: Unlike older datasets that focused on specific urban areas, EROS-Thunder covers various terrains, including forests, agricultural lands, and complex urban environments across different seasons.
: Binary masks where white pixels (255) represent change and black pixels (0) represent no change. EROS_-_Thunder.rar
The dataset and paper focus on advancing change detection (CD) in remote sensing by providing a more diverse and extensive collection of high-resolution satellite imagery than previous benchmarks. Key Details of the Paper : Unlike older datasets that focused on specific
: Pre-change and post-change satellite images. Key Details of the Paper : Pre-change and
: The research was conducted by a team from the School of Remote Sensing and Information Engineering at Wuhan University. The EROS-Thunder Dataset :
The paper introduces a baseline model often referred to as (or similar architecture variations), which utilizes a dual-stream encoder-decoder structure designed to handle the multi-scale variations and noise common in large-scale remote sensing imagery. The "Thunder" naming convention reflects the goal of achieving high-speed and high-accuracy processing for massive geospatial datasets. Typical Use Case If you have downloaded this .rar file, it likely contains:
: It specifically targets the "Easy-to-Retrieve and Open-Source" (EROS) philosophy to make high-quality change detection data accessible to the broader research community. Technical Contribution