Research of ROI image compresssion based on visual attention model

摘要

Region of interest (ROI) coding is important in applications where certain parts of an image are of a higher importance than the rest of the image. Human vision system actively seeks interesting regions in images to reduce the search export in tasks, such as object detection and recognition. Similarly, prominent actions in video sequences are more likely to attract human's first sight than their surrounding neighbors. Based on the mechanism of HVS, this paper proposes a model of the focus of attention for detecting the attended regions in video sequences. It uses the similarity between the adjacent frames, establishes the gray histogram, selects the maximum similarity as predicable model, and gets position of the focus of attention in the next fame. And on the application of an algorithm for visual attention the paper shows the region of interest (ROI) coding in JPEG 2000. JPEG 2000 ROI coding is used in combination with an algorithm for VA to provide a progressive bit-stream where the regions highlighted by the VA algorithm are coded as an ROI and presented first in the bit-stream. It can be seen that there is an improvement in image quality centered on the ROI although this is achieved at the expense of reduced quality in the background of the image.