queried is narrowed step by step by using the search condition, the action route of the target is further analyzed by using the video segment, the segment of the target in other videos is continuously searched according to the action route, the complete monitoring video of the target is obtained, finally, the route map of the action track of the target to be queried is generated according to the monitoring video, the route map of the action track of the target to be queried is automatically obtained through the search condition, the time for manually watching the video to find the target is reduced, and the efficiency of drawing the route map of the action track of the target is improved.

[0016] Optionally, the method further comprises creating a monitoring camera information table;

[0017] The monitoring location information of the monitoring camera and the storage path of the generated monitoring video are stored in the monitoring camera information table as monitoring camera information;

[0018] A monitoring camera database is created for each monitoring camera;

[0019] The monitoring video is segmented and read according to a preset time period, and the feature information of each monitoring target appearing in each segment of the monitoring video is analyzed;

[0020] According to the preset time period and the feature information, a monitoring video table is generated in the monitoring camera database.

[0021] Through the scheme, a database is set up for each monitoring camera, avoiding that the information of all monitoring cameras is stored in one library or one table, and then a table is individually established for each video, and all information of one video is stored in each table, so that the data can be stored in a hierarchical manner, the data structure is more reasonable, and the data security is improved when the table or library is damaged.

[0022] Optionally, the selecting the monitoring video in the monitoring system according to the search condition to obtain the to-be-determined video comprises:

[0023] selecting the monitoring camera information meeting the location condition from the monitoring camera information table;

[0024] accessing the storage location of the monitoring video corresponding to the monitoring camera according to the storage path in the monitoring camera information;

[0025] selecting the segment of the monitoring video containing the time according to the time indicated by the time condition as the to-be-determined video;

[0026] the selecting all to-be-determined targets meeting the search condition according to the to-be-determined video and obtaining the picture of the to-be-determined target comprises:

[0027] connecting the monitoring camera database corresponding to the monitoring camera information according to the monitoring camera information;

[0028] selecting the monitoring video table corresponding to the monitoring camera database according to the to-be-determined video;

[0029] matching the feature information meeting the feature condition from the monitoring video table according to the feature condition, wherein the monitoring target corresponding to each piece of feature information is the to-be-determined target;

[0030] for each to-be-determined target, intercepting the video segment of the to-be-determined video in which the to-be-determined target appears;

[0031] for each video segment, extracting a frame of picture containing the feature information and having the clearest quality as the picture of the to-be-determined target.

[0032] According to the scheme, the location condition is selected from the monitoring camera information table to obtain the monitoring camera that can monitor the target location, then the time condition is used to select the monitoring video of the target time from the database corresponding to the monitoring camera, and finally the feature information is used to determine the target meeting the feature in the monitoring video, so that the full search of the search information is not required during the search, the search pressure on the system is reduced by using the hierarchical search manner, the search time is shortened, and the search accuracy is improved.

[0033] Optionally, the analyzing the feature information of each monitoring target appearing in each monitoring video comprises:

[0034] extracting a plurality of key frames from the monitoring video;

[0035] comparing all the key frames to establish a static environment model;

[0036] Based on the static environment model, all key frames are analyzed to determine whether the monitoring target appears in the key frames;

[0037] If the monitoring target appears, the monitoring target is further analyzed;

[0038] The type of the monitoring target is determined, and if the monitoring target is a human, the body information, face information, and clothing information of the monitoring target are analyzed, and if the monitoring target is a vehicle, the license plate information, vehicle type, and vehicle color of the monitoring target are analyzed.

[0039] By comparing the key frames in the monitoring video to obtain the static environment model, and then using the static environment model to analyze the key frames in reverse, it can be quickly determined whether all targets in the monitoring video appear, and then the characteristics of the targets are analyzed, thereby improving the analysis efficiency of the characteristics of the targets in the monitoring video.

[0040] Optionally, the retrieval condition of the target to be queried is obtained, including:

[0041] A sample picture of the target to be queried is obtained, and feature information in the sample picture is obtained;

[0042] Part of the feature information is further selected as the retrieval condition of the target to be queried.

[0043] By this scheme, the features of the picture are analyzed, and the obtained features are used as the retrieval condition for retrieval, thereby providing the user with more input modes of the retrieval condition and improving the user experience.

[0044] Optionally, the behavior characteristics include: walking speed, entering time, leaving time, entering direction, and leaving direction.

[0045] The behavior characteristics of the target to be queried are analyzed according to the video segment, including:

[0046] The moving distance and the moving time of the target to be queried are obtained from the video segment, and the walking speed of the target to be queried is further calculated;

[0047] The time when the target to be queried first appears in the monitoring area is obtained as the entering time from the video segment, the time when the target to be queried last leaves the monitoring area is obtained as the leaving time, the moving direction when the target to be queried first appears in the monitoring area is obtained as the entering direction, and the moving direction when the target to be queried last leaves the monitoring area is obtained as the leaving direction.

[0048] The behavior characteristics, the distribution of the monitoring cameras in the monitoring system, and all segments of the target to be queried in the monitoring video that meet the retrieval time span are determined, including:

[0049] acquire the road segment information monitored by each monitoring camera in the monitoring system and the adjacent situation of each monitoring camera;

[0050] determine the previous monitoring camera according to the entering direction, the traveling speed and the adjacent situation, and determine the next monitoring camera according to the leaving direction, the traveling speed and the adjacent situation;

[0051] obtain the video segment of the monitoring video of the previous monitoring camera of the target to be queried according to the road segment information monitored by the video segment, the road segment information monitored by the previous monitoring camera, the traveling speed and the entering time, and obtain the video segment of the monitoring video of the next monitoring camera of the target to be queried according to the road segment information monitored by the video segment, the road segment information monitored by the next monitoring camera, the traveling speed and the leaving time;

[0052] circulate the above steps to continuously acquire new video segments, and stop the circulation if the target to be queried leaves the monitoring area of the monitoring system, or the monitoring time of the new video segment has reached the current time, or the target to be queried stops moving, or the monitoring time of the new video segment has exceeded the search time span; after the circulation is stopped, all the video segments acquired are all the segments of the target to be queried in the monitoring video.

[0053] By the scheme, the traveling speed, the traveling direction, the entering time, the leaving time and other behavior characteristics of the target to be queried in the monitoring video are analyzed, the distribution of the monitoring cameras and the road segment information monitored by each monitoring camera are combined, and the monitoring video of the target to be queried shot by the adjacent cameras is obtained in a chain recursive manner according to a piece of monitoring video of the target to be queried, so that the efficiency of searching the video segment of the target to be queried in the monitoring video is improved.

[0054] Optionally, the drawing of the route map of the action track of the target to be queried based on the target video comprises:

[0055] acquire each road segment information monitored in the target video;

[0056] mark each road segment with a solid line in the map according to each road segment information;

[0057] connect the adjacent road segments marked with a solid line in the map with a dashed line according to the traveling direction of the target to be queried in the target video, and draw a direction mark.

