Multiple Face Detection Algorithm for Complex Background Images Based on Edge Structure
Jianguo Shi, Ganyi Tang · 2024
Face detection is widely used in fields such as security monitoring and secure access. In practical detection, the key to facial image detection is to determine the facial region and the edge contour of the face. From the perspective of facial region, a rough facial region is constructed by locating facial key points, and then be corrected based on the structural and edge features of the face. Under complex background conditions, edge features will be weakened. Based on this, we adopted a face target search algorithm based on complex background images and tested multiple face images with complex backgrounds on the VOC2007 dataset. It was proven that this detection algorithm can partially eliminate the influence of edge feature weakening caused by complex background conditions, and effectively improving face detection performance.