Image Recognition Based on a Novel Data Fusion Method

In this paper, a novel data fusion-based method for face image recognition is proposed. Firstly by applying Gabor filter approaches to a image and the image is classified. Secondly, a novel hair locator algorithm, based on the locally dissimilarity pixel feature, has also been applied to the same face image. An image fusion algorithm have been implemented and evaluated on common data sets. We compared our data fusion method with other recognition method. Experiments show that the proposed data fusion method achieves high recognition accuracy and is also not sensitive to noise and other distortions.