Open AccessNatural Images Contour Segmentation
Author(s) -
Khairul Azman Ahmad,
Sharifah Lailee Syed Abdullah,
Mahmod Othman
Publication year - 2018
Publication title -
journal of computing research and innovation
Language(s) - English
Resource type - Journals
ISSN - 2600-8793
DOI - 10.24191/jcrinn.v2i4.62
Subject(s) - artificial intelligence , computer vision , computer science , segmentation , pixel , image segmentation , grayscale , edge detection , range segmentation , segmentation based object categorization , scale space segmentation , interpolation (computer graphics) , process (computing) , pattern recognition (psychology) , image processing , image (mathematics) , operating system
This paper, a combination of edge detection and contour based segmentation approach forobject contour delineation is proposed. The proposed approach employs a new methodologyfor segmenting the fruit contour from the indoor and outdoor natural images more effectively.The overall process is carried out in five steps. The first step is to pre-process the image inorder to convert the colour image to grayscale image. Second step is the adoption of Laplacianof Gaussian edge detection anda new corner template detection algorithm for adjustment ofthe pixels along the edge map in the interpolation process. Third step is the reconstructionprocess by implementing two morphology operators with embedded of inversion condition anddynamic threshold to preserve and reconstruct object contour. Fifth step is ground maskprocess in which the outputs of the inference obtained for each pixel is combined to a finalsegmented output, which provides a segmented foreground against the black background.Thisproposed algorithm is tested over 150 indoor and 40 outdoor fruit images in order to analyseits efficiency. From the experimental results, it has been observed that the proposedsegmentation approach provides better segmentation accuracy of 100 % insegmenting indoorand outdoor natural images. This algorithm also present a fully automatic model based systemfor segmenting fruit images of the natural environment.