Evolutionary Image Thresholding for Image Segmentation

Abstract

Image\ufeff segmentation\ufeff is\ufeff a\ufeffmethod\ufeffof\ufeff segregating\ufeff the\ufeff image\ufeff into\ufeff required\ufeff segments/ regions.\ufeffImage\ufeffthresholding\ufeffbeing\ufeffa\ufeffsimple\ufeffand\ufeffeffective\ufefftechnique,\ufeffmostly\ufeffused\ufefffor\ufeff image\ufeffsegmentation,\ufeff these\ufeff thresholds\ufeffare\ufeffoptimized\ufeffby\ufeffoptimization\ufeff techniques\ufeffby\ufeff maximizing\ufeff the\ufeffTsallis\ufeff entropy.\ufeffHowever,\ufeff as\ufeff the\ufeff two\ufeff level\ufeff thresholding\ufeffextends\ufeff to\ufeff multi-level\ufeff thresholding,\ufeff the\ufeff computational\ufeff complexity\ufeff of\ufeff the\ufeff algorithm\ufeff is\ufeff further\ufeff increased.\ufeffSo\ufeffthere\ufeffis\ufeffneed\ufeffof\ufeffevolutionary\ufeffand\ufeffswarm\ufeffoptimization\ufefftechniques.\ufeffIn\ufeffthis\ufeff article,\ufefffirst\ufefftime\ufeffoptimal\ufeffthresholds\ufeffare\ufeffobtained\ufeffby\ufeffmaximizing\ufeffthe\ufeffTsallis\ufeffentropy\ufeff by\ufeff using\ufeff novel\ufeff hybrid\ufeff bacteria\ufeff foraging\ufeff optimization\ufeff technique\ufeff and\ufeff particle\ufeff swam\ufeff optimization\ufeff (hBFOA-PSO).\ufeffThe\ufeffproposed\ufeffhBFOA-PSO\ufeffalgorithm\ufeffperformance\ufeff in\ufeff segmenting\ufeffthe\ufeffimage\ufeffis\ufefftested\ufeffusing\ufeffnatural\ufeffand\ufeffstandard\ufeffimages.\ufeffExperiments\ufeffshow\ufeff that\ufeffthe\ufeffproposed\ufeffhBFOA-PSO\ufeffis\ufeffbetter\ufeffthan\ufeffparticle\ufeffswarm\ufeffoptimization\ufeff(PSO),\ufeffthe\ufeff cuckoo\ufeffsearch\ufeff(CS)\ufeffand\ufeffthe\ufeffadaptive\ufeffCuckoo\ufeffSearch\ufeff(ACS). KEywoRdS Bacteria Foraging Optimization Technique, Cuckoo Search, Image Segmentation, Image Thresholding, Particle Swam Optimization, Tsallis Entropy International Journal of Computer Vision and Image Processing Volume 9 • Issue 1 • January-March 2019