SatQuery AI: An Interactive Vision-Language Assistant for Multimodal Remote Sensing Image Analysis Through Natural Language Queries

SatQuery AI is an interactive vision-language assistant for multimodal remote sensing image analysis through natural language queries. The system allows users to express satellite imagery analysis tasks using natural language and automatically translates these requests into structured analysis workflows. The system supports dynamic area-of-interest (AOI) selection, temporal and spatial query interpretation, Sentinel-2 satellite imagery retrieval through the Microsoft Planetary Computer STAC infrastructure, image quality control, remote sensing index computation, change detection, image comparison, and geographic visualization. The implemented analysis pipeline includes commonly used remote sensing indices such as Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Normalized Difference Built-up Index (NDBI). An optional vision-language model workflow is also included for scene-level interpretation and multimodal reasoning. This work presents the system architecture, query-processing workflow, satellite data acquisition process, remote sensing analysis methods, change-detection approach, software architecture, API design, and testing methodology of SatQuery AI. Source code:https://github.com/tejask-07/satQuery This record is a preprint and has not been presented as a peer-reviewed journal or conference publication.