Title: Analysis on the spatial dynamic characteristics of land use in the urban agglomeration in central Yunnan based on random forest algorithm

Authors: Lede Niu; Jingzhi Lin; Lifang Zhou; Yan Zhou

Addresses: Faculty of Geography, Yunnan Normal University, Kunming 650500, China ' Faculty of Geography, Yunnan Normal University, Kunming 650500, China ' Faculty of Geography, Yunnan Normal University, Kunming 650500, China ' Faculty of Geography, Yunnan Normal University, Kunming 650500, China

Abstract: In order to improve the accuracy of land use spatial analysis results, this paper takes the urban agglomeration in central Yunnan as an example, and proposes a land use spatial dynamic characteristics analysis method based on random forest algorithm. Using GIS technology to collect and process land remote sensing image data, we extract sparse description features of remote sensing images through the dictionary learning method, build a random forest classification model, classify land use space, and analyse dynamic features. The detailed analysis of land use change in the study area from 2005 to 2020 shows that the cultivated land area in this area has increased by 3,661 km², the dry land area has increased by 3,704 km², and the grassland area has decreased by 2,727 km², with the highest annual change rate of 0.62%.

Keywords: remote sensing image; land use; spatial dynamic feature; random forest algorithm; sparse description feature.

DOI: 10.1504/IJETM.2024.135560

International Journal of Environmental Technology and Management, 2024 Vol.27 No.1/2, pp.92 - 109

Received: 01 Oct 2022
Accepted: 07 Dec 2022
Published online: 18 Dec 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article