VIMARSHAK -- A Web Based Subjective Image Evaluation System

Abstract

In this paper, a secured web based online subjective image evaluation system has been proposed to assess different image processing algorithms. Since many image processing algorithms are designed to enhance the human perception of available image cues, subjective evaluation plays an important role in the assessment of the same. The proposed technique assesses several similar processes by accumulation of votes by individual human evaluators through pair wise comparisons of their outputs. Three tournament strategies are used for the pair wise image comparisons, namely knockout, challenging, and round-robin. The experiments show a satisfactory result in evaluation of accumulated ensemble of evaluators' votes, which is validated using Berkeley boundary detection benchmark dataset.