VesselBench 800k dataset enables better multimodal vessel detection from space
VesselBench-800K: A Large-scale Perception Benchmark for Multimodal Vessel Detection, Counting, and Density Estimation
Computer Vision and Pattern Recognition
Summary
Detecting ships from space is important for monitoring marine traffic and protecting the environment. Existing datasets mostly use only optical satellite images, which struggle in bad weather or at night. The authors created VesselBench-800K, a huge dataset with 800,000 pairs of images that combine optical and radar views to better spot ships under all conditions. They tested many current computer models on this dataset to see how well they detect, count, and estimate ship density. This new dataset could help improve tools for tracking ships worldwide.
What this means in practice
- •For maritime traffic monitoring teams: Improve ship detection accuracy by using combined optical and radar satellite images for better tracking in all weather and lighting conditions.
- •For environmental protection agencies: Enhance vessel counting and density estimation to monitor and protect marine ecosystems globally through more reliable satellite data.
Authors
Danfeng Hong, Chenyu Li, Jocelyn Chanussot
Abstract
Vessel perception from space is crucial for a wide range of maritime applications, from traffic monitoring to environmental protection. However, most existing datasets predominantly focus on general object detection tasks in optical remote sensing (RS) images. Relying solely on single-modality optical RS images proves inadequate for effectively perceiving vessel objects in complex maritime scenarios, where ever-changing weather conditions (e.g., clouds and rain), the need for day-and-night coverage, and the inherent limitations of a single imaging modality pose significant challenges. To fill this gap, we introduce VesselBench-800K, the largest-to-date benchmark dataset on a global scale for vessel perception in multimodal RS images. As its name suggests, VesselBench-800K comprises 800,000 images, each at a resolution of 512x512 pixels, specifically curated for vessel perception tasks such as detection, counting, and density estimation. These multimodal image pairs (i.e., optical, SAR) are collected from diverse platforms, sensors, scenes, shooting heights, and synthetic sources, spanning spatial resolutions from 4.5m to 0.1m. Furthermore, we evaluate numerous state-of-the-art detection, counting, and density estimation models on VesselBench-800K through both qualitative and quantitative comparisons. By revealing previously unrecognized cues, this dataset holds immense potential to significantly advance our understanding of marine traffic. Our VesselBench dataset will be publicly available at https://github.com/danfenghong/IEEE_TGRS_VesselBench to support and contribute to community development.