基于最大Gabor相似度的多光谱人脸识别系统

以精准识别不同特征维数以及噪声情况下的多光谱人脸为目标,设计基于最大Gabor相似度的多光谱人脸识别系统.原始可见光光谱图像以及热红外光谱图像分别输入系统的可见光光谱图像处理模块以及热红外光谱图像处理模块,图像处理模块分别提取原始可见光光谱图像以及热红外光谱图像特征并建立特征集,多光谱图像融合模块基于特征集内特征利用双...

Multi-spectral face recognition system based on maximum Gabor similarity

In order to accurately recognize multi-spectral face with different feature dimensions and noises,a multi-spectral face recognition system based on maximum Gabor similarity is designed.The original visible spectrum image and the thermal infrared spectrum image are respectively input into the visible spectrum image processing module and the thermal infrared spectrum image processing module of the system.The image processing module extracts the features of the original visible spectrum image and the thermal infrared spectrum image respectively and establishes the feature set.The multi-spectral image fusion module uses the double matching image strategy based on the features in the feature set multi-spectral image.The fused image is used to establish the face database,the face image to be recognized is input into the face recognition module.The face recognition module uses the face database to recognize the face through the face recognition method of maximum Gabor similarity,and outputs the recognition results.The system test results show that the system has high recognition accuracy under the condition of different feature dimension images and salt pepper noise.