Multi-dimensional identity recognition method and system

Abstract

The invention discloses a multi-dimensional identity recognition method and system, and the method comprises the steps: collecting the videos of pedestrians entering an entrance from a plurality of angles when the pedestrians are detected to enter the entrance, and obtaining the videos from a plurality of angles; performing face recognition on the videos at multiple angles to obtain a face recognition result; acquiring a pedestrian route track based on the face recognition result, and collecting a plurality of human body photos; and extracting human body features from the plurality of human body photos, comparing the human body features with standard human body features stored in a database, and if the human body features are consistent with the standard human body features, searching pedestrian identity information corresponding to the standard human body features. According to the method, the detection rate of pedestrian identity recognition is improved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a multi-dimensional identification method and system. Background technique [0002] At present, dense pedestrian scenes mainly refer to human-intensive regional scenes, including railway stations, airports, shopping malls, office buildings, factories, and residential communities. The production and life of the society mainly revolve around the dense pedestrian scene, and the daily work and life process of the residents of the social country is also a process of constantly switching between the dense pedestrian scene. Therefore, the continuous order and safety of dense pedestrian scenes is the top priority of maintaining the national social order. For this reason, the current dense pedestrian scene is also equipped with a camera network for video collection and recording, which is convenient for real-time supervision of the situation in the scene, as well as subse...

AI Technical Summary

Problems solved by technology

The identification of the usual intelligent analysis system mainly relies on face recognition, because it has relatively high quality requirements for face pictures, the shooting distance is relatively long, the image is blurred, the pitch angle of the face, and the angle of the side face do not meet the requirements, which will lead to face recognition. failure of

Images