Title: Design and planning of urban ecological landscape using machine learning

Authors: Yajuan Zhang; Tongtong Zhang

Addresses: Environmental Design of Xianyang Normal University, Xianyang City 712000, Shaanxi, China ' Material Forming and Control Engineering, Xi'an University of Architecture and Technology, Xi'an City 710000, Shaanxi, China

Abstract: The purposes are to improve the air quality of the urban ecological environment and increase the green rate of the urban garden ecological landscape. Machine Learning (ML) algorithms are used to analyse and calculate the dust retention outcomes of different plants. Dust retention capabilities and spectral characteristics of several different plants are researched. Results demonstrate a significant correlation between plants and dust retention rate. Red sandalwood has 150 inversion bands, and the optimal inversion algorithm is Random Forest (RF). Zhu Jiao has 74 inversion bands, and the optimal inversion algorithm is the Support Vector Machine (SVM). Ficus microcarpa has 80 inversion bands, and the optimal inversion algorithms are SVM and RF. ML algorithms provide better accuracy than correlation analysis, more suitable for calculating plants' dust retention capabilities. To sum up, ML algorithms can calculate the dust retention amounts of plants to better plan and design regional ecological landscapes.

Keywords: dust retention effect; spectral characteristics; correlation analysis method; machine learning algorithm.

DOI: 10.1504/IJGUC.2022.10045594

International Journal of Grid and Utility Computing, 2022 Vol.13 No.1, pp.3 - 10

Received: 13 Jan 2021
Accepted: 14 May 2021

Published online: 11 Mar 2022 *

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