Vision-Language Agents for Interactive Forest Change Analysis

TL;DR

This paper presents an LLM-driven vision-language system for interactive forest change analysis, integrating change detection and semantic captioning to enhance interpretability and accessibility of satellite imagery data.

Contribution

It introduces a novel multi-level change interpretation model and the Forest-Change dataset for improved remote sensing image change analysis.

Findings

Achieved 67.10% mIoU and 40.17 BLEU-4 scores on Forest-Change dataset.

Achieved 88.13% mIoU and 34.41 BLEU-4 scores on LEVIR-MCI-Trees.

Demonstrated the system's potential to improve interpretability and efficiency in forest change analysis.

Abstract

Modern forest monitoring workflows increasingly benefit from the growing availability of high-resolution satellite imagery and advances in deep learning. Two persistent challenges in this context are accurate pixel-level change detection and meaningful semantic change captioning for complex forest dynamics. While large language models (LLMs) are being adapted for interactive data exploration, their integration with vision-language models (VLMs) for remote sensing image change interpretation (RSICI) remains underexplored. To address this gap, we introduce an LLM-driven agent for integrated forest change analysis that supports natural language querying across multiple RSICI tasks. The proposed system builds upon a multi-level change interpretation (MCI) vision-language backbone with LLM-based orchestration. To facilitate adaptation and evaluation in forest environments, we further introduce the Forest-Change dataset, which comprises bi-temporal satellite imagery, pixel-level change masks, and multi-granularity semantic change captions generated using a combination of human annotation and rule-based methods. Experimental results show that the proposed system achieves mIoU and BLEU-4 scores of 67.10% and 40.17% on the Forest-Change dataset, and 88.13% and 34.41% on LEVIR-MCI-Trees, a tree-focused subset of LEVIR-MCI benchmark for joint change detection and captioning. These results highlight the potential of interactive, LLM-driven RSICI systems to improve accessibility, interpretability, and efficiency of forest change analysis. All data and code are publicly available at https://github.com/JamesBrockUoB/ForestChat.