AN ROI SEARCH METHOD FOR STILL IMAGES BASED ON SET DESCRIPTIONS

Detecting a Region of Interest (ROI) from a digitized image is an impo rtant step towards highly efficient image communication. Although a si gnificant amount of work has been done for time varying images, in whi ch the ROI is expressed in terms of the differences between sequential frames, only limited work has been done for analyzing still images. I n this paper, we present an ROI search algorithm for still images base d on a description of the objects in the image. The proposed approach uses no matching techniques, but rather uses a set descriptor of the p attern. Experimental results obtained by applying this technique to fa cial images show that the proposed methodology is robust with regard t o variations in scale and angle as well as stray variations in the ima ge.