[0058] By the scheme, the road segment information of the to-be-queried target in each monitoring video is utilized to mark the moving routes of the to-be-queried target on a map, and the moving routes are connected according to the moving direction of the to-be-queried target, so as to complete the drawing of the action track graph of the to-be-queried target.

[0059] In a second aspect, the application provides a monitoring intelligent analysis device. The device is used for managing monitoring videos collected by a monitoring system arranged in a target area, and includes:

[0060] The acquisition module is configured to acquire a search condition of a to-be-queried target. The search condition includes a time condition, a location condition, a feature condition, and a search time span.

[0061] The first search module is configured to select monitoring videos in the monitoring system according to the time condition and the location condition, to obtain to-be-determined videos, to select all to-be-determined targets meeting the feature condition based on the to-be-determined videos, and to obtain pictures of the to-be-determined targets.

[0062] The reverse search module is configured to determine pictures of the to-be-queried target from the pictures of the to-be-determined targets, and to determine video segments of the to-be-queried target in the to-be-determined videos according to the pictures of the to-be-queried target.

[0063] The second search module is configured to analyze behavior characteristics of the to-be-queried target according to the video segments, and to determine all segments of the to-be-queried target meeting the search time span in the monitoring videos in combination with the behavior characteristics and the distribution of monitoring cameras in the monitoring system.

[0064] The drawing module is configured to splice the all segments according to time sequences to generate a target video, to draw a route graph of an action track of the to-be-queried target based on the target video, and to call a display module to display the route graph and the target video.

[0065] Optionally, the monitoring intelligent analysis device further includes a database table module.

[0066] The database table module is configured to create a monitoring camera information table.

[0067] The monitoring location information of the monitoring camera and the storage path of the generated monitoring video are stored in the monitoring camera information table as monitoring camera information.

[0068] A monitoring camera database is created for each monitoring camera.

[0069] The monitoring videos are segmented and read according to a preset time period, and the feature information of each monitoring target appearing in each monitoring video is analyzed.

[0070] According to the preset time period, the feature information, a monitoring video table is generated in the monitoring camera database.

[0071] Optionally, the first retrieval module selects the monitoring video in the monitoring system according to the retrieval condition to obtain the pending video, and specifically is used for:

[0072] selecting the monitoring camera information in the monitoring camera information table that meets the location condition;

[0073] According to the storage path in the monitoring camera information, the storage position of the monitoring video generated by the corresponding monitoring camera is entered;

[0074] Based on the time represented by the time condition, the segment containing the time in the monitoring video monitoring time is selected as the pending video;

[0075] When the first retrieval module selects all the pending targets meeting the retrieval condition based on the pending video and obtains the picture of the pending target, specifically is used for:

[0076] According to the monitoring camera information, the corresponding monitoring camera database is connected;

[0077] According to the pending video, the corresponding monitoring video table in the monitoring camera database is selected;

[0078] According to the feature condition, the feature information meeting the feature condition is matched from the monitoring video table, and the monitoring target corresponding to each feature information is the pending target;

[0079] For each of the pending targets, the video segment in which the pending target appears in the pending video is intercepted;

[0080] For each of the video segments, a frame of picture containing the feature information and the clearest picture is extracted, and the picture is taken as the picture of the pending target.

[0081] Optionally, the monitoring intelligent analysis device further comprises a video analysis module;

[0082] When the video analysis module analyzes the feature information of each monitoring target appearing in each monitoring video, specifically is used for:

[0083] The monitoring video is key frame extracted, and a plurality of key frames are obtained;

[0084] All the key frames are compared to establish a static environment model;

[0085] Based on the static environment model, all the key frames are analyzed to determine whether the monitoring target appears in the key frames.

[0086] If the monitoring target appears, further analysis is performed on the monitoring target;

[0087] If the monitoring target is a human, the body information, face information and clothing information of the monitoring target are analyzed; if the monitoring target is a vehicle, the license plate information, vehicle type and vehicle color of the monitoring target are analyzed.

[0088] Optionally, when the first retrieval module acquires the retrieval condition of the target to be queried, the first retrieval module is specifically configured to:

[0089] Acquire a sample picture of the target to be queried, and acquire feature information in the sample picture;

[0090] Further select part of the feature information as the retrieval condition of the target to be queried from the feature information.

[0091] Optionally, the behavior characteristics include a walking speed, an entering time, a leaving time, an entering direction and a leaving direction.

[0092] When the second retrieval module analyzes the behavior characteristics of the target to be queried according to the video clip, the second retrieval module is specifically configured to: acquire a moving distance and a moving time length of the target to be queried through the video clip, and further calculate a walking speed of the target to be queried.

[0093] Acquire, through the video clip, a time when the target to be queried first appears in the monitoring area as the entering time, a time when the target to be queried last leaves the monitoring area as the leaving time, a moving direction when the target to be queried first appears in the monitoring area as the entering direction, and a moving direction when the target to be queried last leaves the monitoring area as the leaving direction.

[0094] When the second retrieval module determines all clips of the target to be queried in the monitoring video that meet the retrieval time span in combination with the behavior characteristics and the distribution of monitoring cameras in the monitoring system, the second retrieval module is specifically configured to: acquire road segment information monitored by each monitoring camera in the monitoring system and adjacent conditions of each monitoring camera.

[0095] Determine a previous monitoring camera according to the entering direction, the walking speed and the adjacent conditions, and determine a next monitoring camera according to the leaving direction, the walking speed and the adjacent conditions.

[0096] According to the road section information monitored by the video clip, the road section information monitored by the last monitoring camera, the travel speed and the entering time, a video clip of the monitoring video of the last monitoring camera is obtained for the target to be queried; according to the road section information monitored by the video clip, the road section information monitored by the next monitoring camera, the travel speed and the leaving time, a video clip of the monitoring video of the next monitoring camera is obtained for the target to be queried;

[0097] The above steps are cycled to continuously obtain new video clips, and the cycle is stopped if the target to be queried leaves the monitoring area of the monitoring system, or the monitoring time of the new video clip reaches the current time, or the target to be queried stops moving, or the monitoring time of the new video clip exceeds the search time span; after the cycle is stopped, all the obtained video clips are all the clips of the target to be queried in the monitoring video.

[0098] Optionally, when the drawing module draws the route map of the action track of the target to be queried based on the target video, the drawing module is specifically configured to:

[0099] Obtain each road section information monitored in the target video;

[0100] According to each road section information, each road section is marked by a solid line in a map;

[0101] According to the travel direction of the target to be queried in the target video, adjacent road sections marked by a solid line in the map are connected by a dashed line, and a direction mark is drawn.

[0102] In a third aspect, the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program capable of being loaded and executed by the processor to execute the method of the first aspect.

[0103] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor to execute the method of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0104] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0105] Figure 1 An application scenario schematic diagram provided by an embodiment of the present application;

[0106] Figure 2 A flow chart of a monitoring intelligent analysis method provided by an embodiment of the present application is shown in FIG. 1.

