Multi-threshold Image Segmentation Based on Simulated Annealing Algorithm

赵于前,李慧芬,王小芳
DOI: https://doi.org/10.3969/j.issn.1001-3695.2010.01.114
2010-01-01
Abstract:This paper proposed a threshold segmentation algorithm based on simulated annealing(SA) considering the vast computation of Otsu algorithm for multi-threshold image segmentation.At first,obtained an initial threshold vector through analyzing and processing of histogram.Then made an improvement of SA by setting this vector as the initial solution and approached an optimum threshold vector well applying this improved SA.Compared with Otsu algorithm,the proposed algorithm decreased the computation greatly.For instance,it increased the efficiency of tri-threshold image segmentation by over 400 times in the experiment.The experimental results demonstrate that this algorithm can cope with multi-threshold image segmentation efficiently and rapidly.
What problem does this paper attempt to address?