International Journal of Advanced Technology and Engineering Exploration ISSN (Print): 2394-5443    ISSN (Online): 2394-7454 Volume-12 Issue-125 April-2025
  1. 4774
    Citations
  2. 2.8
    CiteScore
Enhanced forest fire detection using a hybrid VGG-16 and YOLOv5 model

Indira K1, Abinaya S2 and Aishwarya Shaji2

Department of Computer Science and Engineering,Thiagarajar College of Engineering, Madurai, Tamil Nadu,India1
School of Computer Science and Engineering,Vellore Institute of Technology, Chennai, Tamil Nadu,India2
Corresponding Author : Indira K

Recieved : 03-November-2023; Revised : 28-March-2025; Accepted : 08-April-2025

Abstract

Forests are an essential natural resource for humans, offering a plethora of direct and indirect advantages. The existence of life on Earth and the increasing effects of global warming are significantly impacted by forest fires and other natural disasters. Therefore, automatic forest fire detection is a crucial area of research for disaster prevention. Early detection plays a vital role in strategizing fire suppression efforts and mitigating potential damage. Wildfires exhibit diverse shapes, textures, and colors, making them challenging to detect using standard feature extraction techniques. This study examined the application of computer vision algorithms based on AI for smoke and fire detection in images. Convolutional neural networks (CNNs) demonstrated superior performance over traditional computer vision techniques, such as image classification. However, despite their effectiveness, CNNs required extensive training time, posing computational challenges in real-time fire detection scenarios. To solve this issue, a new hybrid visual geometry group 16 (VGG-16) layer network and the You Only Look Once Version 5 (YOLOv5) model are used for video-based fire detection. The image classifier VGG-16 is utilized in this instance to identify and categorize whether fire is present in the input data. From the categorized output of the VGG classifier, the object detector YOLOv5, is then applied to detect fire. A special effort has also been made to compile and preprocess the dataset, which includes images of fires and non-combustible objects, with as many real-world representative situations as possible. The test results demonstrated that the YOLOv5 detection algorithm has an accuracy of 70% and the VGG image classifier has an accuracy of more than 90%. The accuracy of YOLOv5 was greatly improved to 84% with reduced training time after the hybrid VGG-16-YOLOv5 model was developed. False positives (FP) were also reduced because background mistakes were removed even in videos containing smoke. According to experimental results, the suggested models perform better than the state-of-the-art techniques currently in use.

Keywords

Forest fire detection, Computer vision, Convolutional neural networks (CNNs), VGG-16-YOLOv5 hybrid model, Real-Time fire detection.

Cite this article

K I, S A, Shaji A. Enhanced forest fire detection using a hybrid VGG-16 and YOLOv5 model. International Journal of Advanced Technology and Engineering Exploration. 2025;12(125):670-687. DOI : 10.19101/IJATEE.2023.10102496