[0107] Figure 3 A structural schematic diagram of a monitoring intelligent analysis device provided by an embodiment of the present application is shown in FIG. 2.

[0108] Figure 4 A structural schematic diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 3. DETAILED DESCRIPTION

[0109] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in detail with reference to the accompanying drawings. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0110] In addition, the term "and / or" in the present document is only used to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " in the present document generally represents an "or" relationship between the associated objects unless otherwise specified.

[0111] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0112] With the continuous development of modernization, a large number of monitoring systems are set up in cities. However, each monitoring video contains a large amount of target information, and each monitoring system contains a large amount of monitoring videos. When a monitoring system is called to investigate the action track of a target, a large amount of manpower is needed to read all the monitoring videos, so as to select all the monitoring videos of the target. This investigation process wastes a lot of time and manpower cost, and the action track drawn may not be very accurate due to the possibility of missing some monitoring videos in the process of manual reading.

[0113] Therefore, the present application provides a monitoring intelligent analysis method and device, an electronic device and a storage medium. By inputting the search condition of the target to be queried, the segment of the monitoring video in the monitoring system containing the target to be queried can be quickly searched out, and the action track diagram of the target to be queried can be drawn.

[0114] Figure 1 An application scenario provided by the present application is shown in FIG. 4. Figure 1As shown, the method is specifically carried on a server in the form of software, a user accesses the server through a user equipment, and sends a retrieval condition to the server, the server searches the monitoring video collected by a monitoring system according to the retrieval condition, and returns a target video and a route map meeting the retrieval condition. The monitoring system is composed of a certain number of monitoring cameras. The specific implementation mode can refer to the following embodiments.

[0115] Figure 2 A flowchart of a monitoring intelligent analysis method provided for an embodiment of the present application, the method of the embodiment can be applied to the server in the above scenarios. As shown, Figure 2 The method includes:

[0116] Step S201, obtaining a retrieval condition of a target to be queried, the retrieval condition including a time condition, a place condition, a feature condition, and a retrieval time span.

[0117] The time condition is a certain fixed time or a time period, such as 12:00 or 12:00-12:30, the place condition is a street name or a landmark building name, the feature condition is feature information of the target to be queried, such as clothing information, height information, and name of a person describing a person or license plate information, vehicle model, and vehicle body color of a vehicle describing a vehicle, and the retrieval time span is a time period of video about the target to be queried, such as one day or two days.

[0118] The method is specifically integrated into a software, which provides a search page, and the user accesses the search page provided by the software to input the retrieval condition.

[0119] In some specific implementation modes, the search page includes a search input box and a condition check box, the search input box contains an identifier, and the identifier is used to prompt the user to input only the feature condition of the target to be queried in the input box, and if there are multiple sets of feature conditions, they are separated by spaces, for example, the user needs to search for a vehicle, and inputs "white brand a license plate b1" in the search input box.

[0120] In the embodiment, the license plate is specifically exemplified in order to reflect the difference between the license plates of vehicles, and is irrelevant to the actual vehicle information.

[0121] The condition check box is used for input of the time condition, the place condition, and the retrieval time span, for example, the user clicks the check box of the place condition, a map screen is popped up, and each street name is displayed on the map screen, and the street name is automatically input after being clicked and checked.

[0122] It should be noted that if the user inputs the name of the person to be queried, the name will be searched in the database, the data of the database is connected with the system of some relevant departments, the corresponding picture of the name is obtained, and the face information corresponding to the name is obtained, and the face information is used as the feature information for searching, and the place condition, time condition and search time span can also be input in the form of input box.

[0123] Step S202, according to the time condition and the place condition, the monitoring video in the monitoring system is selected to obtain the to-be-determined video, all to-be-determined targets meeting the feature condition are selected based on the to-be-determined video, and the picture of the to-be-determined target is obtained.

[0124] Among them, the to-be-determined video is a segment of the monitoring video meeting the time condition and the place condition, and the monitoring video refers to the monitoring video generated by the monitoring camera.

[0125] Because the user inputs the search condition with different details, the number of to-be-determined targets meeting the search condition will also be different.

[0126] In some specific implementations, the monitoring cameras in the monitoring system can be first screened by using the place condition in the search condition obtained by the user, and the monitoring cameras in the place represented by the place condition are selected, and then the monitoring videos shot by the selected monitoring cameras are selected by using the time condition in the search condition, and the videos in the monitoring videos containing the time represented by the time condition are selected as to-be-determined videos.

[0127] Then, the to-be-determined video is read, and the feature condition in the search condition is used to compare the features of each target appearing in the to-be-determined video, select the target meeting the feature condition, and correspondingly cut a clear picture of the target.

[0128] For example, the place condition is selected as “A road, B street”, the time condition is 12:00-12:01, and the to-be-determined video is the video shot by the monitoring camera at the intersection of A road and B street from 12:00 to 12:01.

[0129] The feature condition is “white brand a”, the to-be-determined video of the intersection of A road and B street from 12:00 to 12:01 is read, the features of each vehicle in the to-be-determined video are analyzed, and the following three targets meeting the feature condition are selected: “white brand a license plate b2”, “white brand a license plate b3”, “white brand a license plate b1”, and the corresponding pictures.

[0130] Step S203, determining the picture of the target to be queried from the picture of the to-be-determined target, and determining the video segment of the target to be queried in the to-be-determined video according to the picture of the target to be queried.

[0131] Specifically, after obtaining the images of the target objects, all images are returned to the search page for display, allowing the user to further select the target they wish to search for.

[0132] Alternatively, users can provide additional information to further filter the target. For example, if a user enters that the target to be searched is a hit-and-run vehicle, the system will determine whether any of the vehicles among the target targets are damaged and then automatically select the target to be searched.

[0133] Step S204: Based on the video clips, analyze the behavioral characteristics of the target to be queried, and combine the behavioral characteristics with the distribution of surveillance cameras in the monitoring system to determine all segments in the surveillance video that match the retrieval time span for the target to be queried.

[0134] Among them, behavioral characteristics represent features such as the direction and distance of movement of the target being queried. Distribution characteristics represent the adjacent directions and distances of each surveillance camera.

[0135] Specifically, the speed of the target can be calculated by reading video clips and calculating the movement time and distance of the target in the video clips, or by directly selecting the speed at which the target is stable for the longest time while moving as the target's speed.

[0136] Then, using the surveillance camera that captured the video clip as the center, find the adjacent surveillance cameras in the direction of travel. Then, based on the target's travel speed, entry time, and exit time, calculate the time when the target enters or leaves the monitoring area of ​​the adjacent surveillance cameras. Then, based on the target's travel speed and the length of the monitoring area of ​​the adjacent surveillance cameras, calculate the duration for which the target will appear at the adjacent cameras, and then extract the monitoring video from the adjacent surveillance cameras to obtain a new video clip.

