Research Article | Published: 31 July 2025

Empirical analytics of baseline and enhanced CNN architectures with frequency and RGB features using Bayesian Hyperparameter Optimization for wildfire prediction

Devadhas  Crystal Jaba  Kani and Subash Saudia

Journal of Non-Timber Forest Products | Volume: 32 | Issue: 2 | Page No. 112-120 | 2025
DOI: https://doi.org/10.54207/bsmps2000-2025-845K9H | Cite this article

Abstract

­Wildfires have caused long-term economic, ecological, and biological damage, highlighting the need for accurate prediction systems to protect forest wildlife and valuable non-timber resources such as medicinal plants, aromatic products, food, fodder, and fuelwood. This study proposes wildfire prediction models using frequency-domain analytics and Bayesian Optimization (BSO) in designing Convolutional Neural Network (CNN)-based deep learning models, including LeNet-5, AlexNet, and VGG16, applied to the DeepFire dataset. The models are trained and tested on both RGB images and Fast Fourier Transform (FFT)-based frequency-domain representations of fire and non-fire images. BSO, integrated with the Tree-structured Parzen Estimator (TPE), optimizes model parameters to effectively extract fire-related features. Model performance is evaluated using Accuracy and AUC-ROC metrics. Results indicate that BSO-based frequency-aware modified LeNet-5 and AlexNet achieve accuracies of 97% and 96%, respectively. Additionally, RGB-based BSO enhances performance, with modified LeNet-5 and baseline VGG16 reaching up to 98% accuracy. Overall, findings demonstrate that frequency-domain features combined with BSO significantly improve wildfire prediction, supporting ecological and biological conservation efforts.

Keywords

AlexNet, Bayesian Optimization, FFT, LeNet5, TPE, VGG16, Wildfire

Access Options

250/-

Buy Full Access in HTML Format

Instant access to the full article.

Get access to the full version of this article. Buy Full Access in HTML Format

References

1. Akiba, T., Sano, S., Yanase, T., Ohta, T. and Koyama, M., 2019. Optuna: A Next-generation Hyperparameter Optimization Framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, pp.2623–2631.  https://doi.org/10.1145/3292500.3330701

Google Scholar

2. Anonymous, 2022. Ventura country multi-jurisdictional hazard mitigation plan. Jurisdictional Annexes.

Google Scholar

3. Ascoli, D., Moris, J.V., Marchetti, M. and Sallustio, L., 2021. Land use change towards forests and wooded land correlates with large and frequent wildfires in Italy. Annals of Silvicultural Research, pp.177–188.

Google Scholar

4. Bao, M., Liu, J., Ren, H., Liu, S., Ren, C., Chen, C. and Liu, J., 2024. Research Trends in Wildland Fire prediction amidst climate change: A Comprehensive Bibliometric analysis. Forests, 15(7), 1197.  https://doi.org/10.3390/f15071197

Google Scholar

5. Belgiu, M. and Csillik, O., 2018, Sentinel-2 cropland mapping using pixel-based and object-based time-weighted dynamic time warping analysis. Remote sensing of environment, vol. 204, pp.509–523.  https://doi.org/10.1016/j.rse.2017.10.005

Google Scholar

6. Bergstra, J., Bardenet, R., Bengio, Y. and Kégl, B., 2011. Algorithms for Hyper-Parameter Optimization. Advances in Neural Information Processing Systems, pp.2546–2554.

Google Scholar

7. Boer, M.M., Resco de Dios, V. and Bradstock, R.A., 2020. Unprecedented burn area of Australian mega forest fires. Nature Climate Change, pp.171–172.  https://doi.org/10.1038/s41558-020-0716-1

Google Scholar

8. Brigola, R., 2025. Discrete Fourier Transforms, First Applications. Fourier analysis and Distributions: A First Course with Applications. Cham: Springer Nature Switzerland, pp.85–128.  https://doi.org/10.1007/978-3-031-81311-5_6

9. Burley, J., 2002. Forest biological diversity: an overview. UNASYLVA-FAO. 3-9.

Google Scholar

10. Cochran, W.T., Cooley, J.W., Favin, D.L., Helms, H.D., Kaenel, R.A., Lang, W.W., Maling, G.C., Nelson, D.E., Rader, C.M. and Welch, P.D., 1967. What is the fast Fourier transform? Proceedings of the IEEE.1664-74.  https://doi.org/10.1109/PROC.1967.5957

Google Scholar

11. Cook, R.L., 1986. Stochastic sampling in computer graphics. ACM Transactions on Graphics (TOG), 5 (1), pp.51–72. https://doi.org/10.1145/7529.8927

Google Scholar

12. Dictionary, C., 2008. Cambridge advanced learner’s dictionary. Recuperado de.  https:// dictionary. cambridge. org/ es/ dictionary/ ingles/ blended- learning.

