NRCD
收藏资源简介:
NRCD数据集是一个用于非配准变化检测的新型遥感数据集。该数据集由海南大学的研究团队创建,旨在应对自然灾害、人为事故和军事打击等紧急情况下,无法获取配准图像的问题。数据集通过对现有的配准图像进行图像变换,模拟了八种非配准场景,包括随机旋转、随机透视、颜色抖动、光学畸变、太阳化、高斯模糊、粗略删除和灰度化。这些场景的模拟旨在测试和改进现有的变化检测算法,以便在紧急情况下能够快速准确地识别关键地物变化。
The NRCD dataset is a novel remote sensing dataset designed for unregistered change detection. It was developed by a research team from Hainan University to address the challenge of unavailable registered images in emergency scenarios including natural disasters, man-made accidents, and military strikes. The dataset simulates eight types of unregistered scenarios by applying image transformations to existing registered images, namely random rotation, random perspective transformation, color jitter, optical distortion, solarization, Gaussian blur, coarse deletion, and grayscale conversion. The purpose of simulating these scenarios is to test and improve existing change detection algorithms, enabling them to quickly and accurately identify key land cover changes in emergency situations.
NRCD数据集概述
基本信息
- 数据集名称: Non-Registration Change Detection (NRCD)
- 提出任务: 非配准变化检测(针对自然灾害、人为事故和军事打击等紧急情况)
- 相关论文: Non-Registration Change Detection: A Novel Change Detection Task and Benchmark Dataset
- 会议: IGARSS 2025
数据集状态
- 代码与数据集: 即将更新(请持续关注)
引用方式
bibtex @inproceedings{igarss2025nrcd, title={Non-Registration Change Detection: A Novel Change Detection Task and Benchmark Dataset}, author={Zhe Shan and Lei Zhou and Liu Mao and Shaofan Chen and Chuanqiu Ren and Xia Xie}, booktitle={IGARSS}, year={2025} }
备注
- 建议使用时引用上述论文并给仓库点赞(Star)。

- 1Non-Registration Change Detection: A Novel Change Detection Task and Benchmark Dataset海南大学 · 2025年