[0137] Finally, repeat the above steps for the new video clips to recursively obtain new video clips, thereby obtaining all clips within the search time span.

[0138] Step S205: Segment all segments according to time order to generate the target video. Based on the target video, draw a route map of the movement trajectory of the target to be queried, and display the route map and the target video.

[0139] Each segment contains only the target data to be queried.

[0140] Specifically, the video is spliced ​​together based on the start and end times of each video segment to generate a target video. Based on the movement trajectory of the target in the target video, a route map of the target's movement trajectory is drawn on the corresponding map.

[0141] In some specific implementations, the target video is read, and the road surface walked by the target to be queried is color marked. When the target to be queried leaves a segment, a current picture of the road segment is generated by taking a screenshot of the current picture, and the picture ground color features are analyzed to generate a route segment on the corresponding road segment in the map. Finally, all the action route pictures are analyzed, and the action trajectory of the target to be queried in all the monitored road segments is generated as a route segment in the map. Then, the routes between the monitored road segments are connected by the direction of travel of the target to be queried in the video to generate a route map.

[0142] By the scheme, the search condition of the target to be queried is input by the user, and the picture of the target to be determined is generated according to the search condition. Then, the picture of the target to be queried is determined from the picture of the target to be determined. The picture of the target to be queried is confirmed in the video segment in the monitoring, the search range of the target to be queried is narrowed step by step by the search condition, the action route of the target is further analyzed by the video segment, the other segments of the target in the video are continuously searched according to the action route, and the complete monitoring video of the target is obtained. Finally, the route map of the action trajectory of the target to be queried is generated according to the monitoring video. The route map of the action trajectory of the target to be queried is automatically obtained by the search condition, the time for manually watching the video to find the target is reduced, and the efficiency of drawing the route map of the action trajectory of the target is improved.

[0143] In some embodiments, the above method further includes creating a monitoring camera information table, storing the monitoring location information of the monitoring camera, the storage path of the generated monitoring video, as the monitoring camera information into the monitoring camera information table, creating a monitoring camera database for each monitoring camera, segmenting the monitoring video according to a preset time period, analyzing the feature information of each monitoring target appearing in each segment of the monitoring video, and generating a monitoring video table in the monitoring camera database according to the preset time period and the feature information.

[0144] The preset time period is a non-fixed time period, which can be set to 30 minutes, 1 hour, 2 hours, etc.

[0145] The monitoring target is other target that does not belong to the inherent target in the monitoring location, such as vehicle, person, and other movable target, and the ground, tree, fence, etc. are inherent targets.

[0146] Specifically, all the monitoring camera information in the monitoring system is all entered into a table of a database, and an index is created for each monitoring camera information with a monitoring camera label. Each monitoring camera information contains the monitoring location information of the monitoring camera, and the storage path of the monitoring video generated by the monitoring camera. Since the number of monitoring cameras in each monitoring system is not necessarily the same, horizontal table splitting can be performed according to the number.

[0147] Then, a corresponding database is established for each monitoring camera, which is used to store only the data of the monitored target feature information of the corresponding camera. Then, the contents of the monitoring video of the camera for a preset time period are read, the features of all the monitoring targets are extracted, and the feature information in each read video is stored in a monitoring video table. The name of the monitoring video table is set as the time period monitored by the video. For example, the preset time period is 1 hour, the monitoring video is the monitoring video generated by the monitoring camera on February 13, and when reading, the video contents of 00:00-01:00, 01:00-02:00, …, 23:00-24:00 are read respectively, and 00:00-01:00, 01:00-02:00, …, 23:00-24:00 are used as the names of the monitoring video table.

[0148] According to the scheme, a database is established for each monitoring camera, which avoids that all the monitoring camera information is in one library or one table. Then, a table is established for each video, and each table stores all the information of one video. In this way, the data can be stored hierarchically, which can avoid full-text retrieval in subsequent queries, and also avoids that the database accesses one table or one library. When the table or library is damaged, the loss is minimized.

[0149] In some embodiments, according to the search condition, the monitoring video in the monitoring system is selected to obtain a to-be-determined video, including: selecting the monitoring camera information meeting the location condition in the monitoring camera information table, entering the storage location of the monitoring video generated by the corresponding monitoring camera according to the storage path in the monitoring camera information, and selecting the segment containing the time indicated by the time condition in the monitoring time of the monitoring video as the to-be-determined video based on the time condition. Correspondingly, the to-be-determined target meeting the search condition is selected based on the to-be-determined video, and the picture of the to-be-determined target is obtained, including: connecting the corresponding monitoring camera database according to the monitoring camera information, selecting the corresponding monitoring video table in the monitoring camera database according to the to-be-determined video, matching the feature information meeting the feature condition from the monitoring video table according to the feature condition, the monitoring target corresponding to each piece of feature information being the to-be-determined target, intercepting the video segment in which the to-be-determined target appears in the to-be-determined video for each to-be-determined target, extracting a frame of picture containing the feature information and having the clearest quality for each video segment, and taking the picture as the picture of the to-be-determined target.

[0150] Specifically, after obtaining the search condition, the location condition is searched first, the monitoring location field stored in the monitoring camera information table is matched for the location condition, the monitoring location information matching successfully is selected, and the information of the corresponding monitoring camera and the storage path of the generated monitoring video are obtained.

[0151] Secondly, the monitoring video in the storage path is screened by using the time condition, and the monitoring segment meeting the time point indicated by the time condition is selected, for example, the location condition is A road, and the time condition is February 10, 12 o'clock, then all the information of the monitoring location field in the monitoring camera information table including A road is selected, and there are three monitoring cameras at A road B street intersection, A road C street intersection, and A road D street intersection, then the monitoring video storage locations corresponding to the three monitoring cameras are entered, and the monitoring segments from 11:30 to 12:30 on February 10 are selected as to-be-determined videos.

[0152] The connection information of the monitoring camera database is also stored in the information of each monitoring camera in the monitoring camera information table, the database connection is performed based on the connection information in the information of the monitoring camera matching successfully, and then the monitoring video table meeting the time is selected according to the monitoring time of the to-be-determined video, for example, the monitoring time of the to-be-determined video is from 11:30 to 12:30, and the names of the monitoring video tables are from 11:00 to 12:00 and from 12:00 to 13:00, which are consistent with the monitoring time of the to-be-determined video, so the above two tables are selected.

[0153] After selecting the monitoring video table, the feature information in the monitoring video table is screened by using the feature condition, and all the feature information that meets the condition is selected out. Since each feature information describes the information of the same target, each feature information selected out corresponds to a monitoring target. Each feature information includes the time point at which the target appears and the time point at which the target disappears. According to the above time points, the video segment containing only the monitoring target is obtained by intercepting in the to-be-determined video. Then, the feature of the monitoring target contained in each frame of picture is analyzed, and the frame containing the most features of the monitoring target and the clearest picture is selected as the picture of the monitoring target.