Google Scholar

13. Dhillon, A. and Gyanendra, K.V., 2020. Convolutional neural net–work: a review of models, methodologies and applications to object detection. Progress in Artificial Intelligence, 9, pp.85–112.  https://doi.org/10.1007/s13748-019-00203-0

Google Scholar

14. Ferri, C., Hernández-Orallo, J. and Flach, P.A., 2011. A coherent interpretation of AUC as a measure of aggregated classification performance. In Proceedings of the 28th International Conference on Machine Learning, pp.657–664.

Google Scholar

15. Ghibeche, Y., Sellam, A., Nouri, N., Khaldi, A., Harrane, A. and Ghibeche, I., 2024. Machine learning for forest fire prediction: a case study in North Algeria. Ingénierie des Systèmes d'Information, 29(1), 337-346.  https://doi.org/10.18280/isi.290133

Google Scholar

16. Ghosh, R. and Kumar, A., 2022. A hybrid deep learning model by combining convolutional neural network and recurrent neural network to detect forest fire. Multimedia Tools and Applications. 81(27), pp.38643–38660.  https://doi.org/10.1007/s11042-022-13068-8

Google Scholar

17. Gillespie, T.W., Chu, J., Frankenberg, E. and Thomas, D., 2007.   Assessment and prediction of natural hazards from satellite imagery. Progress in Physical Geography: Earth and Environment, 31(5), pp.459–470.

Google Scholar

18. Goodfellow, I., Bengio, Y. and Courville, A., 2016. Deep Learning. MIT Press.

Google Scholar

19. He, K., Zhang, X., Ren, S. and Sun, J.., 2016. Deep residual learning for image recognition. Proceedings of the IEEE conference on computer vision and pattern recognition.  https://doi.org/10.1109/CVPR.2016.90

Google Scholar

20. Kempa, D. and Prezza, N., 2018. At the roots of dictionary compression: string attractors. Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing, pp.827–840.  https://doi.org/10.1145/3188745.3188814

Google Scholar

21. Khan, A., Hassan, B., Khan, S., Ahmed, R. and Abuassba, A., 2022. DeepFire: A novel dataset and deep transfer learning benchmark for forest fire detection. Mobile Information Systems, 5358359.  https://doi.org/10.1155/2022/5358359

Google Scholar

22. Khessiba, S., Blaiech, A.G., Abdallah, A.B., Manzanera, A., Khalifa, K.B. and Bedoui, M.H., 2024. A Novel Hybrid Grid Search and Tree Parzen Estimator for Deep Learning Hyperparameters Optimization. 2024 IEEE/ACS 21st International Conference on Computer Systems and Applications (AICCSA). IEEE.  https://doi.org/10.1109/AICCSA63423.2024.10912622

Google Scholar

23. Kolluru, V.K., Chintakunta, A.N., Nuthakki, Y. and Koganti, S., 2022. Advancements in Wildfire Prediction and Detection: A Systematic Review. International Journal for Multidisciplinary Approach, 4(6).

Google Scholar

24. Kondekar, V.H. and Bodhe, S.K.,  2018. A  Comprehensive investigation of color models used in image processing.  International Journal of Computer Applications. pp.19–24.  https://doi.org/10.5120/ijca2018916507

Google Scholar

25. Krizhevsky, A., Sutskever, I. and Hinton, G.E., 2012. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems. 25

Google Scholar

26. LeCun, Y., Bottou, L., Bengio, Y. and Haffner, P., 1998. Gradient-Based Learning Applied to Document Recognition. Proceedings of the IEEE. pp.2278–2324.  https://doi.org/10.1109/5.726791

Google Scholar

27. Li, L., Jamieson, K., DeSalvo, G., Rostamizadeh, A., and Talwalkar, A., 2017. Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization. Journal of Machine Learning Research, pp.6765–6816.

Google Scholar

28. Li, Z., Wang, Y., Zhang, N., Zhang, Y., Zhao, Z., Xu, D., Ben, G. and Gao, Y., 2022. Deep Learning-Based Object Detection Techniques for Remote Sensing Images: A Survey. Remote Sensing, 14(10), 2385. https://doi.org/10.3390/rs14102385

Google Scholar

29. Pandey, K., 2022. Forest Survey Report 2021: Forest fire counts up 2.7 times. Forest fires in Uttarakhand in November 2020-June 2021 were 28.3 times more compared to November 2019-June 2020. Down To Earth. https://www.downtoearth.org.in/forests/forest-survey-report-2021-forest-fire-counts-up-2-7-times-81123

30. Peñuelas, J. and Sardans, J., 2021. Global change and forest disturbances in the Mediterranean basin: Breakthroughs, knowledge gaps, and recommendations. Forests. 12(5), 603.  https://doi.org/10.3390/f12050603

Google Scholar

31. Porta, H., Dalsasso, E., McCarty, J.L. and Tuia, D., 2025. CanadaFireSat: Toward high-resolution wildfire forecasting with multiple modalities. arXiv preprint arXiv:2506.08690.

