RFI_AI4QC dataset

This dataset was used in the AI4QC project (Artificial Intelligence for Quality Control), in the context of RFI detection through an object detection task. It consists of a set of labeled RFIs (radio frequency interferences). These interferences are caused by man-made sources and can lead to an artefact in the satellite image, typically a bright rectangular pattern. Bounding boxes were defined around RFI artefacts in 3940 Sentinel-1 quick-looks (png images). 

The labeled RFIs are available in three formats: PASCAL VOC (xml files), COCO (json files) and YOLO (txt files). Each is contained in a different zip file. The last zip file contains the 3940 S1 images (quick-looks). One can combine the label files (in a chosen format) with the S1 images to train object detection algorithms to automatically detect RFIs in a satellite image.

本数据集应用于AI4QC项目（Artificial Intelligence for Quality Control），用于通过目标检测任务开展射频干扰（Radio Frequency Interference，RFI）检测相关研究。该数据集包含多组标注后的射频干扰样本，此类干扰由人为活动产生，会在卫星影像中形成伪影，通常呈现为明亮的矩形图案。研究人员在3940张哨兵1号（Sentinel-1）快速预览图（PNG格式图像）中的射频干扰伪影周围标注了边界框。

上述标注后的射频干扰样本提供了三种标注格式：PASCAL VOC（XML格式文件）、COCO（JSON格式文件）以及YOLO（TXT格式文件）。每种格式的标注文件分别存储于独立的压缩包中，最后一个压缩包则包含全部3940张哨兵1号快速预览图。用户可将选定格式的标注文件与哨兵1号影像进行组合，用于训练目标检测算法，以实现卫星影像中射频干扰的自动检测。