Long-time face tracking method based on intelligent cloud platform of patrol robot

Abstract

The invention discloses a long-time face tracking method based on an intelligent cloud platform of a patrol robot. The method comprises the steps that an adaboost face detection module is used to carry out face detection in a region of interest in an input video image frame; a detected face is initialized; a face tracking module tracks the initialized face, and if tracking is effective, a face template updating module is started to update a face template; and if tracking is ineffective, a template matching face detection module is started to carry out template matching detection of the detected face. According to the invention, the face is detected in higher instantaneity, the modules are tracked and learnt, the speed is higher, a target which is lost in tracking can be relocated, the face can be tracked in long time, and multi-target tracking is in real time.

Description

technical field [0001] The invention relates to a face tracking method, in particular to a long-time face tracking method based on an intelligent cloud platform of a patrol robot. Background technique [0002] In the external environmental information perceived by human beings, visual information accounts for a considerable proportion, and dynamic visual information is an important part of it. Because visual information plays a very important role in human perception, video image processing technology has always been a hot spot in scientific research. Machine vision can replace humans in the work of target detection, target tracking, target recognition, etc. It has sharper insight and the ability to continue to work. In the case of quickly obtaining accurate information from video images, machine vision and human Vision has a significant advantage over [0003] Object tracking is the key technology of machine vision, the basis of object recognition and behavior understandi...

AI Technical Summary

Problems solved by technology

However, when used alone, the tracking effect is not very good

For area-based tracking, since each point in the entire target area must correspond, the algorithm is time-consuming

The TLd tracking algorithm takes a lot of time to detect and learn a single target, and is not satisfied with the environment where there are many targets

Images