Landsat30-AU：澳大利亚Landsat影像的视觉语言数据集
Landsat30-AU: A Vision-Language Dataset for Australian Landsat Imagery
AI总结：
针对现有视觉语言数据集忽视低分辨率多卫星长期档案的问题，构建基于澳大利亚Landsat影像的Landsat30-AU数据集，含图像标题对和VQA样本，评估显示微调模型性能显著提升，助力地球观测民主化。
摘要：
支持卫星影像自然语言交互的视觉语言模型（VLMs）可通过加速专家工作流程、让非专业人士获取数据及实现行星级自动化，推动地球观测民主化。但现有数据集主要聚焦少数卫星的短期高分辨率影像，忽视了低分辨率、多卫星、长期档案（如Landsat），而这类档案对经济实惠且鲁棒性强的全球监测至关重要。我们用Landsat30-AU填补此空白，该大规模视觉语言数据集基于澳大利亚上空4颗Landsat卫星（5、7、8、9）收集的30米分辨率影像构建，跨度超36年。数据集含两部分：Landsat30-AU-Cap（196262个图像-标题对）和Landsat30-AU-VQA（覆盖8个遥感领域的17725个人工验证视觉问答样本）。两者均通过引导式管道构建，利用通用VLMs迭代优化和人工验证确保质量。我们在基准上评估8个VLMs，发现现成模型难以理解卫星影像；开源遥感VLM EarthDial的字幕SPIDEr仅0.07、VQA准确率0.48，凸显当前方法局限。令人鼓舞的是，在Landsat30-AU上轻量微调Qwen2.5-VL-7B后，字幕SPIDEr从0.11提升至0.31，VQA准确率从0.74提升至0.87。代码和数据见https://github.com/papersubmit1/landsat30-au。
英文摘要：
Vision language models (VLMs) that enable natural language interaction with satellite imagery can democratize Earth observation by accelerating expert workflows, making data accessible to non-specialists, and enabling planet-scale automation. However, existing datasets focus mainly on short-term, high-resolution imagery from a limited number of satellites, overlooking low-resolution, multi-satellite, long-term archives, such as Landsat, that are essential for affordable and bias-robust global monitoring. We address this gap with Landsat30-AU, a large-scale vision-language dataset built from 30-meter resolution imagery collected by four Landsat satellites (5, 7, 8, and 9) over Australia, spanning more than 36 years. The dataset includes two components: Landsat30-AU-Cap, containing $196,262$ image-caption pairs, and Landsat30-AU-VQA, comprising 17,725 human-verified visual question answering (VQA) samples across eight remote sensing domains. Both datasets are curated through a bootstrapped pipeline that leverages generic VLMs with iterative refinement and human verification to ensure quality. Our evaluation of eight VLMs on our benchmark reveals that off-the-shelf models struggle to understand satellite imagery. The open-source remote-sensing VLM EarthDial achieves only 0.07 SPIDEr in captioning and a VQA accuracy of 0.48, highlighting the limitations of current approaches. Encouragingly, lightweight fine-tuning of Qwen2.5-VL-7B on Landsat30-AU improves captioning performance from 0.11 to 0.31 SPIDEr and boosts VQA accuracy from 0.74 to 0.87. Code and data are available at https://github.com/papersubmit1/landsat30-au.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
