Search

CN-122024119-A - Target track determining method and device, electronic equipment and medium

CN122024119ACN 122024119 ACN122024119 ACN 122024119ACN-122024119-A

Abstract

The embodiment of the application discloses a target trace determining method, a target trace determining device, electronic equipment and a target trace determining medium. The method comprises the steps of determining a first image collector around an initial image collector for monitoring a closed area if a target is identified in videos of the initial image collector, continuously determining a track of the target if the target is identified through the videos of the first image collector, determining a second image collector around the initial image collector for monitoring an open area if the target is not identified through the videos of the first image collector, and determining the track of the target through the videos of the second image collector. According to the scheme, the video analysis of the first image collector at the closed area where the target most possibly arrives can be determined in a targeted manner, the video analysis of the second image collector at the open area of the target path can be used for determining the target trace, the multiple increase of the processing pressure is not caused, and the target trace can be determined rapidly and accurately.

Inventors

  • HUANG LIBIN

Assignees

  • 浙江宇视科技有限公司

Dates

Publication Date
20260512
Application Date
20241111

Claims (10)

  1. 1. A method of determining a target footprint, the method comprising: if a target is identified in the video of the initial image collector, determining a first image collector surrounding the initial image collector for monitoring the closed area; If the target is identified through the video of the first image collector, continuing to determine the track of the target; If the video passing through the first image collector does not identify a target, determining a second image collector which is used for monitoring an open area around the initial image collector; and determining the track of the target through the video of the second image collector.
  2. 2. The method according to claim 1, wherein the method further comprises: determining a closed area and an open area through map data, and determining an in-pair image collector for monitoring the closed area and an out-pair image collector for monitoring the open area; the closed area, the pair of inner image collectors and the pair of outer image collectors are used as nodes, the closed area, the corresponding pair of inner image collectors and the pair of outer image collectors which are positioned on the same traffic road are connected to form a relationship map; The method comprises the steps of taking the in-out direction which can be monitored by the in-pair image collectors on the closed area as an attribute of the side, taking the passing time between the external image collectors as an attribute of the side, and determining target image collectors around the initial image collector according to a relation map with the attribute, wherein the target image collector comprises the first image collector and the second image collector, and the passing time is the time required by a user to pass between two image collectors connected on one side in historical data.
  3. 3. The method of claim 2, wherein determining a target image collector around the initial image collector from a relationship graph with attributes comprises: determining the target number of the image collectors which can be docked according to the residual computing force under the condition of docking with the image collectors to analyze the video of the image collectors; and starting from the node of the initial image collector, searching the target image collector with the shortest passing time of the target number along the edge.
  4. 4. A method according to claim 3, wherein searching for a target image collector with a shortest transit time for a target number along an edge, starting from a node of the initial image collector, comprises: Starting from the node of the initial image collector, searching each first layer node along each branch, and calculating a first passing time from the initial image collector to the searched first layer node; Selecting a target number of candidate branches with the shortest first transit time, continuously searching for subsequent nodes along the candidate branches, and calculating a second transit time from the initial image collector to a relation chain of the searched subsequent nodes; Aiming at a relation chain of candidate branches with second passing time smaller than or equal to the maximum first passing time, taking a node farthest from the initial image collector in the relation chain as a target image collector, wherein the maximum first passing time is the maximum value of the first passing time of the target number of candidate branches; And taking the first layer node in other candidate branches as a target image collector.
  5. 5. The method of claim 2, wherein the determining of the video of the target image collector for identifying and trails determining the target comprises: Determining an initial moment when a target is identified in the video of the initial image collector; determining a target time after the passing time from the initial time, wherein the passing time is the time required by a user to pass between the initial image collector and the target image collector; and taking the video in a preset time period containing the target moment in the target image collector as the video of the target image collector for identifying the target and determining the track.
  6. 6. The method of claim 1, wherein if an object is identified by the video of the first image collector, continuing to determine a track of the object comprises: determining whether the target enters the closed area according to the passing direction of the target in the video of the first image collector; If yes, determining a track of the target through an intra-pair image collector for monitoring the closed area; If not, the first image collector is used as the initial image collector, and the target track determining method is continuously executed.
  7. 7. The method of claim 1, wherein the second image collector comprises a predetermined accent image collector and a plain image collector; correspondingly, determining a second image collector surrounding the initial image collector for monitoring the open area, and determining the track of the target through the video of the second image collector comprises the following steps: determining key image collectors around the initial image collector for monitoring an open area; if the target is identified through the video of the key image collector, continuing to determine the track of the target; if the target is not identified through the video of the key image collector, determining a common image collector which is used for monitoring an open area around the initial image collector, wherein the priority of target identification based on the key image collector is higher than the priority of target identification based on the common image collector; and determining the track of the target through the video of the common image collector.
  8. 8. A target track determining apparatus, the apparatus comprising: The first image collector determining module is used for determining a first image collector which is used for monitoring the closed area around the initial image collector if a target is identified in the video of the initial image collector; The first target track determining module is used for continuously determining the track of the target if the target is identified through the video of the first image collector; The second image collector determining module is used for determining a second image collector which is used for monitoring an open area around the initial image collector if the video passing through the first image collector does not identify a target; and the second target track determining module is used for determining the track of the target through the video of the second image collector.
  9. 9. An electronic device, the electronic device comprising: at least one processor, and A memory coupled to the at least one processor data processing, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the target track determination method of any one of claims 1-7.
  10. 10. A computer readable storage medium storing computer instructions for causing a processor to perform the target track determination method of any one of claims 1-7.

