Lawrence Owusu; Robert B Eshun; Leila Hashemi-Beni; Ali AlQahtani; Masud R Rashel; AKM K. Islam

Reduction of Misclassifications in Wildfire Detection: A Weighted Ensemble Deep Learning Approach

PDF (1176KB), PP.53-67

Views: 0 Downloads: 0

Author(s)

Lawrence Owusu ^1,* Robert B Eshun ¹ Leila Hashemi-Beni ² Ali AlQahtani ² Masud R Rashel ³ AKM K. Islam ¹

1. North Carolina Agricultural and Techincal State University/Department of Computational Data Science and Engineering, Greensboro, NC, 27411, USA

2. North Carolina Agricultural and Techincal State University/Department of Built Environment, Greensboro, NC, 27411, USA

3. University of Evora/Institute of Earth Sciences, Evora, 7002-554, Portugal

* Corresponding author.

DOI: https://doi.org/10.5815/ijisa.2024.05.05

Received: 26 Mar. 2024 / Revised: 17 May 2024 / Accepted: 25 Jul. 2024 / Published: 8 Oct. 2024

Index Terms

Ensemble Learning, Wildfire, Deep Learning, Forest Fire, Confusion Matrix, Classification

Abstract

Governments worldwide are increasingly prioritizing early wildfire detection to safeguard lives, property, and the environment. Although CNN-based models have demonstrated exceptional performance in various computer vision applications, the evolving nature of wildfire images poses significant challenges for a single CNN-based model in wildfire detection. In this study, we addressed this issue by integrating and weighting the differential learning capabilities of three individual transfer learning models: InceptionV3, ResNet50, and VGG16. Experimental results show that the ensemble deep learning models significantly outperformed all single classifiers across all performance metrics. Both the ensemble and weighted ensemble deep learning models achieved 99.7% accuracy, 99.5% precision, 100% recall, 99.8% F1-score, 0.5%false positive rate, 0.0% false negative rate and 0.3% error rate. Additionally, these models reduced the error rate by 98%, 91%, and 40% compared to the error rates of ResNet50, InceptionV3, and VGG16 respectively. A false negative rate of 0% indicates that our proposed ensemble deep learning models identified and predicted all the wildfire instances present in the test set correctly without a single misclassification. This positions our proposed ensemble deep learning models as superior choices for reducing misclassifications in wildfire detection.

Cite This Paper

Lawrence Owusu, Robert B Eshun, Leila Hashemi-Beni, Ali AlQahtani, Masud R Rashel, AKM K. Islam, "Reduction of Misclassifications in Wildfire Detection: A Weighted Ensemble Deep Learning Approach", International Journal of Intelligent Systems and Applications(IJISA), Vol.16, No.5, pp.53-67, 2024. DOI:10.5815/ijisa.2024.05.05

