基于LBP特征匹配算法的红外人脸图像表情识别技术

Infrared facial image expression recognition technology based on LBP feature matching algorithm

红外人脸图像表识别过程中易受到光照不均匀、角度多变、亮度差异大等问题的影响,导致红外人脸图像表情识别效率较差,为解决上述问题,提出基于LBP特征匹配算法的红外人脸图像表情识别方法。首先通过局部优化保留摄影法对红外人脸图像实行降维处理,获取优化后的图像,然后采用多角度分水岭法分割图像,保留图像的细节信息。并采用LBP算子提取预处理后的图像纹理特征,将提取的纹理特征输入支持向量机中,计算出特征的类内比重,完成红外人脸图像表情的识别。实验结果表明,所提方法的平均识别准确率为92%,识别100张人脸图像表情耗时129 s识别效率高、识别效果好、稳定性强。

In the process of infrared face image table recognition,it is easily affected by problems such as uneven illumination,variable angle,large brightness difference,etc.,which leads to the poor efficiency of infrared face image expression recognition.Face image expression recognition method.Firstly,the infrared face image is dimensionally reduced by the local optimization and preservation photography method to obtain the optimized image,and then the multi-angle watershed method is used to segment the image to preserve the detailed information of the image.The LBP operator is used to extract the preprocessed image texture features,and the extracted texture features are input into the support vector machine to calculate the intra-class proportion of the features,and complete the recognition of infrared facial expressions.The experimental results show that the average recognition accuracy of the proposed method is 92%,and it takes 129 s to recognize 100 face images,with high recognition efficiency,good recognition effect and strong stability.