Research And Implementation Of Video Preprocessing Algorithm In Video Codec

The video codec which is a very important aspect in the multimedia technology has passed through quantities of researches. With the constitution of a series of video codec standard, like MPEG-1,MPEG-2,MPEG-4 and H.264, which do lots of work in the motion estimation, transformation, quantization and coding, the video compression has got a lot of good results.But nowadays, huge video data quantity contradicts with the need of more clear, more real-time video. The video preprocess technique is a series of steps which aim at deep compression ratio and assure the final image quality.After the notion of video preprocessing has been made, we bring in a summary and category of the video preprocessing. A detail discussion will be made according to the category of one-way preprocessing and two-way preprocessing.In one-way preprocessing, the common noise category and way of processing have been indicated. Based on the traditional mid-filter which does good in the processing of pulse noise, a pre-adjudge has been made which accelerate the processing and reduce the blur of images. After that, we do out preprocessing in a wireless video communication terminal. The processing module which unifies the data format has been made based on FPGA. At last, the preprocessing based on the decoding process has been indicated.In two-way preprocessing, after the study of the common used DCT transform in video coding process, the fact that different centralizations will be generated as the different direction of the image is been noticed. Considering the directional of the natural images, the preprocessing based on rotation has been advanced. Through the rotation which made the blocks be the best direction as the DCT concentration before DCT transformation, better compression result will be made. In this process, a good way of direction detection has been designed, which fits the rotation algorithm and guarantees the rotation process and the compression result.