Reference

[1]M. Castelli, L. Vanneschi, and A. Popovič, “Predicting burned areas of forest fires: An artificial intelligence approach,” Fire Ecol., vol. 11, no. 1, pp. 106–118, 2015, doi: 10.4996/fireecology.1101106.
[2]K. Bot and J. G. Borges, “A Systematic Review of Applications of Machine Learning Techniques for Wildfire Management Decision Support,” Inventions, vol. 7, no. 1, 2022, doi: 10.3390/inventions7010015.
[3]A. S. Mahdi and S. A. Mahmood, “Analysis of Deep Learning Methods for Early Wildfire Detection Systems: Review,” IICETA 2022 - 5th Int. Conf. Eng. Technol. its Appl., no. October, pp. 271–276, 2022, doi: 10.1109/IICETA54559.2022.9888515.
[4]K. Dimitropoulos, O. Gunay, K. Kose, F. Erden, and F. Chaabene, “Flame Detection for Video-Based Early Fire Warning,” Prog. Cult. Herit. Preserv. 4th Int. Conf. EuroMed 2012, Limassol, Cyprus, Oct. 29 – Novemb. 3, 2012, pp. 378–387, 2012, [Online]. Available: http://link.springer.com/book/10.1007/978-3-642-34234-9/page/5.
[5]J. San-Miguel-Ayanz, N. Ravail, V. Kelha, and A. Ollero, “Active fire detection for fire emergency management: Potential and limitations for the operational use of remote sensing,” Nat. Hazards, vol. 35, no. 3, pp. 361–376, 2005, doi: 10.1007/s11069-004-1797-2.
[6]S. Dogan et al., “Automated accurate fire detection system using ensemble pretrained residual network,” Expert Syst. Appl., vol. 203, no. January, p. 117407, 2022, doi: 10.1016/j.eswa.2022.117407.
[7]P. Barmpoutis, P. Papaioannou, K. Dimitropoulos, and N. Grammalidis, “A review on early forest fire detection systems using optical remote sensing,” Sensors (Switzerland), vol. 20, no. 22, pp. 1–26, 2020, doi: 10.3390/s20226442.
[8]M. J. Sousa, A. Moutinho, and M. Almeida, “Wildfire detection using transfer learning on augmented datasets,” Expert Syst. Appl., vol. 142, p. 112975, 2020, doi: https://doi.org/10.1016/j.eswa.2019.112975.
[9]V. E. Sathishkumar, J. Cho, M. Subramanian, and O. S. Naren, “Forest fire and smoke detection using deep learning-based learning without forgetting,” Fire Ecol., vol. 19, no. 1, 2023, doi: 10.1186/s42408-022-00165-0.
[10]M. Iman, H. R. Arabnia, and K. Rasheed, “A Review of Deep Transfer Learning and Recent Advancements,” Technologies, vol. 11, no. 2, pp. 1–14, 2023, doi: 10.3390/technologies11020040.
[11]X. Gao, C. Shan, C. Hu, Z. Niu, and Z. Liu, “An Adaptive Ensemble Machine Learning Model for Intrusion Detection,” IEEE Access, vol. 7, pp. 82512–82521, 2019, doi: 10.1109/ACCESS.2019.2923640.
[12]R. Xu, H. Lin, K. Lu, L. Cao, and Y. Liu, “A forest fire detection system based on ensemble learning,” Forests, vol. 12, no. 2, pp. 1–17, 2021, doi: 10.3390/f12020217.
[13]A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Commun. ACM, vol. 60, no. 6, pp. 84–90, 2017, doi: 10.1145/3065386.
[14]J. Schmidhuber, “Deep learning in neural networks: An overview,” Neural Networks, vol. 61, pp. 85–117, 2015, doi: https://doi.org/10.1016/j.neunet.2014.09.003.
[15]K. Muhammad, J. Ahmad, Z. Lv, P. Bellavista, P. Yang, and S. W. Baik, “Efficient Deep CNN-Based Fire Detection and Localization in Video Surveillance Applications,” IEEE Trans. Syst. Man, Cybern. Syst., vol. 49, no. 7, pp. 1419–1434, 2019, doi: 10.1109/TSMC.2018.2830099.
[16]B. Kim and J. Lee, “A video-based fire detection using deep learning models,” Appl. Sci., vol. 9, no. 14, 2019, doi: 10.3390/app9142862.
[17]C. Bahhar et al., “Wildfire and Smoke Detection Using Staged YOLO Model and Ensemble CNN,” Electron., vol. 12, no. 1, pp. 1–15, 2023, doi: 10.3390/electronics12010228.
[18]kmalbek B. AbdusalomAov, B. M. S. Islam, R. Nasimov, M. Mukhiddinov, and T. K. Whangbo, “An Improved Forest Fire Detection Method Based on the Detectron2 Model and a Deep Learning Approach,” Sensors, vol. 23, no. 3, 2023, doi: 10.3390/s23031512.
[19]A. S. Almasoud, “Intelligent Deep Learning Enabled Wild Forest Fire Detection System,” Comput. Syst. Sci. Eng., vol. 44, no. 2, pp. 1485–1498, 2023, doi: 10.32604/csse.2023.025190.
[20]S. T. Seydi, V. Saeidi, B. Kalantar, N. Ueda, and A. A. Halin, “Fire-Net: A Deep Learning Framework for Active Forest Fire Detection,” J. Sensors, vol. 2022, 2022, doi: 10.1155/2022/8044390.
[21]A. Khan, B. Hassan, S. Khan, R. Ahmed, and A. Abuassba, “DeepFire: A Novel Dataset and Deep Transfer Learning Benchmark for Forest Fire Detection,” Mob. Inf. Syst., vol. 2022, 2022, doi: 10.1155/2022/5358359.
