Efficient Fire Detection for Uncertain Surveillance Environment

TLDR

Tactile Internet and 5G enable edge intelligence, yet existing CNN‑based fire detection methods struggle in uncertain IoT environments with smoke, fog, or snow, and achieving high accuracy with low latency and small model size on resource‑constrained devices remains difficult. The study proposes an efficient CNN‑based system for detecting fire in videos captured under uncertain surveillance scenarios. The system employs lightweight deep neural networks without dense fully connected layers, reducing computational cost. Experiments on benchmark fire datasets demonstrate that the lightweight CNN outperforms state‑of‑the‑art methods in accuracy, false‑alarm rate, model size, and runtime, making it suitable for mobile and embedded vision applications in uncertain IoT environments.

Abstract

Tactile Internet can combine multiple technologies by enabling intelligence via mobile edge computing and data transmission over a 5G network. Recently, several convolutional neural networks (CNN) based methods via edge intelligence are utilized for fire detection in certain environment with reasonable accuracy and running time. However, these methods fail to detect fire in uncertain Internet of Things (IoT) environment having smoke, fog, and snow. Furthermore, achieving good accuracy with reduced running time and model size is challenging for resource constrained devices. Therefore, in this paper, we propose an efficient CNN based system for fire detection in videos captured in uncertain surveillance scenarios. Our approach uses light-weight deep neural networks with no dense fully connected layers, making it computationally inexpensive. Experiments are conducted on benchmark fire datasets and the results reveal the better performance of our approach compared to state-of-the-art. Considering the accuracy, false alarms, size, and running time of our system, we believe that it is a suitable candidate for fire detection in uncertain IoT environment for mobile and embedded vision applications during surveillance.