Method for detecting human face under backlighting environment

Abstract

The invention relates to a method for detecting a human face under a backlighting environment. The method includes the steps of dynamic camera parameter adjustment, foreground and background cutting, human face positioning, human face tracking and data storage, human face image size filtering, human face image clearness filtering, front-side human face filtering and human face image output. The method has the advantages that parameters of a camera are dynamically adjusted so that the camera can be suitable for detecting human faces under different lighting environments and a better human face image can be acquired under the backlighting environment; a series of simple human face detection methods are designed corresponding to the backlighting environment, the detection effect is good, resources occupied by CPU processing are greatly reduced, and hardware cost of a whole system is lowered.

Description

technical field [0001] The invention relates to a face detection method in a backlight environment. Background technique [0002] With the rapid growth of my country's economy, people's living and material standards have been greatly improved, and people are paying more and more attention to their personal and property safety. At the same time, the rapid economic development accompanied by a sharp increase in urban floating population has added new problems to urban social security. Because most of the traditional security monitoring is just recording video, processing and collecting evidence after the event, it is difficult to prevent the occurrence of potential safety hazards. Therefore, the development of intelligent security monitoring systems has become an inevitable trend in the current construction of safe cities and smart cities. [0003] Generally, the application of building a face detection and recognition system at the entrance is mainly the company's access con...

AI Technical Summary

Problems solved by technology

However, there are very few face detection applications at the entrances of public and entertainment venues. There may be many reasons for this. Considering the technical dimension of face detection alone, the camera is set up at the entrance facing the gate, and there is a strong backlight environment during the day. , it is very difficult to detect faces; the posture of a face in the monitoring area is uncertain, and it is necessary to select an image output that is most suitable for recognition. The general method will have a large amount of calculation, and it is difficult to guarantee high-performance processing equipment. Real-time, if the performance or quantity of processing equipment is increased to solve real-time problems, the hardware cost of the entire system will be increased

Images