Research On The Key Technology Of BTC For Image Compression

With the rapid development of social networks,a large number of digital information data needs to be transmitted over the network,especially the image and video which has a large amount of data.In order to ensure data transmission efficiency and save network bandwidth,the data is compressed and then transmitted.This thesis mainly studies the BTC image compression algorithm.BTC algorithm is a kind of lossy image compression in the spatial domain.The algorithm has the characteristics of fast coding and decoding speed and good quality of compressed image.In order to further understand the BTC algorithm,the influence of block size and quantization threshold on BTC algorithm is studied.The optimal block size is between 4×4 and 8×8,and the Mean method and OTSU method for calculating threshold is better.In order to improve the compression ratio of BTC algorithm,the image blocks are classified according to the characteristics of image blocks,and then different algorithms are used to encode them.The image blocks are divided into uniform blocks,common blocks and complex blocks.The uniform blocks information are expressed by mean value.The common blocks are compressed by BTC algorithm.The complex blocks are compressed by DPCM algorithm.DPCM is also a simple compression algorithm,so the encoding scheme also maintains the advantages of fast encoding.The experimental results show that the coding scheme based on BTC-DPCM has a high compression ratio and can control the compression error well.Because there is a certain similarity between the image blocks,the image has further compressed space.In this thesis,we consider the similarity between image blocks.If the current coding block is similar to the adjacent block,the reconstructed image block of the adjacent block is used directly when the image block is reconstructed.The scheme improves the compression ratio of the scheme when introducing reasonable errors.Compared with the DCT algorithm,the experimental results show that the smooth image has higher compression ratio,the complex image has less compression error.The color image is used often in our daily life,and the scheme of color image compression is proposed.Color image can be divided into R,G,B three color planes,this article uses the brightness bit plane instead of three color bit plane.Because the use of common bit plane,the scheme further improves the coding speed.The experimental results show that the method of using the common plane does not bring more compression error,and the compressed image can not visually see the distortion part.