An image processing approach for intensity detection of tropical cyclone using feature vector analysis. (2nd October 2018)

• Record Type: Journal Article
• Authors: Kar, Chinmoy; Banerjee, Sreeparna
• Is Part Of: International journal of image and data fusion. Volume 9:Number 4(2018)
• Page Start: 338
• Page End: 348
• Publication Date: 2018-10-02
• Subjects: Tropical cyclone -- intensity detection -- euclidian distance -- image processing -- feature vector -- center of gravity; Image processing -- Periodicals; Multisensor data fusion -- Periodicals; Multisensor data fusion; Periodicals; 621.36705
• DOI: 10.1080/19479832.2018.1491896
• Languages: English
• ISSNs: 1947-9832

Abstract

ABSTRACT: Tropical cyclones (TC) prediction and intensity detection is one of the most emerging fields of research. Starting from Dvorak technique many approaches are used by meteorologist towards the TC intensity prediction and estimation. In this paper we have proposed an image processing based approach to categorise the cyclone intensity by using feature vector. The feature vector of a TC constructed using mean, variance, density and decentricity. Further, this vector converges to a distinct weight value by random weight genetic algorithm. Finally weight values under each intensity scale are used to find the intensity of a test image. This approach can be used in cyclone intensity detection technique. This paper focuses on estimate the intensity of tropical cyclones from satellite images over the Bay of Bengal.