Object recognition based on generalized linear regression classification in use of color information

Limited size of object images and lack amount of training data would degrade the performance seriously in modern-day recognition applications. Therefore, how to effectively utilize available information from images becomes more and more important. In this paper, we propose to extend the linear regression classification (GLRC), which can effectively use all the information in cases of multiple inputs, e.g. R, G, and B color components. Experimental results for SOIL-47 object dataset and SDUMLA-HMT face database show that the proposed GLRC method with R, G, and B channels performs better than the original LRC and contemporary popular methods.