References
[1]
Wu Z, Li M, Wang B, Quan Y, Liu J. Using artificial intelligence to estimate the probability of forest fires in Heilongjiang, northeast China. Remote Sensing. 2021; 13(9):1-18.
[Crossref] [Google Scholar]
[2]
Lin X, Li Z, Chen W, Sun X, Gao D. Forest fire prediction based on long-and short-term time-series network. Forests. 2023; 14(4):1-18.
[Crossref] [Google Scholar]
[3]
Pechony O, Shindell DT. Driving forces of global wildfires over the past millennium and the forthcoming century. Proceedings of the National Academy of Sciences. 2010; 107(45):19167-70.
[Crossref] [Google Scholar]
[4]
Fried JS, Gilless JK, Riley WJ, Moody TJ, Simon DBC, Hayhoe K, et al. Predicting the effect of climate change on wildfire behavior and initial attack success. Climatic Change. 2008; 87:251-64.
[Crossref] [Google Scholar]
[5]
Yang S, Huang Q, Yu M. Advancements in remote sensing for active fire detection: a review of datasets and methods. Science of the Total Environment. 2024:173273.
[Crossref] [Google Scholar]
[6]
Fried JS, Torn MS, Mills E. The impact of climate change on wildfire severity: a regional forecast for northern California. Climatic Change. 2004; 64(1):169-91.
[Crossref] [Google Scholar]
[7]
Nagolu C, Cheekula C, Thota DS, Padmanaban K, Bhattacharyya D. Real-time forest fire detection using IoT and smart sensors. In international conference on inventive computation technologies 2023 (pp. 1441-7). IEEE.
[Crossref] [Google Scholar]
[8]
Allison RS, Johnston JM, Craig G, Jennings S. Airborne optical and thermal remote sensing for wildfire detection and monitoring. Sensors. 2016; 16(8):1-29.
[Crossref] [Google Scholar]
[9]
Preeti T, Kanakaraddi S, Beelagi A, Malagi S, Sudi A. Forest fire prediction using machine learning techniques. In international conference on intelligent technologies 2021 (pp. 1-6). IEEE.
[Crossref] [Google Scholar]
[10]
Abinaya S, Kumar KU, Alphonse AS. Cascading autoencoder with attention residual U-Net for multi-class plant leaf disease segmentation and classification. IEEE Access. 2023; 11:98153-70.
[Crossref] [Google Scholar]
[11]
Barmpoutis P, Papaioannou P, Dimitropoulos K, Grammalidis N. A review on early forest fire detection systems using optical remote sensing. Sensors. 2020; 20(22):1-25.
[Crossref] [Google Scholar]
[12]
Sakr GE, Elhajj IH, Mitri G. Efficient forest fire occurrence prediction for developing countries using two weather parameters. Engineering Applications of Artificial Intelligence. 2011; 24(5):888-94.
[Crossref] [Google Scholar]
[13]
Pradeep GS, Danumah JH, Nikhil S, Prasad MK, Patel N, Mammen PC, et al. Forest fire risk zone mapping of Eravikulam National Park in India: a comparison between frequency ratio and analytic hierarchy process methods. Croatian Journal of Forest Engineering: Journal for Theory and Application of Forestry Engineering. 2022; 43(1):199-217.
[Crossref] [Google Scholar]
[14]
Bowman DM, Williamson GJ. River flows are a reliable index of forest fire risk in the temperate tasmanian wilderness world heritage area, Australia. Fire. 2021; 4(2):1-12.
[Crossref] [Google Scholar]
[15]
Gülçin D, Deniz B. Remote sensing and GIS-based forest fire risk zone mapping: the case of Manisa, Turkey. Turkish Journal of Forestry. 2020; 21(1):15-24.
[Crossref] [Google Scholar]
[16]
Maffei C, Lindenbergh R, Menenti M. Combining multi-spectral and thermal remote sensing to predict forest fire characteristics. ISPRS Journal of Photogrammetry and Remote Sensing. 2021; 181:400-12.
[Crossref] [Google Scholar]
[17]
Jin R, Lee KS. Investigation of forest fire characteristics in North Korea using remote sensing data and GIS. Remote Sensing. 2022; 14(22):1-19.
[Crossref] [Google Scholar]
[18]
