Scalable Coding Algorithms Of ROI Image Compression

Along with the development of multimedia and networks continuously, the relation between people and the picture information is more and more closely, so the technique of picture coding relatively is also extensively value. Especially the technique of ROI coding provided a new way of solving the antinomy between the picture compression ratio and rebuild image quantity. Because of different application fields, ROI coding also has the different characteristics. In recent years, in fields of pattern recognition and computer view, an face detection technique becomes the very active topic. This text emphatically research ROI coding matching face detection.For the first, this text introduced a method of face detection currently, and detailed discuss it based on template match. Then put forward to pass to use the Matrox Imaging Library deluxe picture processing plait distance database, efficiently and quickly carried out the examination of the interested in district fixed position, for carried on the coding in the interested in district to provide the good experiment foundation smoothly. Analyzed to promote the basic principle, interested in district of the small wave to code the basic theories immediately after and code according to the SPIHT of a transformation. The picture of ROI coding the technique to is a static picture to compress the standard JPEG2000 a new characteristic, adopt the dissimilarity compression strategy according to the dissimilarity of importance of district. The coding of SPIHT is an excellent built-in coding calculate way, good exploitation a coefficient son generation inherits sex to the father generation, carry out the high-efficiency coding, and the code that pass to enter gradually loses the principle at spread, quantity that carries out to code can ratings. The end apply above-mentioned theories in actual, combine the person face examination application the characteristics of the realm, put forward the SPIHT to combine the ratings of the ROI coding project.Be like the quantity through the heavy composition of interested in examine a district of person of picture coding experiment, to a coding strategy with different from the adoption of the background district of person, come to a have the initiative deliver and reconstruction the result of the picture of a district of person, thus under the condition of reducing the picture to save the space and the network bandwidth, assurance important ROI.