Face illumination compensation based on Armlets
基于Armlets的人脸光照补偿

A new illumination compensation method for face recognition under nonuniform illumination conditions was presented. The method calculated 2-dimensional Armlets (analysis-ready multiwavelets) transform for face illumination compensation in logarithm domain, and then conducted face recognition experiment directly in logarithm domain. The proposed approach was compared with other illumination compensation methods on Yale B face database. The experimental results show that the average error rate of the proposed approach is only 0.18%, which is superior to that of all the existing methods.