SOD-CHIEF: Salient Object Detection using Convex Hull with multI-scale Energy Functions

Abstract

In the era of multimedia, the process of detection of salient object has emerged as an interesting research area in the field of image processing and computer vision. However, the existing SOD models are unable to make the strike balance in the trade-off between the accuracy of object detection and the computational time required by humans, In this work the authors addresses the above concerns and focuses to improve the SOD accuracy by considering the requirement of less computation time using the Convex Hull with multi-scale energy functions. Experimental analysis of the proposed technique has been done with the measurement parameters of computational time, F-measure, recall and precision on three publicly available image datasets. The computation time of the SOD-CHIEF model is comparatively less than most of the existing techniques with better accuracy for detecting the Salient object.