A face detection method based on skin color and adaboost algorithm

Abstract

The invention discloses a face detection method based on skin color and Adaboost algorithm. After the input image or video is preprocessed, the RGB space is converted into YCb'Cr' space by nonlinear segmentation, and then binarized and processed. Morphological processing to obtain the segmented skin color area; the cascade classifier trained by the Adaboost algorithm is used to perform face detection on the segmented skin color area, and the result is output to complete the detection. The face detection algorithm of the present invention not only guarantees the detection rate but also guarantees the efficiency, uses the non-linear color space subsection conversion to make up for the inaccuracy brought by the traditional linear conversion, and ensures that the skin color pixels are extracted to the maximum extent; the retrained Adaboost The cascade classifier can reduce the number of stages of the original cascade classifier, reduce training time, and improve detection speed.

Description

technical field [0001] The invention belongs to the technology about face detection in the field of pattern recognition, in particular to a method for realizing face detection by using computer technology, digital image processing technology, pattern recognition technology, etc., specifically a kind of face detection based on skin color and Adaboost algorithm Detection method. Background technique [0002] 1. Face recognition technology [0003] Face recognition technology specifically refers to computer technology that uses analysis and comparison of facial visual feature information for identity identification. Face detection is a key link in the automatic face recognition system. Early face recognition research mainly focused on face images with strong constraints (such as images without background), often assuming that the position of the face remains unchanged or easy to detect. Obtained, so the problem of face detection has not been taken seriously. With the develop...

AI Technical Summary

Problems solved by technology

[0015] Since the face detection method based on skin color features has poor detection performance and a high false detection rate, it is generally used in combination with other methods.

However, there is still no face detection algorithm combined with the Adaboost algorithm. Making full use of the combined application of the two has become a research direction in this professional field.

Images