This research presents a novel framework for real-time disaster response using autonomous drone image analysis, integrating deep learning-based object detection with fuzzy decision-making for optimized deployment. The research introduces a unique approach that combines the YOLOv8 model with a fuzzy multi-criteria evaluation system to enhance object detection, communication infrastructure, and autonomous landing in complex disaster environments. A key innovation lies in the incorporation of multimodal data, including both visible-spectrum and thermal drone imagery, which improves detection accuracy in low-visibility conditions. The system is trained and validated on a custom dataset comprising over 10,000 annotated images collected from simulated disaster scenarios, achieving over 90% detection accuracy and a 15–20% improvement in response time and resource allocation compared to conventional methods. This integration enables more reliable real-time identification of affected individuals and infrastructure, contributing to faster and more effective emergency response. The research advances the current state of autonomous systems for disaster management by providing an affordable, scalable, and performance-optimized solution for emergency detection and decision-making.