A Novel Method for Analyzing Infrared Images Taken by Unmanned Aerial Vehicles for Forest Fire Monitoring

Abstract

Conventional forest fire monitoring methods, such as ranger patrol and satellite remote sensing, possess certain limitations. Unmanned Aerial Vehicles (UAVs) have been identified as valuable tools for firefighting due to their high mobility, rapid deployment, and low cost, enabling quick identification of fire sources during the initial stages. To maximize the potential of UAVs in forest fire monitoring, this study investigates a novel approach for analyzing infrared images captured by UAVs for forest fire monitoring purposes. Initially, a diagram illustrating the hardware utilized in a typical forest fire monitoring system is provided. Subsequently, based on static features (such as color and texture) and dynamic features (such as size, location, and shape) of the infrared images captured by UAVs, a new algorithm for detecting suspected fire areas is proposed and employed to make final judgments on potential forest fire regions. Finally, a forest fire identification model is developed based on an improved Probability Neural Network (PNN), and its effectiveness is verified through experimentation. The research presented in this paper could offer valuable insights for forest fire monitoring.

Forest Defender Fusion System for Early Detection of Forest Fires

“…This implies that the Forest Defender Fusion System has a high detection accuracy. Competitive performance is also shown by Wang et al [10], Khan et al [22], and Khan et al [23] at 98.73%, 98.42%, and 95.0%, respectively. The lower accuracy of Sousa et al [24], Tang et al [25], and Sun et al [26] at 93.6%, 92.0%, and 94.1%, respectively, suggest the need for improvement.…”

“…However, the components of the Forest Defender Fusion System including the ECP-LEACH protocol, drone functionalities, and the Intermediate Fusion VGG16 model exhibit consistent behavior in both simulated and real-world environments, minimizing the impact of the simulation-real-world disparity on the overall evaluation of the system's efficacy. [10] 98.73% ---Khan, A. et al [23] 95.0% 95.7% 94.2% 94.96% Sousa, M.J. et al [24] 93.6% 94.1% 93.1% -Tang, Y. et al [25] 92.0% ---Sun, X. et al [26] 94.1% ---Khan, S. et al [22] 98.42% 97.42% 99.47% 98.43% Forest Defender Fusion System 99.86% 99.86% 99.86% 99.86%…”

“…Mashraqi, A. M. [9] proposed a model for the early detection of forest fires using RGB images taken from drones, while Wang, Y. [10] proposed a model using IR images. Chen, X.…”