Acquatic Rare Species Habitat Detection and Tracking
Ramya R.
Mrs. Ramya. R, Assistant Professor, A.V.C College of Engineering, Mayiladuthurai (Tamil Nadu), India.
Manuscript received on 13 April 2017 | Revised Manuscript received on 20 April 2017 | Manuscript Published on 30 April 2017 | PP: 134-137 | Volume-6 Issue-4, April 2017 | Retrieval Number: D4928046417/17©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Computer vision has gained paramount significance in recent times due to the increased use of cameras as portable devices and their incorporation in standard PC hardware, mobile devices, machines etc. Computer vision techniques such as detection, tracking, segmentation, recognition and so on, aim to mimic the human vision system. Humans hardly realize the complexities involved in vision, but in fact, our eye is more powerful than it seems. It processes around 60 images per second, with each image consisting of millions of points. Computer vision is still a long away from its goal of replicating the human eye, but in the meantime various computer vision techniques are being applied to complex applications. The proposed algorithm is resistant to small illumination changes and also involves a module that reduces effects of camera movement. . In this system four static cameras are used to capture the moving objects. Background subtraction method subtracts the moving object from static underwater place. This procedure is done by pixel by pixel. Area of the species is also main consideration. Once the species are detected from static underwater place, using background subtraction method tracking is done on each of the four sides. Gaussian mixture model (GMM. and BLOB analysis method is applied for counting the rare species. Gaussian mixture model gives the better segmentation to the original images. BLOB analysis produces the bounding boxes to the species.
Keywords: Blob Analysis, Gaussian Mixture Model, MATLAB.
Scope of the Article: GIS and GPS