The dataset contains 6,000 large-scale ultra-high-resolution remote sensing images, with nearly 60,000 bridges across different backgrounds manually annotated. The image format is 2048×2048-16384×16384 pixels, and has two sets of target detection labels: rotation box and horizontal box. The GLH-Bridge dataset provides strong data support for the task of bridge target detection integrated with large-scale ultra-high-resolution remote sensing images.

This dataset is the latest achievement of Professor Zhang Yongjun's team from the School of Remote Sensing and Information Engineering of Wuhan University on bridge target detection in large-format ultra-high-resolution remote sensing images. It has been included in the top international journal in the field of artificial intelligence, IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI, impact factor: 23.6), with the title 「Learning to Holistically Detect Bridges From Large-Size VHRRemote Sensing Imagery」.