[0154] For example, the feature condition is "180 male red shirt black shorts white sneakers", all feature information containing the feature condition is selected out. If there are two pieces of feature information "180 male red shirt black shorts white sneakers black hat 11:20-11:21" and "180 male red shirt black shorts white sneakers 12:20-12:21" that meet the condition, both of them are selected out, and the video segments corresponding to 11:20-11:21 and 12:20-12:21 in the to-be-determined video from 11:00 to 12:00 are intercepted.

[0155] By the scheme, the target place can be monitored by selecting the monitoring camera from the monitoring camera information table according to the place condition, the monitoring video of the target time is selected from the database corresponding to the monitoring camera according to the time condition, and finally the target meeting the feature is selected out from the monitoring video according to the feature information. Therefore, the search of all the search information is not needed in the search, the search pressure on the system is reduced by using the hierarchical search mode, the search time is shortened, and the search accuracy is improved.

[0156] In some embodiments, analyzing the feature information of each monitoring target appearing in the monitoring video includes: extracting key frames from the monitoring video to obtain a plurality of key frames, comparing all the key frames to establish a static environment model, and analyzing all the key frames based on the static environment model to determine whether the monitoring target appears in the key frame. If the monitoring target appears, the monitoring target is further analyzed to determine the type of the monitoring target. If the monitoring target is a human, the body information, face information and clothing information of the monitoring target are analyzed. If the monitoring target is a vehicle, the license plate information, vehicle type and vehicle color of the monitoring target are analyzed.

[0157] The key frame records all the information of the picture in the current time, and the static environment model is a model picture of the monitoring place without the monitoring target and containing only the inherent environmental features of the monitoring place.

[0158] Specifically, all key frame pictures in the monitoring video are extracted, and then the key frame pictures are compared one by one. When a static target appears in more than half of the key frame pictures, it can be determined that the static target is a fixed scene in the monitoring place. The static target is selected and added to the static environment model, and then a picture of the static environment model is generated.

[0159] It should be noted that when a static target appears in more than half of the key frame pictures, it can be determined that the static target is a fixed scene in the monitoring place.

[0160] Then, the picture of the static environment model is compared with the key frame again. If a target appears in the key frame that does not appear in the picture of the static environment model, the target is set as a monitoring target, and the monitoring target is subjected to preliminary image analysis.

[0161] If the monitoring target is analyzed to be a human, the picture recognition technology is used to further analyze the upper garment color information, upper garment type, lower garment color, lower garment type, shoe color, hat color, height information, and face information of the human. If the monitoring target is analyzed to be a vehicle, the picture recognition technology is used to further analyze the license plate information, vehicle color, and vehicle type of the vehicle.

[0162] Through the comparison of the key frame in the monitoring video, the static environment model is obtained, and then the key frame is analyzed in reverse using the static environment model, so that whether all targets in the monitoring video appear can be quickly obtained, and feature analysis of the targets is performed, thereby improving the analysis efficiency of the feature information of the targets in the monitoring video.

[0163] In some embodiments, a retrieval condition of a target to be queried is obtained, including: obtaining a sample picture of the target to be queried, obtaining feature information in the sample picture, and further selecting part of the feature information as the retrieval condition of the target to be queried.

[0164] The sample picture is a picture containing the retrieval condition of the target to be queried.

[0165] Specifically, after a user uploads a sample picture of a target to be queried on a search page, the sample picture is recognized by using image recognition technology, all features in the picture are extracted, and then all the features are returned to the search page for display. The user selects one or more features from all the features as the retrieval condition. For example, a user has only one picture of a place whose name is unknown. After uploading the picture, the place features of the place are automatically extracted, and the user selects the features as the retrieval condition.

[0166] In another implementation manner of the embodiment, after all the features in the picture are extracted, the user continues to input supplementary feature information, and then the supplementary feature information and the feature information of the picture are used as the retrieval condition.

[0167] By the scheme, the analysis of the picture features is used, and the obtained features are used as the retrieval condition for retrieval, so that more retrieval condition input manners are provided for the user, and the use experience of the user is improved.

[0168] In some embodiments, the behavior characteristics include a moving speed, an entering time, a leaving time, an entering direction, and a leaving direction. The behavior characteristics of the target to be queried are analyzed according to the video segment, including: the moving distance and the moving time of the target to be queried are obtained from the video segment, and the moving speed of the target to be queried is further calculated; the time when the target to be queried first appears in the monitoring area is obtained from the video segment as the entering time, and the time when the target to be queried last leaves the monitoring area is obtained as the leaving time; the moving direction when the target to be queried first appears in the monitoring area is obtained as the entering direction; and the moving direction when the target to be queried last leaves the monitoring area is obtained as the leaving direction. The all segments of the target to be queried in the monitoring video that meet the retrieval time span are determined in combination with the behavior characteristics and the distribution of the monitoring cameras in the monitoring system, including: the road segment information monitored by each monitoring camera in the monitoring system and the adjacent situation of each monitoring camera are obtained; the last monitoring camera is determined according to the entering direction, the moving speed, and the adjacent situation; the next monitoring camera is determined according to the leaving direction, the moving speed, and the adjacent situation; the video segment of the target to be queried in the monitoring video of the last monitoring camera is obtained according to the road segment information monitored by the video segment, the road segment information monitored by the last monitoring camera, the moving speed, and the entering time; the video segment of the target to be queried in the monitoring video of the next monitoring camera is obtained according to the road segment information monitored by the video segment, the road segment information monitored by the next monitoring camera, the moving speed, and the leaving time, the above steps are repeatedly performed to obtain new video segments, and the cycle is stopped when the target to be queried leaves the monitoring area of the monitoring system, or the monitoring time of the new video segment reaches the current time, or the target to be queried stops moving, or the monitoring time of the new video segment exceeds the retrieval time span; after the cycle is stopped, all the obtained video segments are all the segments of the target to be queried in the monitoring video.

[0169] In a specific implementation manner, since the monitoring range of each monitoring camera is fixed, the moving distance of the target to be queried in the real world can be calculated according to the moving distance of the target to be queried in the video segment, and then the moving speed of the target to be queried in the moving process is calculated according to the moving time of the target to be queried in the video segment.

[0170] Record the time when the target appears in the monitoring area and the time when it leaves the monitoring area, and use these as the target's entry time and exit time, respectively. Record the target's movement direction when it appears in the monitoring area and the movement direction when it leaves the monitoring area, and use these as the target's entry direction and exit direction, respectively.

[0171] Based on the direction of travel of the target in the video clip and the distribution of the surveillance cameras in the monitoring system, the next surveillance camera in the direction in which the target leaves is determined.

[0172] Next, the distance between the monitored locations of the current and next surveillance cameras is determined using road segment information. Dividing this distance by the target's travel speed yields the travel time. Adding this travel time to the target's departure time calculates the target's entry time into the next surveillance camera's monitoring area. Based on this entry time, travel speed, and road segment information from the next camera's monitoring area, the target's departure time from the adjacent camera's monitoring area can be calculated. Finally, the corresponding video clips from the next surveillance camera are extracted based on the entry and departure times, generating new video segments. Similarly, video clips from the next surveillance camera in the target's direction of entry can be obtained.

