Optimum indices for vegetation cover change detection in the Zayandeh-rud river basin: a fusion approach. (3rd July 2019)

• Record Type: Journal Article
• Authors: Gholinejad, Saeid; Fatemi, Sayyed Bagher
• Is Part Of: International journal of image and data fusion. Volume 10:Number 3(2019)
• Page Start: 199
• Page End: 216
• Publication Date: 2019-07-03
• Subjects: Change detection -- vegetation indices -- majority voting -- particle swarm optimisation -- Zayandeh-rud river basin; Image processing -- Periodicals; Multisensor data fusion -- Periodicals; Multisensor data fusion; Periodicals; 621.36705
• DOI: 10.1080/19479832.2019.1601642
• Languages: English
• ISSNs: 1947-9832

Abstract

ABSTRACT: In this study, vegetation changes in the Zayandeh-rud river basin in the period 2001 to 2016 have been investigated based on the combining of the 15 different vegetation indices. Two main plans were applied and tested to produce the final vegetation change map. In the first plan, change maps were produced by differencing the original vegetation indices individually. The second plan, which was a fusion perspective, included two algorithms. In the first one, change maps, obtained from vegetation indices, were fused at the decision level using the Majority Voting Method. The second algorithm included a particle swarm optimisation (PSO) based weighted combination of the different vegetation maps. The results show that the high correlation between vegetation indices does not necessarily provide the same results and combination of all them can be done automatically by applying PSO. Although PSO-based combination could not significantly improve the change detection, it solved the problem of finding optimal threshold value for the change detection process. However, from the vegetation change point of view, they all indicate that during the years of study, the vegetation cover has increased in the study area due to irregular water usage of the Zayandeh-rud river in its western part.