Face detection in complicated backgrounds and different illumination conditions by using YCbCr color space and neural network

This investigation develops an efficient face detection scheme that can detect multiple faces in color images with complex environments and different illumination levels. The proposed scheme comprises two stages. The first stage adopts color and triangle-based segmentation to search potential face regions. The second stage involves face verification using a multilayer feedforward neural network. The system can handle various sizes of faces, different illumination conditions, diverse pose and changeable expression. In particular, the scheme significantly increases the execution speed of the face detection algorithm in the case of complex backgrounds. Results of this study demonstrate that the proposed method performs better than previous methods in terms of speed and ability to handle different illumination conditions.