Advanced Wildfire Detection Using
Deep Learning Algorithms: A
Comparative Study of CNN Variants
PROJECT GUIDE
Ms. O. Abhila Anju
Assistant Professor
PROJECT MEMBER
Vishali. K.S (22ADR122)
Yashica .S (22ADR123)
Yuvaraj. M (22ADR124)
PROBLEM STATEMENT
⮚ Wildfires cannot always be detected promptly using traditional optical
spectrum analysis technology due to obstacles in forested areas and
external interference, leading to high-speed spreading and casualties.
⮚ Early wildfire detection is challenging, especially in remote areas with
limited infrastructure, and there is a need for a more effective and timely
detection method.
OBJECTIVE
⮚ Develop a deep learning model for real-time wildfire detection.
⮚ Integrate satellite imagery, drones, and sensor networks for
enhanced surveillance.
⮚ Enable swift response coordination and early warning systems for
effective firefighting efforts.
LITERATURE SURVEY
Authors Title of the Paper Journal Name Year of Publication
Lidong Wang A Deep Learning-Based IEEE 2023
Experiment on Forest
Wildfire Detection in
Machine Vision Course.
Hoor Ul Ain Wildfire detection in aerial IEEE 2022
Tahir images using deep learning.
Dai Quoc Tran Forest-Fire Response System IEEE 2023
Using Deep-Learning-Based
Approaches With CCTV
Images and Weather Data
Rafik Ghali , Deep Learning and Sensors 2022
Moulay A. Transformer Approaches for
Akhloufi UAV-Based
Wildfire Detection and
Segmentation
EXISTING SYSTEM
⮚ The experimental results show that the proposed Reduce-VGGNet
model can reach 91.20% in accuracy, and the optimized CNN
model with the combination of spatial and temporal features can
reach 97.35% in accuracy.
⮚ On the FireNet and FLAME aerial picture datasets, we evaluated
the proposed method's performance and achieved the F1-score of
94.44%.
⮚ Classifying fire and no-fire photos, the proposed RBFN-RAISR
fire detection approach achieves 97.55% accuracy, 93.33%
F-Score, 96.44% recall, 94.19% precision, and an error rate of
24.89.
ALGORITHM USED
⮚ CNN
⮚ InceptionV3
⮚ DenseNet121
⮚ NASNetMobile
⮚ MobileNetV2
⮚ Xception
BLOCK DIAGRAM
Data
Collection
Algorithm 1.CNN
2.InceptionV3
3.DenseNet121
4.NASNetMobile
Evaluate 5.MobileNetV2
Models 6.Xception
Deploy
model
Test Images
Predicted
Output
RESULT
⮚ CNN NASNetMobile
Accuracy – 98.53% Accuracy – 98.40%
⮚ DenseNet121 InceptionV3
Accuracy - 95.59% Accuracy - 97.79%
MobileNetV2 Xception
Accuracy - 97.06% Accuracy - 98.32%
REFERENCES
⮚ Aytekin, E. Wildfires Ravaging Forestlands in Many Parts of Globe. 2021.
Available online: https://www.aa.com.tr/en/world/wildfires-ravaging-forestlands-
in-many-parts-of-globe/2322512 (accessed on 20 November 2021).
⮚ Canadian Wildland Fire Information System. Canada Wildfire Facts. 2021.
Available online: https://www.getprepared.gc.ca/cnt/hzd/wldfrs-en.aspx (accessed
on 20 November 2021).
⮚ Gaur, A.; Singh, A.; Kumar, A.; Kumar, A.; Kapoor, K. Video flame and smoke
based fire detection algorithms: A literature review.Fire Technol. 2020, 56, 1943–
1980. [CrossRef]
⮚ Chen, C.J.; Huang, Y.Y.; Li, Y.S.; Chang, C.Y.; Huang, Y.M. An AIoT Based
Smart Agricultural System for Pests Detection. IEEE Access 2020, 8, 180750–
180761. [CrossRef]
⮚ Giang, T.L.; Dang, K.B.; Toan Le, Q.; Nguyen, V.G.; Tong, S.S.; Pham, V.M. U-
Net Convolutional Networks for Mining Land Cover Classification Based on
High-Resolution UAV Imagery. IEEE Access 2020, 8, 186257–186273.
[CrossRef]
THANK YOU
