基于混合Gabor滤波器与加权中心对称LBP的掌纹识别

掌纹识别是一种比较新颖的生物特征识别技术,提 取最佳分类特征一直是掌纹识别研究领域的一个重要方向。掌纹图像纹理特征丰富,但传统 方法难以准确将其表征。针对此问题,将固定尺度及自适应多尺度Gabor滤波器结合起来, 提出基于混合Gabor滤波器与加权中心对称局部二值模式(weighted center symmetric loca l binary pattern,WCS-LBP)的掌纹识别方法。首先,利用混合Gabor滤波器对掌纹感兴趣 区域进行滤波得到特征图像,并将其串联在掌纹特征空间；然后,使用WCS-LBP提取该空间 掌纹特征形成特征向量；最后,通过匹配WCS-LBP直方图序列的相似度来实现分类。在Poly U图库、同济大学图库、IIT-D图库和自建非接触图库中进行实验。结果显示,该算法获得 的识别率最高分别可达99.768%、99. 510%、99.091%和98.501%,最低等误率分别为 0.794%、1.235%、1.672%和2.339%,且识别时间都在1s以内,相比其他传统和流行算法具有优势 ,显示出方法良好的效果。

Palmprint recognition based on hybrid Gabor filter and weighted central symmetri c LBP

Palmprint recognition is a relatively new biometric recognition technology,extracting the optimal classifying features from palmprint always is an important research area in the palmprint recognitio n field.Palmprint images have rich texture features,but traditional methods are difficult to accurately characterize them.In order to solve this problem,a palmp rint recognition method based on hybrid Gabor filter and weighted center symmetr ic local binary pattern (WCS-LBP) is proposed by combining fixed scale and adap tive multi-scale Gabor filter.Firstly,using the hybrid Gabor filter to extract the region of interest of palmprint to obtain a feature image,and connect it in series in the palmprint feature space.Then,using the WCS-LBP to extract the spa tial palmprint features to form a feature vector.Finally,the classification is a chieved by matching the similarity of WCS-LBP histogram sequences.Experiments w ere carried out in PolyU library,Tongji University library,IIT-D library and se lf-built non-contact library.The results show that the highest recognition rat es obtained by this algorithm are 99.7685%,99.5109%,99.0916% and 98.5010%,and the l owest equal error rate rates are 0.7945%,1.2357%,1.6725% and 2.3391%,respectively,an d recognition time is within 1s,which is superior to other traditional and popul ar algorithms and shows good results.