Surgically altered face detection using log-Gabor wavelet

Plastic surgery can be reconstructive to correct facial feature anomalies or cosmetic to improve the appearance. Both corrective and cosmetic surgeries modify the original facial information to a large extent thereby posing a great challenge for face recognition algorithms. This paper employs an edge-based log-Gabor feature representation approach for the recognition of surgically altered faces. A log-Gabor based feature, termed as Edge-based Histogram of log-Gabor Magnitude Patterns (Edge-based HLGMP), is proposed and is very simple but effective. To ensure that the edge information richly captures significant features of the faces, a simple illumination normalization process is applied prior to edge information extraction. Experimental results on plastic surgery database shows that the proposed method performs well in comparison to existing plastic surgery face recognition methods reported in the literature.

整形手术可以是修复性的，用于矫正面部特征异常，也可以是美容性的，用于改善外貌。矫正手术和美容手术都会在很大程度上改变原始面部信息，从而给人脸识别算法带来巨大挑战。本文采用一种基于边缘的对数伽柏（log - Gabor）特征表示方法来识别经过手术改变的面部。提出了一种基于对数伽柏的特征，称为基于边缘的对数伽柏幅值模式直方图（基于边缘的HLGMP），它非常简单但很有效。为了确保边缘信息能充分捕捉面部的重要特征，在提取边缘信息之前应用了一个简单的光照归一化过程。在整形手术数据库上的实验结果表明，与文献中报道的现有整形手术人脸识别方法相比，所提出的方法表现良好。