Detection method and device and computer readable storage medium

Abstract

The invention provides a detection method, which comprises the following steps of: classifying and identifying a first type of frame and a second type of frame in a neural network model, and independently limiting the positions of the first type of frame and the second type of frame to obtain the position of the first type of frame and the position of the second type of frame; extracting comprehensive semantic information from the first class frame and the second class frame; judging the inclusion relation of the first type of frame and the second type of frame, and when the inclusion relation of the first type of frame and the second type of frame is judged to be correct, performing data correction to reduce the weight when the relative position is correct, so as to perform constraint limitation on the relative position of the position of the first type of frame and the position of the second type of frame. And the first type of frames are positioned in the second type of frames. The problem that the second type of frames and the first type of frames lack relative position constraints is solved, and the accuracy of the relative positions of the second type of frames and the first type of frames is improved. The invention further provides equipment and a computer readable storage medium.

Description

technical field [0001] The present invention relates to the technical field of detection, and in particular, to a detection method, a device and a computer-readable storage medium. Background technique [0002] In recent years, with the continuous development of intelligent identification technology, intelligent video surveillance has played an increasingly important role in many fields such as security and intelligent transportation. The camera can continuously record all the information within the field of view, but the video generated by the camera is a kind of unstructured data, in which the target can only be viewed visually, and cannot be directly read and recognized by the computer. [0003] Pedestrians are the focus of the video surveillance system. Pedestrian structuring refers to extracting pedestrian-related information from video content, transforming it into machine-understandable information, and further transforming it into intelligence, realizing the transfor...

AI Technical Summary

Problems solved by technology

[0006] The object of the present invention is to provide a detection method, equipment and computer-readable storage medium, which solves the problem of the lack of relative position constraints between the second type of frame and the first type of frame, and improves the relative position of the second type of frame and the first type of frame. location accuracy

Images