Cunningham AA, Martell DL. A stochastic model for the occurrence of man-caused forest fires. Canadian Journal of Forest Research. 1973; 3(2):282-7.
[Crossref] [Google Scholar]
[19]
Tian Y, Wu Z, Li M, Wang B, Zhang X. Forest fire spread monitoring and vegetation dynamics detection based on multi-source remote sensing images. Remote Sensing. 2022; 14(18):1-20.
[Crossref] [Google Scholar]
[20]
Shi S, Yao C, Wang S, Han W. A model design for risk assessment of line tripping caused by wildfires. Sensors. 2018; 18(6):1-15.
[Crossref] [Google Scholar]
[21]
Kalantar B, Ueda N, Idrees MO, Janizadeh S, Ahmadi K, Shabani F. Forest fire susceptibility prediction based on machine learning models with resampling algorithms on remote sensing data. Remote Sensing. 2020; 12(22):1-24.
[Crossref] [Google Scholar]
[22]
Ye Q, Huang P, Zhang Z, Zheng Y, Fu L, Yang W. Multiview learning with robust double-sided twin SVM. IEEE Transactions on Cybernetics. 2021; 52(12):12745-58.
[Crossref] [Google Scholar]
[23]
Dampage U, Bandaranayake L, Wanasinghe R, Kottahachchi K, Jayasanka B. Forest fire detection system using wireless sensor networks and machine learning. Scientific Reports. 2022; 12(1):1-11.
[Crossref] [Google Scholar]
[24]
Qiu J, Wang H, Shen W, Zhang Y, Su H, Li M. Quantifying forest fire and post-fire vegetation recovery in the Daxin’anling area of northeastern China using landsat time-series data and machine learning. Remote Sensing. 2021; 13(4):1-17.
[Crossref] [Google Scholar]
[25]
Xu R, Lin H, Lu K, Cao L, Liu Y. A forest fire detection system based on ensemble learning. Forests. 2021; 12(2):1-16.
[Crossref] [Google Scholar]
[26]
Tien BD, Le KT, Nguyen VC, Le HD, Revhaug I. Tropical forest fire susceptibility mapping at the Cat Ba National Park Area, Hai Phong City, Vietnam, using GIS-based kernel logistic regression. Remote Sensing. 2016; 8(4):1-15.
[Crossref] [Google Scholar]
[27]
Chang Y, Zhu Z, Bu R, Chen H, Feng Y, Li Y, et al. Predicting fire occurrence patterns with logistic regression in Heilongjiang Province, China. Landscape Ecology. 2013; 28:1989-2004.
[Crossref] [Google Scholar]
[28]
Bisquert M, Caselles E, Sánchez JM, Caselles V. Application of artificial neural networks and logistic regression to the prediction of forest fire danger in Galicia using MODIS data. International Journal of Wildland Fire. 2012; 21(8):1025-9.
[Crossref] [Google Scholar]
[29]
Oliveira S, Oehler F, San-miguel-ayanz J, Camia A, Pereira JM. Modeling spatial patterns of fire occurrence in mediterranean Europe using multiple regression and random forest. Forest Ecology and Management. 2012; 275:117-29.
[Crossref] [Google Scholar]
[30]
Satir O, Berberoglu S, Donmez C. Mapping regional forest fire probability using artificial neural network model in a Mediterranean forest ecosystem. Geomatics, Natural Hazards and Risk. 2016; 7(5):1645-58.
[Crossref] [Google Scholar]
[31]
Fan R, Pei M. Lightweight forest fire detection based on deep learning. In 31st international workshop on machine learning for signal processing 2021 (pp. 1-6). IEEE.
[Crossref] [Google Scholar]
[32]
Sathishkumar VE, Cho J, Subramanian M, Naren OS. Forest fire and smoke detection using deep learning-based learning without forgetting. Fire Ecology. 2023; 19(1):1-17.
[Crossref] [Google Scholar]
[33]
Bui DT, Bui QT, Nguyen QP, Pradhan B, Nampak H, Trinh PT. A hybrid artificial intelligence approach using GIS-based neural-fuzzy inference system and particle swarm optimization for forest fire susceptibility modeling at a tropical area. Agricultural and Forest Meteorology. 2017; 233:32-44.
[Crossref] [Google Scholar]
[34]
Kang Y, Jang E, Im J, Kwon C. A deep learning model using geostationary satellite data for forest fire detection with reduced detection latency. GIScience & Remote Sensing. 2022; 59(1):2019-35.