Description

Target track determining method and device, electronic equipment and medium Technical Field The present application relates to the field of target trace determination technologies, and in particular, to a target trace determination method, device, electronic apparatus, and medium. Background Intelligent services are widely used at present, for example, services that perform intelligent analysis on video acquired by an image acquisition device to identify a target. The large amount of intelligent analysis needs to consume a large amount of calculation power, and particularly, the urban intelligent monitoring system needs a large amount of calculation power to cover all cameras, so that a large amount of hardware resources are needed to be consumed, and the cost is high. At present, in a city-level monitoring system, intelligent analysis is not started by all cameras, and an analysis task is established only when needed. At present, the scheme for determining the target track based on intelligent analysis business is that if a target is identified in a picture of a certain image collector in the background monitoring process, the video of the image collectors around the image collector is gradually intelligently analyzed by gradual expansion so as to determine whether the target exists. However, in the process of gradual expansion analysis, the number of image collectors to be analyzed is multiplied, so that the demand for calculation force is multiplied, huge processing pressure is generated, and the efficiency of determining the target trails is influenced. Disclosure of Invention The embodiment of the application provides a target trace determining method, a target trace determining device, electronic equipment and a target trace determining medium, which can accurately and rapidly determine a target trace under the condition of not causing the multiple increase of processing pressure. According to an aspect of the present application, there is provided a target track determining method, the method comprising: if a target is identified in the video of the initial image collector, determining a first image collector surrounding the initial image collector for monitoring the closed area; If the target is identified through the video of the first image collector, continuing to determine the track of the target; If the video passing through the first image collector does not identify a target, determining a second image collector which is used for monitoring an open area around the initial image collector; And determining the track of the target through the video of the second image collector. According to an aspect of the present application, there is provided a target track determining apparatus, the apparatus comprising: The first image collector determining module is used for determining a first image collector which is used for monitoring the closed area around the initial image collector if a target is identified in the video of the initial image collector; The first target track determining module is used for continuously determining the track of the target if the target is identified through the video of the first image collector; The second image collector determining module is used for determining a second image collector which is used for monitoring an open area around the initial image collector if the video passing through the first image collector does not identify a target; and the second target track determining module is used for determining the track of the target through the video of the second image collector. According to another aspect of the present application, there is provided an electronic device including: at least one processor, and A memory coupled to the at least one processor data processing, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the target track determination method of any one of the embodiments of the application. According to another aspect of the present application, there is provided a computer readable storage medium storing computer instructions for causing a processor to perform the target track determining method of any of the embodiments of the present application. According to the technical scheme, if a target is identified in the video of the initial image collector, a first image collector which is used for monitoring a closed area around the initial image collector is determined, if the target is identified through the video of the first image collector, the track of the target is continuously determined, if the target is not identified through the video of the first image collector, a second image collector which is used for monitoring an open area around the initial image collector is determined, and the track of the target is determined through the video of the second image collector. According to the scheme, the problem that when the video of the image collector is