CAJ | 학술논문

为了更好地解决光照变化对人脸识别系统的干扰问题，提出一种融合了NSCT和自适应平滑的算法，以提取带有更多人脸结构信息的光照不变量。首先用NSCT分解对数域人脸图像，并对各高频子带进行NormalShrink阈值滤波；再将滤波后的高频子带和未经处理的低频子带进行逆NSCT 处理得到人脸图像的模糊图像；然后对NSCT分解后的低频子带使用自适应平滑提取出低频子带中的人脸细节信息；最后结合该人脸细节信息和模糊图像进行计算，得到人脸图像的光照不变量。该不变量有效地弥补了 NSCT 方法中缺乏低频子带中的人脸细节信息的不足，提高了人脸信息的利用率。在Yale B和CM U PIE人脸库上的实验结果表明，该算法能够有效地消除光照变化的影响，具有更优的人脸识别性能，提高人脸识别系统的光照鲁棒性。
위료경호지해결광조변화대인검식별계통적간우문제，제출일충융합료NSCT화자괄응평활적산법，이제취대유경다인검결구신식적광조불변량。수선용NSCT분해대수역인검도상，병대각고빈자대진행NormalShrink역치려파；재장려파후적고빈자대화미경처리적저빈자대진행역NSCT 처리득도인검도상적모호도상；연후대NSCT분해후적저빈자대사용자괄응평활제취출저빈자대중적인검세절신식；최후결합해인검세절신식화모호도상진행계산，득도인검도상적광조불변량。해불변량유효지미보료 NSCT 방법중결핍저빈자대중적인검세절신식적불족，제고료인검신식적이용솔。재Yale B화CM U PIE인검고상적실험결과표명，해산법능구유효지소제광조변화적영향，구유경우적인검식별성능，제고인검식별계통적광조로봉성。
To better handle light changes in face recognition ,a method combining NSCT and adaptive smoothing is proposed ,to extract illumination invariants with more structural information of faces . T he method first decomposes a face image in the logarithmic domain , then applies NormalShrink filtering to each high‐frequency subband .Next ,a blurred face image is obtained by performing inverse NSCT on the filtered high‐frequency and original low‐frequency subbands . After that , adaptive smoothing is used to extract detailed facial information from the original low‐frequency subbands . Finally ,a facial illumination invariant is estimated by combining the detailed facial information and the blurred face image .T he resulting illumination invariant has more detailed facial information in the low‐frequency subbands ,compared with results from NSCT . The proposed method better exploits the structural information of faces .Experiments on Yale B and CM U PIE face databases show that the method effectively eliminates illumination variations ,has better performance compared with existing methods ,and improves the robustness of face recognition systems .