T/CSES 17-2020 Guidelines for Remote Sensing Extraction of Typical Urban Surface Features for Monitoring Urban Non-point Source Pollution
T/CSES 17-2020 Guidelines for Remote Sensing Extraction of Typical Urban Surface Features for Monitoring Urban Non-point Source Pollution
Basic Information
Scope
Main technical content: This document aims to accurately identify the structural and spatial distribution of typical urban underlying surface pollution sources that may cause urban non-point source pollution. Based on high-resolution remote sensing images, with the support of ground-based measured spectral data and remote sensing knowledge bases, we use image classification algorithms and some manual assistance to finely extract and classify typical urban underlying surface land cover types such as roofs, roads, green spaces, water bodies, construction sites, farmland, forest land, and greenhouse greenhouses. A remote sensing extraction technology process for urban non-point source pollution monitoring of typical urban underlying surfaces has been established, which mainly includes the selection of high-resolution remote sensing data, data preprocessing, sample collection, feature extraction, image classification, post-processing and result verification, and output of results. The aim is to provide theoretical guidance and technical support for the extraction of typical urban underlying surfaces based on high spatial resolution remote sensing images.