Improved technique for segmenting images under natural environment

Abstract

Segmentation is an essential step used for analyzing images in computer vision because digital images must be partitioned into different regions. The objective of this study is to investigate the techniques in thresholding-based method for segmenting images under natural environment. Although thresholding is popular and widely used for image segmentation, it was discovered that modifications to threshold value must be made for several images to produce clearer segmented images.