Research on Identification and Analysis of Land Cover Change Supported by Multi Source Remote Sensing Data
多源遥感数据支持下地表覆盖变化的识别与分析研究

摘要

地表覆盖变化能够综合反映区域生态环境演变、资源利用调整和人类活动影响。单一遥感数据受空间 分辨率、时间分辨率、云覆盖及地物光谱混淆等因素限制，难以稳定识别复杂地表变化。文章分析光学遥感、 合成孔径雷达和全球土地覆盖产品的互补特征，构建数据预处理、特征融合、分类识别、变化检测及精度评 价的分析框架，并结合典型土地覆盖产品的公开参数与精度数据讨论多源数据的应用效果。研究认为，多源 遥感能够增强复杂地物的区分能力和连续监测水平，分类体系差异、尺度效应与时相不一致仍是影响变化识 别可靠性的关键因素。

Abstract

 Land cover change can comprehensively reflect regional ecological environment evolution, resource utilization adjustment, and the impact of human activities. Single-source remote sensing data are limited by factors such as spatial resolution, temporal resolution, cloud coverage, and spectral confusion among ground objects, making it difficult to stably identify complex land surface changes. This paper analyzes the complementary characteristics of optical remote sensing, synthetic aperture radar, and global land cover products, and constructs an analytical framework covering data preprocessing, feature fusion, classification recognition, change detection, and accuracy evaluation. It also discusses the application effects of multi-source data in combination with publicly available parameters and accuracy data of typical land cover products. The study suggests that multi-source remote sensing can enhance the discrimination ability for complex ground objects and improve continuous monitoring, while differences in classification systems, scale effects, and temporal inconsistency remain key factors affecting the reliability of change identification.