ACADSTAFF UGM
| Title | : | Real-Time Forest Fire Detection Framework Based on Artificial Intelligence using Color Probability Model and Motion Feature Analysis |
| Author | : |
Wahyono, Ph.D. (1) Prof. Drs. Agus Harjoko, M.Sc., Ph.D. (2) Dr. Andi Dharmawan, S.Si., M.Cs. (3) Faisal Dharma Adhinata (4) Gamma Kosala (5) Kang-Hyun Jo (6) |
| Date | : | 12 2022 |
| Keyword | : | color probability,motion feature analysis,forest fire,fire detection,intelligent surveillance system,real-time process,intersection over union color probability,motion feature analysis,forest fire,fire detection,intelligent surveillance system,real-time process,intersection over union |
| Abstract | : | As part of the early warning system, forest fire detection has a critical role in detecting fire in a forest area. In this case, the speed of the detection process is the most important factor to support a fast response by the authorities. Thus, this paper proposes a new framework for fire detection based on combining color-motion-shape features with machine learning technology. The characteristics of the fire are not only red but also from their irregular shape and movement that tends to be constant at certain locations. These characteristics are represented by color probabilities in the segmentation stage, color histograms in the classification stage, and image moments in the verification stage. A frame-based evaluation and an Intersection over Union (IoU) ratio will be applied to evaluate the proposed framework. Frame-based evaluation is used to measure the performance of the method in detecting fires. In contrast, the IoU ratio is used to measure the performance of the framework in localizing the fires. The experiment found that the proposed framework produces 89.97% and 10.04% in true positive rate and false-negative rate, respectively, when using the VisiFire dataset. Meanwhile, in terms of processing time, the proposed method is able to produce an average of 21.70 frames per second. These results proved that the proposed method is fast in the detection process and able to maintain performance accuracy. Thus, the proposed method is suitable and reliable to be integrated into the early warning system. |
| Group of Knowledge | : | Ilmu Komputer |
| Original Language | : | English |
| Level | : | Internasional |
| Status | : |
Published
|