Potential Application of Hough Transform to Extract Hot Spot Delineated Boundaries Using Landsat-8 Satellite Images
Tasneem Ahmed1, Mohammad Usama2
1Tasneem Ahmed*, Department of Computer Application, Integral University, Lucknow, India.
2Mohammd Usama, Department of Environmental Science, Integral University, Lucknow, India.
Manuscript received on December 15, 2019. | Revised Manuscript received on December 26, 2019. | Manuscript published on January 10, 2020. | PP: 3550-3557 | Volume-9 Issue-3, January 2020. | Retrieval Number: B7342129219/2020©BEIESP | DOI: 10.35940/ijitee.B7342.019320
Open Access | Ethics and 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: Coal fires, also known as subsurface fires or hot spots are all-inclusive issues in coal mines everywhere throughout the globe. Aimless mining over a period of past 100 years has prompted large scale damages to the ecosystem of the earth. For example, debasement in nature of water, soil, air, vegetation dissemination and variations in land topography have caused degradation. Research is needed to be more attentive on developing the prospective use of the satellite image analysis for hot spot detection because ground-based hot spots monitoring is time-taking, complex, cumbrous and very expensive. In this paper, a two-stage model has been developed to extract the hot spot delineated boundaries in Jharia coal field (JCF) region. In the first stage, contextual thresholding (CT) technique has been used to classify the hot spot and non-hot spot regions. After thorough processing, hot spots regions have been retrieved and for performance evaluation sensitivity and specificity are calculated, which suggest that hot spots were detected accurately in successful and efficient way. In second stage, the Canny edge detection algorithm is applied to detect the edges of the hot spot regions and then the binary image is generated, which is later converted into a vector image. Finally Hough transform is implemented on the obtained vector images for delineating hot spot boundaries. In future, delineated hot spot boundaries may be used to obtain the expansion or shrinking information of hot spot regions and it can be used for area estimation also.
Keywords: Hot Spot, Contextual Thresholding, Hough Transform, Sensitivity, Specificity, Edge Detection and Boundary Delineation.
Scope of the Article: Image analysis and Processing