基于超分辨激光成像的人脸图像模糊纹理识别方法

Fuzzy texture recognition of face image based on super-resolution laser imaging

纹理特征依赖于图像的高频细节信息,当图像出现模糊时,单纯利用纹理特征的识别方法,识别人脸特征精度会急剧下降,提出基于超分辨激光成像的人脸图像模糊纹理识别方法。通过线激光三目立体视觉人脸扫描成像原理,获取低分辨率的人脸图像,依据该图像采用基于特征脸和匹配追踪算法重建超分辨人脸图像。采用LPQ与Fisherfaces融合的人脸识别算法识别超分辨人脸图像,分块处理超分辨人脸图像,通过LPQ算子获取人脸分块直方图并串接直方图,形成超分辨率人脸直方图,提取人脸图像模糊纹理特征,采用PCA算法和LDA算法对超分辨人脸图像直方图降维,获取2个特征子空间并融合得到融合算子空间,将训练样本的超分辨人脸图像直方图投影到融合算子空间,获取待识别样本特征,通过最近邻分类器对待识别样本特征进行模糊纹理识别。实验表明,所提识别方法识别人脸图像模糊纹理的识别率和准确率都很高,识别效果佳。

The texture feature relies on high frequency detail information of the image.When the image is blurred,the accuracy of face feature recognition will decrease sharply by simply using recognition method based on texture feature.A face image fuzzy texture recognition method based on super-resolution laser imaging is proposed.The low-resolution face images are obtained by the principle of line-laser three-eye stereoscopic vision,and the super-resolution face image is reconstructed based on feature face and matching tracking algorithm.The LPQ fused with Fisherfaces face recognition is used for super-resolution face image to deal the super-resolution face image through block processing.The face block histogram is obtained through LPQ operator and then to be concatenated to form the super-resolution face histogram and the fuzzy texture features of face image are extracted.Using PCA and LDA algorithms for the reduction of super-resolution face image histogram dimensionality to get two feature sub-spaces and get fusion operator space.The histogram of super-resolution face image of training sample is projected into fusion operator space to obtain the sample features to be identified,and the fuzzy texture recognition is performed on the features of the sample to be identified through nearest neighbor classifier.Experimental results show that the proposed method has high recognition rate and accuracy in fuzzy texture recognition.