In this paper, a parallel loop filtering algorithm based on GPU is proposed to accelerate the processing speed for the loop filter module in H.264/AVC protocol. By analyzing the process of two edges are filtering in parallel, this paper proposed a method that changes the boundary strength to make the calculation of deblocking filter adapt to the GPU hardware. Through mapping the deblocking filter onto CUDA threads, this algorithm is able to complete the deblocking filter operation once for all. The experiment shows that the proposed algorithm has little loss on the PSNR of image while the acceleration of HD video can be up to 400%.
In this paper, we proposed a novel real-time mosaic block detection method based on intensity order and shape feature to automatically detect mosaic blocks in video. In the proposed method, the video frame is firstly converted to gray image, and it is divided into several layers according to the gray-scale levels to obtain the binary images. Secondly, performed the morphological operations on binary image include dilation and erosion to exclude the influence of single pixel and small area. Finally, extracted the shape features of connected domain of each binary image are used to detect the mosaic block. Experiments show that the proposed method can effectively detect the abnormal image which containing mosaic blocks in video and has the good performance on accuracy.