Beard detection method and system in video image

Abstract

The invention relates to a beard detection method and system in a video image. According to the method, a to-be-detected area of interest in a video image is set based on the width and height of the video image; according to a preset foreground detection method, foreground detection is carried out on the area of interest to obtain a foreground area; Haar-feature-based face detection is carried out in the foreground area to obtain a face area; skin color detection is carried out on the face area to localize a mouth part area; and a proportion of a skin color pixel point of the mouth part area in a total pixel point of the mouth part area is calculated and whether the proportion exceeds a set threshold value is determined; if so, a beard is determined. Therefore, an effective detection way is provided for effective detection of a target with a beard in a video image.

Description

Technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a method and system for beard detection in video images. Background technique [0002] In some unsafe areas, for the needs of social security, people wearing special clothing or with special characteristics need to be specially detected and identified, and early warnings are given in advance to remind people to take preventive measures in advance. Although the current image detection can collect images of the monitored area for real-time monitoring, it cannot perform real-time analysis of the collected images, especially for some people who may be unstable, such as those with a big beard. Because people have a big beard, it is difficult to recognize their facial features using ordinary facial recognition technology. It is difficult to effectively monitor and prevent such people, and it is impossible to recognize in real time and give early warning. Summary of th...

AI Technical Summary

Problems solved by technology

Although the current image detection can collect images in the monitoring area for real-time monitoring, it cannot analyze the collected images in real time, especially for some potentially unstable people, such as people with beards, because people with beards Because people have big beards, it is difficult to identify their facial features with ordinary facial recognition technology. It is difficult to effectively monitor and prevent such people, and it is impossible to achieve real-time recognition and early warning.

Images