Google Scholar

32. Raiaan, M.A.K., Sakib, S., Fahad, N.M., Al Mamun, A., Rahman, M.A., Shatabda, S. and Mukta, M.S.H., 2024. A systematic review of hyperparameter optimization techniques in Convolutional Neural Networks. Decision analytics journal, 11, p.100470.  https://doi.org/10.1016/j.dajour.2024.100470

Google Scholar

33. Rao, K.R., Kim, D.N. and Hwang, J.J.  2011. Fast Fourier transform-algorithms and applications. Springer Science & Business Media. https://doi.org/10.1007/978-1-4020-6629-0

Google Scholar

34. Sadowska, B., Grzegorz, Z. and Stępnicka, N., 2021. Forest fires and losses caused by fires–an economic approach. WSEAS Trans Environ Dev, 17, pp.181–191.  https://doi.org/10.37394/232015.2021.17.18

Google Scholar

35. Simonyan, K. and Zisserman, A., 2015. Very Deep Convolutional Networks for Large-Scale Image Recognition. International Conference on Learning Representations (ICLR).

Google Scholar

36. Stipaničev, D., Vuko, T., Krstinić, D., Štula, M. and Bodrožić, L., 2006.  Forest fire protection by advanced video detection system-Croatian experiences. Third TIEMS Workshop-Improvement of Disaster Management System, Trogir, pp.26–27.

Google Scholar

37. Sweeney, C., Ennis, E., Mulvenna, M., Bond, R. and O’Neill, S., 2022. How machine learning classification accuracy changes in a happiness dataset with different demographic groups. Computers. 11(5) 83.  https://doi.org/10.3390/computers11050083

Google Scholar

38. Tammina, S., 2019. Transfer learning using vgg-16 with deep convolutional neural network for classifying images. International Journal of Scientific and Research Publications (IJSRP), pp.143–150.  https://doi.org/10.29322/IJSRP.9.10.2019.p9420

Google Scholar

39. Üstek, İ., Arana‐Catania, M., Farr, A. and Petrunin, I.,  2024. Deep autoencoders for unsupervised anomaly detection in wildfire prediction. Earth and Space Science, 11(11), p.e2024EA003997.  https://doi.org/10.1029/2024EA003997

Google Scholar

40. Willman, J., 2021. GUIs for computer vision. Modern PyQt: Create GUI Applications for Project Management, Computer Vision, and Data Analysis, pp.163–208.  https://doi.org/10.1007/978-1-4842-6603-8_5

Google Scholar

41. Xavier, V.D., 1998. Forest fire propagation. Philosophical Transactions of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, pp. 2907–2928.  https://doi.org/10.1098/rsta.1998.0303

Google Scholar

42. Xu, K.,  Qin, M. , Sun, F. , Wang, Y. , Chen, Y. -K.  and Ren, F. 2020. Learning in the frequency domain.  Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR), Seattle, WA, USA,.  https://doi.org/10.1109/CVPR42600.2020.00181

Google Scholar

43. Zhang, Z.H., Yang, Z., Sun, Y., Wu, Y.F. and Xing, Y.D., 2019. Lenet-5 convolution neural network with mish activation function and fixed memory step gradient descent method. In 2019 16th international computer conference on    wavelet active media technology and information processing.196–199. IEEE.  https://doi.org/10.1109/ICCWAMTIP47768.2019.9067661

Google Scholar

44. Zhu, M., and Gupta, S., 2017. To prune, or not to prune: exploring the efficacy of pruning for model compression. arXiv preprint arXiv:1710.01878.

Google Scholar

About this article

How to cite

Kani, D.C.J. and Saudia, S., 2025. Empirical analytics of baseline and enhanced CNN architectures with frequency and RGB features using Bayesian Hyperparameter Optimization for wildfire prediction. Journal of Non-Timber Forest Products, 32(2), pp.112-120. https://doi.org/10.54207/bsmps2000-2025-845K9H

Publication History

Manuscript Received on 02 July 2025

Manuscript Revised on 25 July 2025

Manuscript Accepted on 28 July 2025

Manuscript Published on 31 July 2025

Share this article

Anyone you share the following link with will be able to read this content: