Importance of Defining Roi: a Semi- Automated Algorithm for Predicting Importance Maps
Vamsi Kadiyala · SHAREOK (University of Oklahoma; Oklahoma State University; Central Oklahoma University) · 2007
The importance of defining ROI in gray scale images is clearly established. A new way of proportionating the distortion between ROI and non-ROI is found out. The various factors influencing the ROI have been studied and the relationship between these factors and the perceived interest has been established. A new Algorithm has been proposed to produce an importance map for a given gray scale image. The algorithm takes into account various factors like Size, Location, Blur and Contrast. The results of the algorithm have been found to be satisfactory. A comparison was made between the proposed algorithm and two other widely used algorithm and the proposed algorithm has significantly better results.