[22]A. V. Jonnalagadda and H. A. Hashim, “SegNet: A segmented deep learning based Convolutional Neural Network approach for drones wildfire detection,” Remote Sens. Appl. Soc. Environ., vol. 34, no. February, 2024, doi: 10.1016/j.rsase.2024.101181.
[23]S. Khan and A. Khan, “FFireNet: Deep Learning Based Forest Fire Classification and Detection in Smart Cities,” Symmetry (Basel)., vol. 14, no. 10, 2022, doi: 10.3390/sym14102155.
[24]J. Brownlee, “Develop Deep Learning Models On Theano And TensorFlow Using Keras,” J. Chem. Inf. Model., vol. 53, no. 9, pp. 1689–1699, 2019.
[25]D. Singh and B. Singh, “Investigating the impact of data normalization on classification performance,” Appl. Soft Comput., vol. 97, p. 105524, 2020, doi: 10.1016/j.asoc.2019.105524.
[26]V. N. G. Raju, K. P. Lakshmi, V. M. Jain, A. Kalidindi, and V. Padma, “Study the Influence of Normalization/Transformation process on the Accuracy of Supervised Classification,” Proc. 3rd Int. Conf. Smart Syst. Inven. Technol. ICSSIT 2020, no. Icssit, pp. 729–735, 2020, doi: 10.1109/ICSSIT48917.2020.9214160.
[27]K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 770–778, 2016, doi: 10.1109/CVPR.2016.90.
[28]N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, “Dropout: A simple way to prevent neural networks from overfitting,” J. Mach. Learn. Res., vol. 15, pp. 1929–1958, 2014.
[29]N. S. Shadin, S. Sanjana, and N. J. Lisa, “COVID-19 Diagnosis from Chest X-ray Images Using Convolutional Neural Network(CNN) and InceptionV3,” 2021 Int. Conf. Inf. Technol. ICIT 2021 - Proc., vol. 3, no. September 2012, pp. 799–804, 2021, doi: 10.1109/ICIT52682.2021.9491752.
[30]K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf. Track Proc., pp. 1–14, 2015.
[31]C. Lin, L. Li, W. Luo, K. C. P. Wang, and J. Guo, “Transfer learning based traffic sign recognition using inception-v3 model,” Period. Polytech. Transp. Eng., vol. 47, no. 3, pp. 242–250, 2019, doi: 10.3311/PPtr.11480.
[32]S. Tammina, “Transfer learning using VGG-16 with Deep Convolutional Neural Network for Classifying Images,” Int. J. Sci. Res. Publ., vol. 9, no. 10, p. p9420, 2019, doi: 10.29322/ijsrp.9.10.2019.p9420.
[33]Z. Zahisham, C. P. Lee, and K. M. Lim, “Food Recognition with ResNet-50,” IEEE Int. Conf. Artif. Intell. Eng. Technol. IICAIET 2020, pp. 0–4, 2020, doi: 10.1109/IICAIET49801.2020.9257825.
[34]J. Cao, M. Yan, Y. Jia, X. Tian, and Z. Zhang, “Application of a modified Inception-v3 model in the dynasty-based classification of ancient murals,” EURASIP J. Adv. Signal Process., vol. 2021, no. 1, 2021, doi: 10.1186/s13634-021-00740-8.
[35]A. F. Agarap, “Deep Learning using Rectified Linear Units (ReLU),” no. 1, pp. 2–8, 2018, [Online]. Available: http://arxiv.org/abs/1803.08375.
[36]X. Wang, Y. Zhong, L. Jin, and Y. Xiao, “Scale Adaptive Graph Convolutional Network for Skeleton-Based Action Recognition,” Tianjin Daxue Xuebao (Ziran Kexue yu Gongcheng Jishu Ban)/Journal Tianjin Univ. Sci. Technol., vol. 55, no. 3, pp. 306–312, 2022, doi: 10.11784/tdxbz202012073.
[37]D. Chicco and G. Jurman, “The Matthews correlation coefficient (MCC) should replace the ROC AUC as the standard metric for assessing binary classification,” BioData Min., vol. 16, no. 1, pp. 1–23, 2023, doi: 10.1186/s13040-023-00322-4.
[38]H. El Massari, Z. Sabouri, S. Mhammedi, and N. Gherabi, “Diabetes Prediction Using Machine Learning Algorithms and Ontology,” J. ICT Stand., vol. 10, no. 2, pp. 319–338, 2022, doi: 10.13052/jicts2245-800X.10212.
[39]C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the Inception Architecture for Computer Vision,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2016-Decem, pp. 2818–2826, 2016, doi: 10.1109/CVPR.2016.308.
[40]L. Zhang, M. Wang, Y. Fu, and Y. Ding, “A Forest Fire Recognition Method Using UAV Images Based on Transfer Learning,” Forests, vol. 13, no. 7, 2022, doi: 10.3390/f13070975.
[41]S. Treneska and B. R. Stojkoska, “Wildfire detection from UAV collected images using transfer learning,” no. August, 2021, [Online]. Available: https://www.researchgate.net/publication/353732371.
[42]M. Mukhiddinov, A. B. Abdusalomov, and J. Cho, “A Wildfire Smoke Detection System Using Unmanned Aerial Vehicle Images Based on the Optimized YOLOv5,” Sensors, vol. 22, no. 23, 2022, doi: 10.3390/s22239384.

International Journal of Intelligent Systems and Applications (IJISA)