JPEG2000 ROI coding with fine-grain accuracy through rate-distortion optimization techniques

Abstract

Region of interest (ROI) coding is a feature of prominent image coding systems that enables the specification of different coding priorities to certain regions of the image. JPEG2000 provides ROI coding through two mechanisms: either modifying wavelet coefficients or using rate-distortion optimization techniques. Although ROI coding methods based on the modification of wavelet coefficients provide an excellent accuracy to delimit the ROI area (referred to as fine-grain accuracy), they significantly penalize the coding efficiency. On the other hand, methods based on rate-distortion optimization improve the coding efficiency but, so far, have not been able to achieve the intended fine-grain accuracy. This letter introduces two ROI coding methods that, using rate-distortion optimization techniques, achieve a fine-grain accuracy, comparable to the one obtained when wavelet coefficients are modified, and are competitive in terms of coding efficiency.