ACO-TSSCD: An Optimized Deep Multimodal Temporal Semantic Segmentation Change Detection Approach for Monitoring Agricultural Land Conversion DOI Creative Commons

Henggang Zhang,

Kaiyue Luo, Alim Samat

et al.

Agronomy, Journal Year: 2024, Volume and Issue: 14(12), P. 2909 - 2909

Published: Dec. 5, 2024

With the acceleration of urbanization in agricultural areas and continuous changes land-use patterns, transformation land presents complexity dynamism, which puts higher demands on precise monitoring. And most existing monitoring methods are constrained by limited spatial temporal resolution, high computational demands, challenges distinguishing complex cover types. These limitations hinder their ability to effectively detect rapid subtle use changes, particularly experiencing urban expansion, where shortcomings become more pronounced. To address these challenges, this study a multimodal deep learning framework using semantic segmentation change detection (TSSCD) model optimized with ant colony optimization (ACO) analyze conversion Zhengzhou City, major grain-producing area China. This utilizes Landsat 7/8 imagery Sentinel-2 satellite from 2003 2023 capture spatiotemporal cropland driven infrastructure development, population over last two decades. The TSSCD achieves superior classification accuracy, kappa coefficient improving 0.871 0.892, F1 score 0.903 0.935, 0.848 0.879, indicating its effectiveness identifying changes. significant variation characteristics City were revealed through model, transformations initially concentrated near Zhengzhou’s core expanding outward, east north. results highlight remote sensing techniques conversion.

Language: Английский

The Influence of Human Activities and Climate Change on the Spatiotemporal Variations of Eco-Environmental Quality in Shendong Mining Area, China from 1990 to 2023 DOI Creative Commons
Yu Tian, Zhile Wang, Chuning Ji

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(5), P. 2296 - 2296

Published: Feb. 21, 2025

The Shendong mining area is the largest coal production base in western China. Due to long-term activities, ecological environment quality (EEQ) of has undergone significant changes. Investigating evolution EEQ during process mineral resource exploitation great importance for sustainable development area. However, current research lacks a quantitative assessment contributions climate change and human activities spatiotemporal variations In this study, Remote Sensing Ecological Index (RSEI) was used as an evaluation metric. Theil–Sen slope estimation Mann–Kendall test were applied analyze changes from 1990 2023. Additionally, partial derivative method investigate response characteristics climatic factors quantify relative these two driving factors. results indicate that, over past 34 years, overall study shown improving trend. Compared 1990, proportions areas with good-grade excellent-grade 2023 increased by 28% 23.78%, respectively. second phase (2011–2023), average RSEI time series value significantly compared first (1990–2010). Among factors, annual precipitation had greatest impact on EEQ, contribution rate 0.085. conversion unused land forestland improved showing very increase RSEI, accounting 82.30%. region significant, slight increases smaller than conclusion, trend, being dominant factor 71.52% where increased, while 26.89% decreased.

Language: Английский

Citations

0
ACO-TSSCD: An Optimized Deep Multimodal Temporal Semantic Segmentation Change Detection Approach for Monitoring Agricultural Land Conversion DOI Creative Commons

Henggang Zhang,

Kaiyue Luo, Alim Samat

et al.

Agronomy, Journal Year: 2024, Volume and Issue: 14(12), P. 2909 - 2909

Published: Dec. 5, 2024

With the acceleration of urbanization in agricultural areas and continuous changes land-use patterns, transformation land presents complexity dynamism, which puts higher demands on precise monitoring. And most existing monitoring methods are constrained by limited spatial temporal resolution, high computational demands, challenges distinguishing complex cover types. These limitations hinder their ability to effectively detect rapid subtle use changes, particularly experiencing urban expansion, where shortcomings become more pronounced. To address these challenges, this study a multimodal deep learning framework using semantic segmentation change detection (TSSCD) model optimized with ant colony optimization (ACO) analyze conversion Zhengzhou City, major grain-producing area China. This utilizes Landsat 7/8 imagery Sentinel-2 satellite from 2003 2023 capture spatiotemporal cropland driven infrastructure development, population over last two decades. The TSSCD achieves superior classification accuracy, kappa coefficient improving 0.871 0.892, F1 score 0.903 0.935, 0.848 0.879, indicating its effectiveness identifying changes. significant variation characteristics City were revealed through model, transformations initially concentrated near Zhengzhou’s core expanding outward, east north. results highlight remote sensing techniques conversion.

Language: Английский

Citations

0