سال انتشار: ۱۳۸۳
محل انتشار: سومین کنفرانس ماشین بینایی و پردازش تصویر
تعداد صفحات: ۸
Hadi Fatemi – ECE Department University of Tehran
Carol Lucas – ECE Department University of Tehran
In this paper we have proposed a new method of image compression, utilizing non-uniform sampling and interpolation (NSI) of the image intensity surface, and genetic algorithm as an important optimization tool. The goal was creating a compression algorithm with the following properties: (1) having satisfactory qualities in high compression rates, (2) losing minimum correlation data during the compression, (3) well reconstruction of edges in image after decompression, (4) Asymmetric, i.e. with lower decompression complexity and potentially higher compression complexity, (5) somehow intelligent, to get satisfactory results for most kind of images. (Lossless if possible, otherwise a good lossy one) To achieve these goals we have used NSI, and GA as sampling points selection algorithm. Finally, we’ll see results of some experiments and compare this method with JPEG compression.