2-D Gabor filter based transition region extraction and morphological operation for image segmentation. (August 2017)

• Record Type: Journal Article
• Authors: Parida, Priyadarsan; Bhoi, Nilamani
• Is Part Of: Computers & electrical engineering. Volume 62(2017)
• Page Start: 119
• Page End: 134
• Publication Date: 2017-08
• Subjects: Transition region -- Thresholding -- 2D-Gabor filter (two dimensional Gabor filter) -- Morphological operation; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2016.10.019
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Transition region is extracted using 2D-Gabor filter and global thresholding. The method uses morphological operations for image segmentation containing single and multiple objects. The method is robust as it gives better performance in noisy conditions. Abstract: Transition region-based image segmentation techniques have proved effective due to their simplicity and efficient computation. These techniques greatly depend on the accurate extraction of transition regions. Transition region extraction becomes difficult when there is grey level overlapping between foreground and background. Further, the performance of these methods deteriorates when the background and foreground are textured or are of varying intensities. Also, these are applied mostly for single object segmentation. To overcome these shortcomings, we propose a robust hybrid method for image segmentation containing single and multiple objects. The proposed method uses a two-dimensional Gabor filter which enhances the boundaries of object regions for better extraction of transition regions. These transition regions undergo morphological operations to get the object contours and object regions. Finally, objects are extracted from the object regions. Experimental results show that the proposed method yield superior performance for image segmentation containing single and multiple objects. Graphical abstract: