In order to solve the problem of low resolution of iamge, a new image super-resolution reconstruction method based on four-channel convolutional sparse coding was proposed. Firstly, the input image was turned over 90° in turn as the input of four channels, and an input image was decomposed into the high frequency part and the low frequency part by low pass filter and gradient operator. Then, the high frequency part and low frequency part of the low resolution image in each channel were reconstructed by convolutional sparse coding method and cubic interpolation method respectively. Finally, the four-channel output images were weighted for mean to obtain the reconstructed high resolution image. The experimental results show that the proposed method has better reconstruction effect than some classical super-resolution methods in Peak Signal-to-Noise Ratio (PSNR), Structural SIMilarity (SSIM) and noise immunity. The proposed method can not only overcome the shortcoming of consistency between image patches destroyed by overlapping patches, but also improve the detail contour of reconstructed image, and enhance the stability of reconstructed image.

In recent years, Sparse Representation based Classifier (SRC) has become a hot issue which has been great successful in face recognition. However, when the SRC reconstructed test samples, it is possible to use the training samples with large difference from the test samples, meanwhile, SRC tends to lose locality information and thus produces unstable classification results. A Self-Paced Learning Weighted Sparse Representation based Classifier (SPL-WSRC) was proposed. It could effectively eliminate the training samples with large difference from the test samples. In addition, locality information between the samples was considered by weighting to improve the classification accuracy and stability. The experimental results on three classical face databases show that the proposed SPL-WSRC algorithm is better than the original SRC algorithm. The effect is more obvious, especially when the training samples are enough.