ATM abnormal human face detection method based on video monitoring

Abstract

The invention discloses an ATM abnormal human face detection method based on video monitoring. Firstly, binarization processing is performed on a video image through a skin color model; secondly, an effective block mass zone, an eye zone and a mouth zone are selected, and the average number N of white dots in each row of the effective block mass zone, the average number Ne of white dots in each row of the eye zone, and the average number Nm of white dots in each row of the mouth zone are calculated; whether N-Ne is larger than a threshold value N' or not or whether N-Nm is larger than the threshold value N' or not is judged; if N-Ne or N-Nm is larger than N', whether a symmetric index is larger than a threshold value R or not is further judged, if yes, a human face abnormal mark position is determined; finally, whether human face abnormity alarming is performed or not is determined by the statistics of data of the human face abnormal mark position. The ATM abnormal human face detection method based on video monitoring has the advantages that classification and identification are performed based on the shape features of the human eyes or the human mouth, judgment is performed through change of pixel values of the human eyes or the human mouth which is shielded, the robustness of the abnormal human face is greatly improved, in addition, the algorithm is simple, the operand is small, and a low omission rate and a low false detection rate are achieved under the ATM environment.

Description

technical field [0001] The invention relates to image detection technology in video surveillance, specifically, a video surveillance-based ATM machine abnormal face detection method. Background technique [0002] With the rapid development of the banking industry, ATMs (Automated Tellermachines) are developing rapidly. However, while ATMs bring convenience to our lives, they also expose some security problems. In recent years, security incidents against ATMs have occurred frequently, such as masked people stealing illegal money from others. Therefore, how to reduce the occurrence of ATM security accidents has gradually become a topic of widespread concern, especially how to use modern prevention technology to contain security incidents in the bud is an urgent problem to be solved. [0003] As a kind of image data storage and reproduction technology, video surveillance has become an important tool to improve ATM security capabilities. The existing ATM video monitoring syste...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the defects of the prior art for the high false detection rate and missed detection rate of abnormal face detection in front of ATM machines, the present invention provides a method for detecting abnormal face faces in ATM machines based on video monitoring, mainly for wearing sunglasses and The face wearing a mask is detected and the alarm information is output in time, thereby simplifying the amount of calculation, increasing the calculation speed, and reducing the missed detection rate

Images