An Incremental Sparse Linear Regression Classification Algorithm for Face Recognition
WU Guang-chao, Jiao Liu, Sen-tao CHEN · DEStech Transactions on Computer Science and Engineering · 2017
This paper mainly aims at improving the performance of Linear Regression Classification(LRC) when facing large scale training data. Based on LRC, we propose an Incremental Sparse Linear Regression Classification Algorithm (ISLRC) for face recognition. The ISLRC can easily obtain the representation coefficients without inverting a matrix and lower the negative effect of redundant samples or noise. These bring the benefits of faster training speed and higher recognition rate when the amount of the training data is large. Numerical experiments show that it provides on average a 46.23% reduction in training time and a 0.18% raise in recognition rate over the LRC at large scale face data.