Data processing method, electronic equipment and storage medium

Abstract

The invention provides a data processing method, electronic equipment and a storage medium, and is applied to a server, the server is connected with a plurality of cameras, the method comprises the following steps: in response to the acquisition of Lx, adding Lx to a historical track segment set to obtain an updated track segment set M = {L1, L2,..., Ln} and La = {Pa1, Pa2,..., Paf (a)}, M including Lx; according to the updated trajectory segment set M, determining a trajectory feature set N = {T1, T2,..., Tn} and Ta = Avg (ta1, ta2,..., taf (a)), where Ta is a trajectory feature corresponding to La, and tab is an object feature obtained by performing feature extraction on an object in Pab; generating a similarity matrix S according to the track feature set N; and clustering the updated trajectory segment set M according to the similarity matrix S to obtain at least one global trajectory, each global trajectory including at least one trajectory segment, and each trajectory segment in the same global trajectory corresponding to the same object.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a data processing method, electronic equipment and a storage medium. Background technique [0002] In recent years, cross-camera object tracking has been a research focus in the field of security and management. Its purpose is to be able to integrate the action trajectories of the same target in different shots. Most of the existing cross-camera target tracking recognizes the same target based on the similarity between pictures, but in many scenes (such as shopping malls), because there are too many targets and each target’s running track is very long, it is necessary to perform track fusion , it is necessary to extract the features of the targets in all the pictures, and then put the features of the pictures together for analysis (feature comparison between each two pictures is required), so as to realize trajectory fusion, and because the number of pictures is often huge Howeve...

AI Technical Summary

Problems solved by technology

Most of the existing cross-camera target tracking recognizes the same target based on the similarity between pictures, but in many scenes (such as shopping malls), because there are too many targets and each target’s running track is very long, it is necessary to perform track fusion , it is necessary to extract the features of the targets in all the pictures, and then put the features of the pictures together for analysis (feature comparison between each two pictures is required), so as to realize trajectory fusion, and because the number of pictures is often huge However, it takes a long time to calculate and process the trajectory fusion, so it can only be processed offline at the set time point, and cannot achieve effective real-time trajectory fusion.

Images