A Novel Evaluation Method Based on Entropy for Image Segmentation
Xiaoling Chen, Riming Wang, Yunfei Cao, Weiyu Yu, Jiuchao Feng · Procedia Engineering · 2012
Image segmentation evaluation is still a hotspot problem. Various methods of image segmentation evaluation have been proposed. Amongst all the evaluation approaches, Segmentation Entropy Quantitative Assessment (SEQA) is one of the most popular methods. In this paper, segmentation entropy is proposed. In experiments, some standard images which are segmented by multi-level thresholds are tested and used to conclude the characteristics of SEQA, including its application conditions, advantages and disadvantages in image segmentation evaluation.