In emergency scenarios such as fires or earthquakes, rapid and accurate situational awareness of the scene is critical for decision-making processes. Operators may inaccurately visualize descriptions received from victims under stress. In this study, a web-based system is proposed that instantly converts natural language environment descriptions into three-dimensional (3D) scene visualizations. Although Generative AI approaches produce photorealistic images, they carry the risk of hallucination. Therefore, this study adopts a deterministic Scene Assembly approach that prioritizes spatial consistency. The system converts user text into a structured JSON format via an LLM-based parser, validates physical consistency through a rulebased spatial inference layer, and computes evacuation routes using the A* algorithm. Experimental results on 100 scenarios show that while LLM-only achieves 76.3% overall accuracy, the addition of the spatial constraint layer raises this to 85.7%.
Sena Yapsu, Selim Balcisoy· Signal Processing and Commun...· 0 citations
This paper presents a modular decision support system that infers the primary location of the user or the reported incident and a situation-aware risk level from multi-turn Turkish disaster dialogues between a help-seeking user and an AI-supported emergency assistant. The assistant guides the user with follow-up questions about health status, number of affected people, structural damage, environmental hazards, and known nearby landmarks to complete missing information. The system manages the dialogue with a finite state machine, determines the location by linking user cues to a local GeoJSON point-of-interest database and by landmark verification, and produces explainable risk scores with Multi-Criteria Decision Analysis. The key novelty is treating landmarks as an evidence layer that verifies the current location hypothesis through proximity and clustering instead of directly replacing candidates based on a landmark signal. In a ten-scenario pilot evaluation, accuracy, end-to-end latency, and token usage are reported for three configurations.
Eren Varlıker, Yusuf Sinan Özmen, Selim Balcisoy· Signal Processing and Commun...· 0 citations