Restoring clear image from degraded images has long been one challenging problem in digital imaging. In this paper, we present a deblurring method with edge prior and wavelet analysis. A preprocessing step is proposed to remove noise before deblurring, then uses shock filter to enhance edges and canny operation to detect edges for estimating kernel. After that, maximize the sparsity of clear image under tight wavelet frame system. Furthermore, the split Bregman method is proposed to solve the optimization problem. Finally, we will get the clear image. The experiments show that our algorithm can remove motion blurring effectively.

No abstract

Palm will produce the problem of scale inconsistent, rotation and translation during acquisition period,which may cause difficulties in identifying. To solve these problems, a novel palmprint recognition algorithm based on binary horizontal gradient orientation and local information intensity (referred BHOG-LII) has been proposed. First, we use the horizontal gradient template for palmprint image to obtain the gradient image in the horizontal orientation and binarization. Then, the image we get is divided into some girds, and we statistic information intensity of each block as a statistical feature, which are paralleled integration to generate the final feature vector. At last, The chi-square distance is used to classification. Experimental results on PolyU palmprint experiment shows that the proposed method can obtain recognition accuracy up to 99.50%.Compared with some traditional methods, the recognition rate improved significantly. In addition, the proposed algorithm has important significance on the rotation, translation, scaling issues of palmprint recognition.

No abstract