This article proposes a novel method for detecting forest fires, through the use of a new color index, called the Forest Fire Detection Index (FFDI), developed by the authors. The index is based on methods for vegetation classification and has been adapted to detect the tonalities of flames and smoke; the latter could be included adaptively into the Regions of Interest (RoIs) with the help of a variable factor. Multiple tests have been performed upon database imagery and present promising results: a detection precision of 96.82% has been achieved for image sizes of 960 × 540 pixels at a processing time of 0.0447 seconds. This achievement would lead to a performance of 22 f/s, for smaller images, while up to 54 f/s could be reached by maintaining a similar detection precision. Additional tests have been performed on fires in their early stages, achieving a precision rate of p = 96.62%. The method could be used in real-time in Unmanned Aerial Systems (UASs), with the aim of monitoring a wider area than through fixed surveillance systems. Thus, it would result in more cost-effective outcomes than conventional systems implemented in helicopters or satellites. UASs could also reach inaccessible locations without jeopardizing people’s safety. On-going work includes implementation into a commercially available drone.

An implementation of an artifact for security in 5G networks using deep learning methods

In digital image processing and computer vision, a fairly frequent task is the performance comparison of different algorithms on enormous image databases. This task is usually time-consuming and tedious, such that any kind of tool to simplify this work is welcome. To achieve an efficient and more practical handling of a normally tedious evaluation, we implemented the automatic detection system, with the help of MATLAB®’s Parallel Computing Toolbox™. The key parts of the system have been parallelized to achieve simultaneous execution and analysis of segmentation algorithms on the one hand and the evaluation of detection accuracy for the nonforested regions, such as a study case, on the other hand. As a positive side effect, CPU usage was reduced and processing time was significantly decreased by 68.54% compared to sequential processing (i.e., executing the system with each algorithm one by one).