
Optimal partitioning methods for image segmentation
Author(s) -
Fadnavis Shreyas
Publication year - 2015
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2015.0171
Subject(s) - thresholding , artificial intelligence , image segmentation , computer science , computer vision , merge (version control) , image processing , region growing , segmentation based object categorization , digital image processing , segmentation , scale space segmentation , binary image , image texture , digital image , pattern recognition (psychology) , image (mathematics) , information retrieval
The importance of image processing is increasing in the digitally connected world due to its numerous applications in various fields of medical science, astronomy, weather prediction and video surveillance systems etc. The latest research and development in this field has helped the authors to obtain finer details of a particular image under study. The image segmentation technique, a part of digital image processing, helps to obtain meaningful information of the object. This study discusses the three widely used important image segmentation techniques: namely, split and merge, image growing and thresholding and their effects on a sample image. The authors results thus depict a significant difference in the segmented image by split and merge, image growing and thresholding. Split and merge is the optimal method of image segmentation as compared with the other two techniques mentioned above. The choice of the method varies with type of image, its colour, intensity and noise level.