CN-121982276-A - Target detection method, electronic device, and computer-readable storage medium

CN121982276ACN 121982276 ACN121982276 ACN 121982276ACN-121982276-A

Abstract

The application discloses a target detection method, electronic equipment and a computer readable storage medium, wherein the method comprises the steps of obtaining a target object type of an object to be identified in a target identification area, and a target projection area and a target projection size of the object to be identified on a projection plane; the method comprises the steps of obtaining a verification size matched with a target object type, carrying out size verification on an object to be identified based on the verification size and the target projection size, responding to the object to be identified to pass the size verification, obtaining a reference object in a preset range of a target identification area, determining a reference identification area of the reference object and a reference projection area of the reference object on a projection plane, carrying out area verification on the object to be identified based on at least one of the identification area combination of the target identification area and the reference identification area and the projection area combination of the target projection area and the reference projection area, and taking the object to be identified, which passes the area verification, as the target object. By the aid of the scheme, accuracy of statistics of the target object can be improved.

Inventors

LI JIACHEN
ZHOU RENJIE
WU LIFENG
WANG ZHENGJUN
LU CHAO
SU ZHEN

Assignees

浙江大华技术股份有限公司

Dates

Publication Date: 20260505
Application Date: 20251209

Claims (10)

1. A method of detecting an object, comprising: acquiring a target object type of an object to be identified in a target identification area, and a target projection area and a target projection size of the object to be identified on a projection plane; Acquiring a verification size matched with the target object type, and performing size verification on the object to be identified based on the verification size and the target projection size; responding to the object to be identified passing the size verification, obtaining a reference object in a preset range of the target identification area, and determining a reference identification area of the reference object and a reference projection area of the reference object on the projection plane; Performing region verification on the object to be identified based on at least one of the combination of the target identification region and the identification region corresponding to the reference identification region and the combination of the target projection region and the projection region corresponding to the reference projection region; And taking the object to be identified, which passes the area verification, as a target object.
2. The method of claim 1, wherein the obtaining a verification size matching the target object type, and performing size verification on the object to be identified based on the verification size and the target projection size, comprises: Acquiring a tolerance coefficient and a size threshold matched with the target object type, and acquiring the verification size based on the tolerance coefficient and the size threshold; And performing size verification on the target projection size of the object to be identified by using the verification size.
3. The method according to claim 2, further comprising, after the verifying the size of the target projection size of the object to be identified using the verification size: acquiring a historical projection size corresponding to the identified target object corresponding to the target object type, and determining a verification adjustment size matched with the target object type based on the historical projection size; And performing size verification on the target projection size of the object to be identified by utilizing the verification adjustment size.
4. The method of claim 3, wherein the determining a verification resizing that matches the target object type based on the historical projection size comprises: determining a projection size mean value, a projection size standard deviation and a projection size distribution matched with the target object type based on the historical projection size, wherein the projection size distribution corresponds to a confidence interval; Obtaining a limit coefficient based on a confidence interval corresponding to the projection size distribution; And obtaining a verification adjustment size based on the projection size mean value, the projection size standard deviation and the limit coefficient.
5. The method of claim 1, wherein the obtaining a verification size matching the target object type, and performing size verification on the object to be identified based on the verification size and the target projection size, comprises: Acquiring a historical projection size corresponding to the identified target object corresponding to the target object type, and determining a verification size matched with the target object type based on the historical projection size; And performing size verification on the target projection size of the object to be identified by using the verification size.
6. The method of claim 1, wherein the combination of identification areas corresponds to an identification intersection area between the target identification area and the reference identification area, the identification intersection area matching the target identification area by a first duty cycle; the projection area combination corresponds to a projection intersection area between the target projection area and the reference projection area, and the projection intersection area is matched with the target projection area by a second duty ratio; The object to be identified and the reference object are identified from video frames, and the region verification is verified based on the number of frames of the plurality of video frames, the first duty ratio of which exceeds a first threshold, and/or the number of frames of the plurality of video frames, the second duty ratio of which exceeds a second threshold.
7. The method of claim 6, wherein the performing region verification on the object to be identified comprises: Obtaining matching frames with a first duty ratio exceeding a first threshold value and a second duty ratio exceeding a second threshold value from a plurality of video frames, and determining a check frame number corresponding to the matching frames; and determining the regional verification result of the object to be identified based on the time sequence relevance between the matched frames and the magnitude relation of the verification frame number compared with a frame number threshold.
8. The method according to claim 1, wherein the obtaining the reference object within the preset range of the target recognition area includes: Acquiring candidate objects in a preset range of the target identification area; and taking the candidate object which is mutually different from the target object type of the object to be identified as the reference object.
9. An electronic device comprising a memory and a processor coupled to each other, the memory having program instructions stored therein, the processor being configured to execute the program instructions to implement the method of any of claims 1-8.
10. A computer readable storage medium, characterized in that program instructions executable by a processor for implementing the method of any of claims 1-8 are stored.

Description

Target detection method, electronic device, and computer-readable storage medium Technical Field The present application relates to the field of computer vision, and in particular, to a target detection method, an electronic device, and a computer readable storage medium. Background With the rapid development and popularization of mobile internet technology and the development of transportation, massive video frames are generated. At present, corresponding target objects are usually detected and counted based on video frames, however, in the prior art, the detection of the target objects has higher misjudgment, so that the accuracy of the target object counting is lower. In view of this, how to improve the accuracy of the statistics of the target object is a problem to be solved. Disclosure of Invention The application mainly solves the technical problem of providing a target detection method, electronic equipment and a computer readable storage medium, which can improve the accuracy of target object statistics. In order to solve the technical problems, the first aspect of the application provides a target detection method, which comprises the steps of obtaining a target object type of an object to be identified in a target identification area and at least one of a target projection area and a target projection size of the object to be identified on a projection plane, obtaining a verification size matched with the target object type, carrying out size verification on the object to be identified based on the verification size and the target projection size, responding to the fact that the object to be identified passes the size verification, obtaining a reference object in a preset range of the target identification area, determining a reference identification area of the reference object and a reference projection area of the reference object on the projection plane, and carrying out area verification on the object to be identified based on at least one of the target identification area and the reference projection area corresponding to the reference identification area and the projection area corresponding to the target projection area. To solve the above technical problem, a second aspect of the present application provides an electronic device, including a memory and a processor coupled to each other, where the memory stores program instructions, and the processor is configured to execute the program instructions to implement the method described in the first aspect. To solve the above technical problem, a third aspect of the present application provides a computer-readable storage medium storing program instructions executable by a processor for implementing the method described in the first aspect. According to the scheme, the target object type of the object to be identified in the target identification area is obtained, the target projection area and the target projection size of the object to be identified on the projection plane are obtained, the verification size matched with the target object type is obtained, the size verification is carried out on the object to be identified according to the verification size and the target projection size, when the object to be identified passes the size verification, the reference object in the preset range of the target identification area is obtained, the reference identification area corresponding to the reference object and the reference projection area of the reference object on the projection plane are determined, the area verification is carried out on the object to be identified according to at least one of the combination of the target identification area and the reference identification area and the combination of the projection area corresponding to the target projection area and the reference projection area, the object to be identified which passes the area verification is used as the object to be counted, the object to be identified which is firstly subjected to unified verification standard by defining the target object type, and then the double verification of the size verification and the area verification are combined to form complementation, the size abnormal object is filtered, the suspected object is eliminated, the probability of missed judgment is reduced, and the accuracy of the statistics of the object can be counted. Drawings In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein: FIG. 1 is a flow chart of an embodiment of a target detection method according to the present application; FIG. 2 is a schematic flow chart of another embodiment of th