Adaptive image compression based on segmentation and block classification

This article presents a new digital image compression scheme which exploits a human visual system property-namely, recognizing images by their regions -to achieve high compression ratios. It also assigns a variable bit count t o each image region that is proportional to the amount of information it co nveys to the viewer. The new scheme copes with image nonstationarity by ada ptively segmenting the image into variable block-sized regions and classify ing them into statistically and perceptually different classes. These class es include a smooth class, a textural class, and an edge class. Blocks in e ach class are separately encoded. For smooth blocks, a new adaptive predict ion technique is used to encode block averages. Meanwhile, an optimized DCT -based technique is used to encode both edge and textural blocks. Based on extensive testing and comparisons with other existing compression technique s, the performance of the new scheme surpasses the performance of the JPEG standard and goes beyond its compression limits. In most test cases, the ne w compression scheme results in a maximum compression ratio that is at leas t twice of JPEG, while exhibiting lower objective and subjective image degr adations. Moreover, the performance of the new block-based compression is c omparable to the performance of the state-of-the-art wavelet-based compress ion technique and provides a good alternative when adaptability to image co ntent is of interest, (C) 1999 John Wiley & Sons, Inc.