[0173] Repeat the steps in this implementation method, continuously obtaining new video clips based on the behavioral characteristics of the target to be queried in the new video clips, the distribution of each surveillance camera in the monitoring system, and the road segment information monitored by the surveillance cameras, until the target to be queried leaves the monitoring area of ​​the monitoring system in the new video clip, or the monitoring time of the new video clip reaches the current time, or the target to be queried stops moving, or the monitoring time of the new video clip exceeds the retrieval time span, then stop the loop.

[0174] For example, a surveillance camera is used to monitor a section of Road A, which is 20 meters long. At 11:30:40 on February 10, it captured a person walking into the monitored area from east to west, and at 11:31:00, it captured the person leaving the monitored area from east to west. Then, the person's walking speed is 60 meters per minute, the entry time is 11:30:40, the departure time is 11:31:00, the entry direction is from east to west, and the departure direction is from east to west.

[0175] The range of the road section monitored by the current monitoring camera is the A Road B Street intersection, the range of the road section monitored by the monitoring camera adjacent to the monitoring camera is the A Road C Street intersection, and the distance between the east intersection of the A Road B Street intersection and the west intersection of the A Road C Street intersection is 1200 meters, so the distance between the monitoring locations of the corresponding two monitoring cameras is also 1200 meters.

[0176] The time when the person arrives at the A Road C Street intersection is 11:51:00, the length of the road section of the A Road C Street intersection is 20 meters, so the time when the person leaves the A Road C Street intersection is 11:51:20, and the video corresponding to the monitoring video of the A Road C Street intersection monitoring camera is 11:51:00-11:51:20.

[0177] The example is only for clear description and easy understanding, and the time is relatively accurate, and the actual interception time will be adjusted accordingly.

[0178] In other implementations, when determining the adjacent camera, a monitoring camera within a certain range in the direction in which the target to be queried leaves is determined.

[0179] By the scheme, the behavior characteristics of the target to be queried in the monitoring video, such as the travel speed, the travel direction, the entering time, and the leaving time, are analyzed, the distribution of the monitoring cameras and the road section information monitored by each monitoring camera are combined, and the monitoring video of the target to be queried shot by the adjacent camera is obtained in a chain recursive manner according to a monitoring video of the target to be queried, so that the efficiency of searching for the video segment about the target to be queried in the monitoring video is improved.

[0180] In some embodiments, based on the target video, a route map of the action track of the target to be queried is drawn, including: obtaining the monitoring information of each road section in the target video, marking each road section in the map with a solid line according to each road section information, connecting the adjacent road sections marked with a solid line in the map with a dashed line according to the travel direction of the target to be queried in the target video, and drawing a direction mark.

[0181] Specifically, since the target video is composed of video segments in which the target to be queried appears, and the road section monitored by each video segment is the road section in which the target to be queried appears, after the target video is completely read, the road section information represented by the video segments constituting the target video is marked in the map respectively, then the adjacent road sections in the map are connected with a dashed line according to the travel direction of the target to be queried in each video segment, and the direction information is marked on the dashed line, so that the behavior track map of the target to be queried is completed.

[0182] By the scheme, the road segment information of the to-be-queried target in each monitoring video is utilized to mark the moving routes of the to-be-queried target on a map, and the moving routes are connected according to the moving direction of the to-be-queried target, so as to complete the drawing of the action track graph of the to-be-queried target.

[0183] Figure 3 A structural schematic diagram of a monitoring intelligent analysis device provided for an embodiment of the present application is shown in FIG. 3. Figure 3 The monitoring intelligent analysis device 300 is used for managing the monitoring videos collected by the monitoring system arranged in a target area, and includes:

[0184] The acquisition module 301 is used for acquiring a search condition of a to-be-queried target; the search condition includes a time condition, a location condition, a feature condition, and a search time span.

[0185] The first search module 302 is used for selecting the monitoring videos in the monitoring system according to the time condition and the location condition, to obtain to-be-determined videos; selecting all to-be-determined targets meeting the feature condition based on the to-be-determined videos, and obtaining pictures of the to-be-determined targets.

[0186] The reverse search module 303 is used for determining the pictures of the to-be-queried target from the pictures of the to-be-determined targets, and determining the video segments of the to-be-queried target in the to-be-determined videos according to the pictures of the to-be-queried target.

[0187] The second search module 304 is used for analyzing the behavior characteristics of the to-be-queried target according to the video segments, and determining all segments of the to-be-queried target meeting the search time span in the monitoring videos in combination with the behavior characteristics and the distribution of the monitoring cameras in the monitoring system.

[0188] The drawing module 305 is used for splicing all the segments according to the time sequence to generate a target video, and drawing a route graph of the action track of the to-be-queried target based on the target video, and calling the display module to display the route graph and the target video.

[0189] Optionally, the monitoring intelligent analysis device 300 further includes a database table module 306.

[0190] The database table module 306 is used for creating a monitoring camera information table.

[0191] The monitoring location information of the monitoring camera and the storage path of the generated monitoring video are stored in the monitoring camera information table as the monitoring camera information.

[0192] A monitoring camera database is created for each monitoring camera.

[0193] The monitoring video is segmented and read according to a preset time period, and feature information of each monitoring target appearing in each segment of the monitoring video is analyzed.

[0194] According to the preset time period and the feature information, a monitoring video table is correspondingly generated in the monitoring camera database.

[0195] Optionally, when the first retrieval module 302 obtains the to-be-determined video by selecting the monitoring video in the monitoring system according to the retrieval condition, the first retrieval module 302 is specifically configured to:

[0196] The monitoring camera information meeting the location condition is selected from the monitoring camera information table;

[0197] According to the storage path in the monitoring camera information, a storage position of the monitoring video generated by the corresponding monitoring camera is entered;

[0198] According to the time indicated by the time condition, a segment containing the time in the monitoring time of the monitoring video is selected as the to-be-determined video;

[0199] When the first retrieval module 302 selects all to-be-determined targets meeting the retrieval condition based on the to-be-determined video and obtains pictures of the to-be-determined targets, the first retrieval module 302 is specifically configured to:

[0200] According to the monitoring camera information, a corresponding monitoring camera database is connected;

[0201] According to the to-be-determined video, a corresponding monitoring video table in the monitoring camera database is selected;

[0202] According to the feature condition, feature information meeting the feature condition is matched from the monitoring video table, and each piece of feature information corresponds to the to-be-determined target;

[0203] For each to-be-determined target, a video segment in which the to-be-determined target appears in the to-be-determined video is intercepted;

[0204] For each video segment, a frame of picture containing the feature information and having the clearest quality is extracted, and the picture is taken as the picture of the to-be-determined target.