[Crossref] [Google Scholar]
[35]
Guo FT, Hu HQ, Ma ZH, Zhang Y. Applicability of different models in simulating the relationships between forest fire occurrence and weather factors in Daxing'an Mountains. The Journal of Applied Ecology. 2010; 21(1):159-64.
[Google Scholar]
[36]
Fu L, Zhang D, Ye Q. Recurrent thrifty attention network for remote sensing scene recognition. IEEE Transactions on Geoscience and Remote Sensing. 2020; 59(10):8257-68.
[Crossref] [Google Scholar]
[37]
Mohajane M, Costache R, Karimi F, Pham QB, Essahlaoui A, Nguyen H, et al. Application of remote sensing and machine learning algorithms for forest fire mapping in a Mediterranean area. Ecological Indicators. 2021; 129:107869.
[Crossref] [Google Scholar]
[38]
Guo F, Zhang L, Jin S, Tigabu M, Su Z, Wang W. Modeling anthropogenic fire occurrence in the boreal forest of China using logistic regression and random forests. Forests. 2016; 7(11):1-14.
[Crossref] [Google Scholar]
[39]
Natekar S, Patil S, Nair A, Roychowdhury S. Forest fire prediction using LSTM. In 2nd international conference for emerging technology 2021 (pp. 1-5). IEEE.
[Crossref] [Google Scholar]
[40]
Mohan KV, Satish AR, Rao KM, Yarava RK, Babu GC. Leveraging machine learning to predict wild fires. In 2nd international conference on smart electronics and communication 2021 (pp. 1393-400). IEEE.
[Crossref] [Google Scholar]
[41]
Jiang K, Chen L, Wang X, An F, Zhang H, Yun T. Simulation on different patterns of mobile laser scanning with extended application on solar beam illumination for forest plot. Forests. 2022; 13(12):1-18.
[Crossref] [Google Scholar]
[42]
Fu JJ, Wu ZW, Yan SJ, Zhang YJ, Gu XL, Du LH. Effects of climate, vegetation, and topography on spatial patterns of burn severity in the Great Xing’an Mountains. Acta Ecologica Sinica. 2020; 40:1672-82.
[Google Scholar]
[43]
Poirier F, Jaziri R, Srour C, Bernard G. Enhancing anomaly detection in videos using a combined YOLO and a VGG GRU approach. In 20th ACS/IEEE international conference on computer systems and applications 2023 (pp. 1-6). IEEE.
[Crossref] [Google Scholar]
[44]
Wang Z, Wu L, Li T, Shi P. A smoke detection model based on improved YOLOv5. Mathematics. 2022; 10(7):1-13.
[Crossref] [Google Scholar]
[45]
Vijayakumar A, Vairavasundaram S. Yolo-based object detection models: a review and its applications. Multimedia Tools and Applications. 2024; 83(35):83535-74.
[Crossref] [Google Scholar]
[46]
Abinaya S, Alphonse AS, Indira K, Karthiga S. Enhanced detection of brain tumour cells using visual geometry group. International Journal of Intelligent Systems and Applications in Engineering. 2022; 10(4):296-300.
[Google Scholar]
[47]
https://www.kaggle.com/datasets/brsdincer/wildfire-detection-image-data. Accessed 25 November 2024.
[48]
Abinaya S, Rajasenbagam T. Enhanced visual analytics technique for content-based medical image retrieval. International Journal of Electrical and Electronics Research. 2022; 10(2):93-9.
[Crossref] [Google Scholar]
[49]
Alphonse AS, Abinaya S, Abirami S. Alibaba and forty thieves algorithm and novel prioritized Prewitt pattern (PPP)-based convolutional neural network (CNN) using hyperspherically compressed weights for facial emotion recognition. Journal of Visual Communication and Image Representation. 2023; 97:103948.
[Crossref] [Google Scholar]
[50]
Abinaya S, Indira K, Karthiga S, Rajasenbagam T. Time cluster personalized ranking recommender system in multi-cloud. Mathematics. 2023; 11(6):1-17.
[Crossref] [Google Scholar]
[51]
Chen H, Liu H, Sun T, Lou H, Duan X, Bi L, et al. MC-YOLOv5: a multi-class small object detection algorithm. Biomimetics. 2023; 8(4):1-15.
[Crossref] [Google Scholar]