用GPU加速实现超光谱影像快速无误配准，提升交互体验。
Spacecube: A fast inverse hyperspectral georectification system
- 基于OpenGL的逆向配准算法，突破传统方法速度瓶颈。
- 处理速度远超实时，有效消除像素覆盖伪影。
- 适合遥感数据交互分析与高质量成果导出，开源可用。
超光谱相机每像素可捕获数百个数据点，显著优于RGB或多光谱系统。航空平台远程获取此类数据，但需经过地理配准才能用于分析。传统直接配准方法速度慢且易产生伪影。为此，我们提出Spacecube，一个集成完整超光谱配准流程的程序，包含自研的快速逆向配准技术，利用OpenGL图形编程实现。Spacecube运行速度远超实时，有效消除像素覆盖伪影，支持高质量交互查看、数据探索及最终产品导出。代码已公开，供社区使用。
原文摘要 · Abstract (English)
Hyperspectral cameras provide numerous advantages in terms of the utility of the data captured. They capture hundreds of data points per sample (pixel) instead of only the few of RGB or multispectral camera systems. Aerial systems sense such data remotely, but the data must be georectified to produce consistent images before analysis. We find the traditional direct georectification method to be slow, and it is prone to artifacts. To address its downsides, we propose Spacecube, a program that implements a complete hyperspectral georectification pipeline, including our own fast inverse georectification technique, using OpenGL graphics programming technologies. Spacecube operates substantially faster than real-time and eliminates pixel coverage artifacts. It facilitates high quality interactive viewing, data exploration, and export of final products. We release Spacecube's source code publicly for the community to use.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。