期刊名称:International Journal of Image Processing (IJIP)
电子版ISSN:1985-2304
出版年度:2014
卷号:8
期号:1
页码:9-16
出版社:Computer Science Journals
摘要:This paper presents a new technique able to provide a very good compression ratio in preserving the quality of the important components of the image called main objects. It focuses on applications where the image is of large size and consists of an object or a set of objects on background such as identity photos. In these applications, the background of the objects is in general uniform and represents insignificant information for the application. The results of this new techniques show that is able to achieve an average compression ratio of 29% without any degradation of the quality of objects detected in the images. These results are better than the results obtained by the lossless techniques such as JPEG and TIF techniques.