A method for abnormal detection and localization of crowd emotions based on deep neural network

Abstract

The invention relates to a method for detecting and locating abnormal crowd emotions based on a deep neural network. Video data is obtained through a monitoring device, video key frame data is extracted from the video key frame data, and face image data of each frame is obtained from the video key frame data. , and perform alignment, grouping, and sorting preprocessing, and then input it into the trained face emotion recognition model based on convolutional neural network, and through the trained crowd emotion detection and positioning model, obtain the abnormal crowd emotion in the monitoring video data The detection and positioning results are fed back to the staff of the monitoring equipment. The design of the present invention is reasonable, and the relationship between abnormal crowd emotions and crowd abnormalities can be obtained through the model, which avoids the problem of detection limitations caused by the association between crowd abnormalities and specific abnormal events. In addition, the model uses a mixed deep neural network structure model, which further improves the efficiency of abnormal detection and location of video crowd emotions.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method for detecting and locating crowd emotional anomalies based on a deep neural network. Background technique [0002] With the continuous and stable development of society and the continuous improvement of people's material and cultural living standards, entertainment activities and commercial activities in urban crowd gathering areas have attracted more and more public attention. These activities often have limited venue space and a large number of participants. It is very easy to cause serious harm to the lives and property safety of the general public. In order to detect abnormal situations as early as possible and take timely measures, we mainly rely on the monitoring equipment widely present in the city to detect and locate abnormal situations through monitoring equipment; [0003] Anomaly detection is mainly divided into two categories: overall anomaly detecti...

AI Technical Summary

Problems solved by technology

[0008] Aiming at the deficiencies of the prior art, the present invention provides a method for detecting and locating abnormal crowd emotions based on a deep neural network, so as to solve the detection limitations caused by the direct correlation between the abnormal definition and specific abnormal events in the existing crowd abnormal detection technology question

Images