Pedestrian attribute refined recognition method based on deep learning

Abstract

The invention relates to a pedestrian attribute refined recognition method based on deep learning, and the method comprises the steps: automatically learning the pedestrian attribute features throughimproving a deep learning model frame; inputting the attribute features into classifiers, and training independent pedestrian attribute classifiers, so that the posterior probability that a pedestriansample has the attribute; obtaining the posterior probability of an attribute category through calculating the proportional relationship between the attribute in the pedestrian training sample and the attribute category; and finally obtaining the attribute category of the pedestrian sample according to the Bayesian formula. The method can improve the recognition precision and recognition speed.

Description

technical field [0001] The present invention relates to the technical field of pattern recognition, in particular to a method for refined recognition of pedestrian attributes based on deep learning. Background technique [0002] With the introduction of the concept of safe city in recent years. Surveillance videos are distributed in every corner of the city, maintaining city security. If an accident occurs, it will inevitably consume a lot of manpower and material resources to find effective information from a large number of surveillance images. Pedestrians are the main monitoring targets, if the attributes of pedestrians can be effectively identified, it will bring great convenience to the surveillance video retrieval work. Pedestrian attribute recognition has broad application prospects in video surveillance, intelligent commercial video, pedestrian re-identification, face recognition and other fields, and has attracted more and more researchers' attention. A tradition...

AI Technical Summary

Problems solved by technology

The advantage of artificial neural network is that it has strong nonlinear mapping ability, self-learning and self-adaptive ability, generalization ability and certain fault tolerance ability, but it has the following disadvantages, the convergence speed is slow in the training of pedestrian recognition samples, and its training process is The supervision process, but the labeling of training samples is time-consuming and laborious, and video pedestrian recognition involves the calculation and analysis of a large amount of data, plus the interference of some environmental factors, the traditional pedestrian attribute recognition algorithm cannot extract the preferred features of the image, resulting in the recognition rate limited

Images