[0205] Optionally, the monitoring intelligent analysis apparatus 300 further comprises a video analysis module 307.

[0206] When the video analysis module 307 analyzes feature information of each monitoring target appearing in each segment of the monitoring video, the video analysis module 307 is specifically configured to:

[0207] Key frames are extracted from the monitoring video to obtain a plurality of key frames.

[0208] By comparing all the aforementioned keyframes, a static environment model is established;

[0209] Based on the static environment model, all key frames are analyzed to determine whether the monitoring target appears in the key frames.

[0210] If it occurs, the monitored target will be further analyzed;

[0211] If the target type is determined, the target's body shape, facial features, and clothing information are analyzed. If the target type is determined, the target's license plate information, vehicle type, and vehicle color are analyzed.

[0212] Optionally, when the first retrieval module 302 obtains the retrieval conditions for the target to be queried, it is specifically used for:

[0213] Obtain a sample image of the target to be queried, and obtain the feature information in the sample image;

[0214] Further select some feature information from the aforementioned feature information as retrieval conditions for the target to be queried.

[0215] Optionally, the behavioral characteristics include: travel speed, entry time, exit time, entry direction, and exit direction;

[0216] When the second retrieval module 304 analyzes the behavioral characteristics of the target to be queried based on the video clip, it is specifically used to: obtain the moving distance and moving duration of the target to be queried through the video clip, and further calculate the moving speed of the target to be queried;

[0217] By analyzing video clips, we obtain the time when the target first appears in the monitored area as the entry time; the time when the target last leaves the monitored area as the exit time; the direction of movement when the target first appears in the monitored area as the entry direction; and the direction of movement when the target last leaves the monitored area as the exit direction.

[0218] When the second retrieval module 304 determines all segments in the surveillance video that match the retrieval time span based on the behavioral characteristics and the distribution of surveillance cameras in the monitoring system, it is specifically used to: obtain the road segment information monitored by each surveillance camera in the monitoring system and the adjacent situation of each surveillance camera;

[0219] The previous surveillance camera is determined based on the direction of entry, the speed of travel, and the adjacent conditions; the next surveillance camera is determined based on the direction of exit, the speed of travel, and the adjacent conditions.

[0220] According to the road section information monitored by the video clip, the road section information monitored by the last monitoring camera, the travel speed and the entering time, a video clip of the monitoring video of the last monitoring camera of the target to be queried is obtained; according to the road section information monitored by the video clip, the road section information monitored by the next monitoring camera, the travel speed and the leaving time, a video clip of the monitoring video of the next monitoring camera of the target to be queried is obtained;

[0221] The above steps are cycled to continuously obtain new video clips, and the cycle is stopped if the target to be queried leaves the monitoring area of the monitoring system, or the monitoring time of the new video clip reaches the current time, or the target to be queried stops moving, or the monitoring time of the new video clip exceeds the search time span; after the cycle is stopped, all the obtained video clips are all the clips of the target to be queried in the monitoring video.

[0222] Optionally, when the drawing module 305 draws the route map of the action track of the target to be queried based on the target video, the drawing module 305 is specifically configured to:

[0223] Obtain each road section information monitored in the target video;

[0224] According to each road section information, each road section is marked by a solid line in the map;

[0225] According to the travel direction of the target to be queried in the target video, adjacent road sections marked by a solid line in the map are connected by a dashed line, and a direction mark is drawn.

[0226] The device of the embodiment can be used to execute the method of any of the above embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0227] Figure 4 An electronic device provided in an embodiment of the present application has a structural schematic diagram as shown in Figure 4 The electronic device 400 of the embodiment can include a memory 401 and a processor 402.

[0228] The memory 401 stores a computer program capable of being loaded by the processor 402 and executing the method in the above embodiments.

[0229] The processor 402 and the memory 401 are connected, for example, through a bus.

[0230] Optionally, the electronic device 400 can further include a transceiver. It should be noted that the transceiver in actual application is not limited to one, and the structure of the electronic device 400 does not constitute a limitation on the embodiments of the present application.

[0231] The processor 402 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure. The processor 402 can also be a combination of computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0232] The bus can include a path for transmitting information between the above-mentioned components. The bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is shown in the figure, but it does not mean that there is only one bus or only one type of bus.

[0233] The memory 401 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, an optical disk storage (including a compact disk, a laser disk, an optical disk, a digital versatile disk, a Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program codes in the form of instructions or data structures and can be accessed by a computer, but not limited thereto.

[0234] The memory 401 is used to store application program codes for implementing the scheme of the present application, and is controlled by the processor 402 to execute. The processor 402 is used to execute the application program codes stored in the memory 401 to realize the content shown in the foregoing method embodiments.

[0235] The electronic device includes, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (Personal Digital Assistant), a PAD (Tablet Personal Computer), a PMP (Portable Multimedia Player), a car terminal (for example, a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. It can also be a server or the like. Figure 4 The illustrated electronic device is merely an example and should not impose any limitation on the function and use range of the embodiments of the present application.

[0236] The electronic device of the embodiments can be used to execute the method of any of the above embodiments, and has similar implementation principles and technical effects, which will not be described here.

[0237] The present application also provides a computer readable storage medium storing a computer program capable of being loaded and executed by a processor to execute the method in the above embodiments.

[0238] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The foregoing program can be stored in a computer readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the foregoing storage medium includes ROM, RAM, magnetic disk or optical disk and various media that can store program codes.

Claims

1. A method for intelligent monitoring and analysis, characterized in that, The method for managing surveillance videos collected by a monitoring system located in a target area includes: obtaining search conditions for a target to be queried; the search conditions include: time conditions, location conditions, feature conditions, and search time span; selecting segments of the surveillance video in the monitoring system based on the time and location conditions to obtain a pending video; selecting all pending targets that meet the feature conditions based on the pending video and obtaining images of the pending targets; determining the image of the target to be queried from the images of the pending targets; determining the video segments of the target to be queried in the pending video based on the images of the target to be queried; analyzing the behavioral characteristics of the target to be queried based on the video segments; combining the behavioral characteristics and the distribution of surveillance cameras in the monitoring system to determine all segments of the target to be queried in the surveillance video that meet the search time span; splicing all the segments in chronological order to generate a target video; drawing a route map of the target to be queried's movement trajectory based on the target video, and displaying the route map and the target video. The behavioral characteristics include: travel speed, entry time, exit time, entry direction, and exit direction; the analysis of the behavioral characteristics of the target to be queried based on the video clips includes: obtaining the movement distance and movement duration of the target to be queried through the video clips, and further calculating the travel speed of the target to be queried; obtaining the time when the target to be queried first appears in the monitored area, as the entry time; obtaining the time when the target to be queried last leaves the monitored area, as the exit time; obtaining the movement direction when the target to be queried first appears in the monitored area, as the entry direction; obtaining the movement direction when the target to be queried last leaves the monitored area, as the exit direction; the determination of all segments in the monitored video that match the retrieval time span based on the behavioral characteristics and the distribution of monitoring cameras in the monitoring system includes: obtaining the road segment information monitored by each monitoring camera in the monitoring system and the adjacent situation of each monitoring camera; based on the entry direction, travel speed, The adjacent conditions are used to determine the previous surveillance camera; based on the departure direction, the travel speed, and the adjacent conditions, the next surveillance camera is determined; based on the road segment information monitored by the video clip, the road segment information monitored by the previous surveillance camera, the travel speed, and the entry time, a video clip of the target being searched is obtained from the surveillance video of the previous surveillance camera; based on the road segment information monitored by the video clip, the road segment information monitored by the next surveillance camera, the travel speed, and the departure time, a video clip of the target being searched is obtained from the surveillance video of the next surveillance camera; the above steps are repeated to continuously acquire new video clips; if the target being searched leaves the monitoring area of ​​the monitoring system, or the monitoring time of a new video clip has reached the current time, or the target being searched stops moving, or the monitoring time of a new video clip has exceeded the retrieval time span, the loop stops; after the loop stops, all acquired video clips are all clips of the target being searched in the surveillance video.

