Research On An Application Of The Wavelet Transform In Video Image Compression Coding

With the arrival of age of Internet and communication,video and image compression is widely used in the fields of wireless communication,interactive multimedia and remote sensing.Its outstanding practical technology has expanded from military,medical and other specific areas to the mass field dominated by ordinary users,and with a promising development.But in the actual application process,we should consider not only its construction cost,but also its reconstruction quality and compression efficiency.But the real-time application is restricted by the limited transmission bandwidth.How to ensure the quality of video and image reconstruction while improving the compression efficiency and reducing the complexity of algorithm has gradually become one of the research hot-spots.Which leads to the research on an application of the wavelet transform in video image compression coding,and we carry out the following specific research:(1)Research on the static image compression coding algorithm.Mainly researched the principle of the embedded zero-tree wavelet encoding algorithm,which based on the background of wavelet transform.Given that the increase of the number of encoding bits and computation during the repeated scans,and a large number of zero-tree root which leads to the complexity of the algorithm,symbol redundancy and other issues,and according to the properties of human visual,an improved embedded zero-tree wavelet encoding algorithm is proposed.In this method,the important coefficient is reclassified by the way of expanding the coded symbol,and the edge information is also defined as an important factor;Secondly,in the process of successive approximation quantification,the way to find zero-tree root and isolated zeros is changed to achieve the rapid judgment of zero-tree structure;Finally,combining the improved algorithm with Huffman coding instead of the arithmetic coding to eliminate symbolic redundancy,which will make it easier.(2)Research on the video image compression coding algorithm.Taking the existing video compression scheme,which based on wavelet transform as the background,in view of the problem of the time sequence redundancy and the poor restoration quality of the image reconstruction in video compression,an improved video compression scheme is proposed.This scheme is mainly considered from two aspects of intra-frame coding and inter-frame coding,the spatial information video segmentation method based on video sequence is adopted.First of all,the video sequence is transformed in the time direction to get each successive image frames,then grouping and the first frame is chosen as reference frame;secondly,the spatial two-dimensional wavelet transform is carried out for each frame image;thirdly,after transformation,in the intra-frame encoding module,the improved embedded zero-tree wavelet encoding algorithm is introduced to encode the first frame,and In the inter-frame coding module,the overlapped block multi-resolution motion compensation technology is used to compare the high-frequency components of other frames representing their motion displacements with the reference frames,and the residual values are encoded;Finally,To completed the related experiments and applied them to natural scene analysis.Experimental results show that the proposed video image compression scheme has some improvements in PSNR and coding efficiency compared with other methods and is suitable for the application in natural scenes.Thus can improve the validity and feasibility of the scheme that proposed in this paper.