2. The method according to claim 1, characterized in that, Also includes: Create a surveillance camera information table; The monitoring location information of the surveillance camera and the storage path of the generated surveillance video are stored as surveillance camera information in the surveillance camera information table; For each surveillance camera, create a surveillance camera database; read the surveillance video in segments according to preset time periods, and analyze the feature information of each surveillance target appearing in each segment of the surveillance video; Based on the preset time period and the feature information, a corresponding monitoring video table is generated in the monitoring camera database.

3. The method according to claim 2, characterized in that, The step of selecting surveillance videos from the monitoring system according to the search criteria to obtain pending videos includes: selecting surveillance camera information that meets the location criteria from the surveillance camera information table; entering the storage location where the corresponding surveillance video is generated according to the storage path in the surveillance camera information; selecting segments of the surveillance video containing the time indicated by the time criteria as pending videos; and selecting all pending targets that meet the search criteria based on the pending videos and obtaining images of the pending targets, including: connecting to the corresponding surveillance camera database according to the surveillance camera information; selecting the corresponding surveillance video table in the surveillance camera database according to the pending videos; matching feature information that meets the feature criteria from the surveillance video table according to the feature criteria, where each feature information corresponds to a surveillance target as a pending target; for each pending target, extracting video segments in the pending videos that have appeared as the pending target; and for each video segment, extracting a frame containing the feature information and having the clearest image quality, and using that frame as an image of the pending target.

4. The method according to claim 2, characterized in that, The analysis of the feature information of each monitored target appearing in each monitoring video segment includes: extracting keyframes from the monitoring video to obtain multiple keyframes; comparing all the keyframes to establish a static environment model; analyzing all keyframes based on the static environment model to determine whether the monitored target appears in the keyframes; if it appears, further analyzing the monitored target; determining the type of the monitored target; if it is determined to be a human, analyzing the body shape, facial features, and clothing information of the monitored target; if it is determined to be a vehicle, analyzing the license plate information, vehicle type, and vehicle color of the monitored target.

5. The method according to claim 1, characterized in that, The step of obtaining the search conditions for the target to be searched includes: obtaining a sample image of the target to be searched, obtaining feature information from the sample image, and further selecting some feature information from the feature information as search conditions for the target to be searched.

6. The method according to claim 1, characterized in that, The step of drawing a route map of the movement trajectory of the target to be queried based on the target video includes: obtaining information on each road segment monitored in the target video; marking each road segment with a solid line on the map according to the information on each road segment; connecting adjacent road segments marked with solid lines on the map with dashed lines according to the direction of travel of the target to be queried in the target video, and drawing direction marks.

7. A monitoring and intelligent analysis device, characterized in that, This system is used to manage surveillance videos collected by a monitoring system located in a target area. It includes: an acquisition module for acquiring search criteria for a target to be queried; the search criteria include: time conditions, location conditions, feature conditions, and a search time span; a first retrieval module for selecting segments of the surveillance video from the monitoring system based on the time and location conditions to obtain a pending video; selecting all pending targets that meet the feature conditions based on the pending video and obtaining images of the pending targets; a reverse retrieval module for determining the image of the target to be queried from the images of the pending targets; and determining the video segments of the target to be queried in the pending video based on the images of the target to be queried; a second retrieval module for analyzing the behavioral characteristics of the target to be queried based on the video segments; and, combining the behavioral characteristics and the distribution of surveillance cameras in the monitoring system, determining all segments of the target to be queried in the surveillance video that meet the search time span; and a drawing module for stitching all the segments together in chronological order to generate a target video; and drawing a route map of the target to be queried's movement trajectory based on the target video, and calling a display module to display the route map and the target video. The behavioral characteristics include: travel speed, entry time, exit time, entry direction, and exit direction; When the second retrieval module analyzes the behavioral characteristics of the target to be queried based on the video clip, it is specifically used to: obtain the moving distance and moving duration of the target to be queried through the video clip, and further calculate the moving speed of the target to be queried; Using the video clip, the time when the target first appeared in the monitored area is obtained as the entry time; the time when the target last left the monitored area is obtained as the departure time; the direction of movement when the target first appeared in the monitored area is obtained as the entry direction; and the direction of movement when the target last left the monitored area is obtained as the departure direction. When the second retrieval module combines the behavioral characteristics and the distribution of surveillance cameras in the monitoring system to determine all segments in the surveillance video that match the retrieval time span for the target to be queried, it is specifically used to: obtain the road segment information monitored by each surveillance camera in the monitoring system and the adjacent situation of each surveillance camera; The previous surveillance camera is determined based on the direction of entry, the speed of travel, and the adjacent conditions; the next surveillance camera is determined based on the direction of exit, the speed of travel, and the adjacent conditions. Based on the road segment information monitored by the video clip, the road segment information monitored by the previous surveillance camera, the travel speed, and the entry time, a video clip of the target being queried in the surveillance video of the previous surveillance camera is obtained; based on the road segment information monitored by the video clip, the road segment information monitored by the next surveillance camera, the travel speed, and the departure time, a video clip of the target being queried in the surveillance video of the next surveillance camera is obtained. Repeat the above steps to continuously acquire new video clips. If the target to be queried leaves the monitoring area of ​​the monitoring system, or the monitoring time of the new video clip has reached the current time, or the target to be queried stops moving, the loop stops, or the monitoring time of the new video clip has exceeded the retrieval time span; after the loop stops, all the acquired video clips are all the clips of the target to be queried in the monitoring video.

8. An electronic device, characterized in that, include: Memory and processor; The memory is used to store program instructions; The processor is used to call and execute program instructions in the memory to perform the intelligent monitoring and analysis method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program; when the computer program is executed by a processor, it implements the intelligent monitoring and analysis method as described in any one of claims 1-6.

Images