Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection images
In recent years, the adoption of advanced machine learning techniques has revolutionized approaches to solving complex problems, such as identifying occurrences of forest fires. Among these techniques, the use of Convolutional Neural Networks (CNNs) combined with ensemble methods is particularl...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Growing Science
2025-01-01
|
| Series: | International Journal of Data and Network Science |
| Online Access: | http://www.growingscience.com/ijds/Vol9/ijdns_2024_193.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850057045724626944 |
|---|---|
| author | Dima Suleiman Ruba Obiedat Rizik Al-Sayyed Shadi Saleh Wolfram Hardt Yazan Al-Zain |
| author_facet | Dima Suleiman Ruba Obiedat Rizik Al-Sayyed Shadi Saleh Wolfram Hardt Yazan Al-Zain |
| author_sort | Dima Suleiman |
| collection | DOAJ |
| description |
In recent years, the adoption of advanced machine learning techniques has revolutionized approaches to solving complex problems, such as identifying occurrences of forest fires. Among these techniques, the use of Convolutional Neural Networks (CNNs) combined with ensemble methods is particularly promising. To investigate the feasibility of detecting fires using video streams from Unmanned Aerial Vehicles (UAVs), the lightweight CNN architecture MobileNetV2 was utilized for real-time detection. Several experiments were conducted on the DeepFire dataset, which comprises an equal number of images with and without fire, to evaluate MobileNetV2's performance. Notably, the architecture's linear bottlenecks and the efficient use of inverted residuals ensure high accuracy without compromising on feature extraction capabilities. For a comprehensive assessment, MobileNetV2 was benchmarked against other models, including DenseNet121, EfficientNetV2S, and VGG16. Accuracy was enhanced by averaging predictions through methods such as voting or summing results. As documented in the literature, MobileNetV2 consistently outperforms other architectures in computational efficiency and provides an excellent balance between efficiency and the quality of learned features over multiple epochs. This study underscores the suitability of MobileNetV2 for real-time applications on drones, particularly for the detection of forest fires in resource-constrained environments. The results show that MobileNetV2 achieves the highest accuracy (0.994), sensitivity (0.994), and specificity (0.998) among the tested models, with low standard deviations across all metrics. In contrast, EfficientNetV2S exhibited the lowest accuracy and sensitivity, both at 0.779, with a specificity of 0.829. The ensemble (Sum) method achieved an average accuracy of 0.989, sensitivity of 0.989, and specificity of approximately 0.988. Therefore, MobileNetV2 not only delivers the highest accuracy and stability but also demonstrates that the choice of ensemble method significantly affects the results. |
| format | Article |
| id | doaj-art-de7b8f4e720b40dd97458ee758e5dd57 |
| institution | DOAJ |
| issn | 2561-8148 2561-8156 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Growing Science |
| record_format | Article |
| series | International Journal of Data and Network Science |
| spelling | doaj-art-de7b8f4e720b40dd97458ee758e5dd572025-08-20T02:51:31ZengGrowing ScienceInternational Journal of Data and Network Science2561-81482561-81562025-01-019229731610.5267/j.ijdns.2024.10.004Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection imagesDima SuleimanRuba ObiedatRizik Al-SayyedShadi SalehWolfram HardtYazan Al-Zain In recent years, the adoption of advanced machine learning techniques has revolutionized approaches to solving complex problems, such as identifying occurrences of forest fires. Among these techniques, the use of Convolutional Neural Networks (CNNs) combined with ensemble methods is particularly promising. To investigate the feasibility of detecting fires using video streams from Unmanned Aerial Vehicles (UAVs), the lightweight CNN architecture MobileNetV2 was utilized for real-time detection. Several experiments were conducted on the DeepFire dataset, which comprises an equal number of images with and without fire, to evaluate MobileNetV2's performance. Notably, the architecture's linear bottlenecks and the efficient use of inverted residuals ensure high accuracy without compromising on feature extraction capabilities. For a comprehensive assessment, MobileNetV2 was benchmarked against other models, including DenseNet121, EfficientNetV2S, and VGG16. Accuracy was enhanced by averaging predictions through methods such as voting or summing results. As documented in the literature, MobileNetV2 consistently outperforms other architectures in computational efficiency and provides an excellent balance between efficiency and the quality of learned features over multiple epochs. This study underscores the suitability of MobileNetV2 for real-time applications on drones, particularly for the detection of forest fires in resource-constrained environments. The results show that MobileNetV2 achieves the highest accuracy (0.994), sensitivity (0.994), and specificity (0.998) among the tested models, with low standard deviations across all metrics. In contrast, EfficientNetV2S exhibited the lowest accuracy and sensitivity, both at 0.779, with a specificity of 0.829. The ensemble (Sum) method achieved an average accuracy of 0.989, sensitivity of 0.989, and specificity of approximately 0.988. Therefore, MobileNetV2 not only delivers the highest accuracy and stability but also demonstrates that the choice of ensemble method significantly affects the results.http://www.growingscience.com/ijds/Vol9/ijdns_2024_193.pdf |
| spellingShingle | Dima Suleiman Ruba Obiedat Rizik Al-Sayyed Shadi Saleh Wolfram Hardt Yazan Al-Zain Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection images International Journal of Data and Network Science |
| title | Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection images |
| title_full | Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection images |
| title_fullStr | Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection images |
| title_full_unstemmed | Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection images |
| title_short | Employing CNN mobileNetV2 and ensemble models in classifying drones forest fire detection images |
| title_sort | employing cnn mobilenetv2 and ensemble models in classifying drones forest fire detection images |
| url | http://www.growingscience.com/ijds/Vol9/ijdns_2024_193.pdf |
| work_keys_str_mv | AT dimasuleiman employingcnnmobilenetv2andensemblemodelsinclassifyingdronesforestfiredetectionimages AT rubaobiedat employingcnnmobilenetv2andensemblemodelsinclassifyingdronesforestfiredetectionimages AT rizikalsayyed employingcnnmobilenetv2andensemblemodelsinclassifyingdronesforestfiredetectionimages AT shadisaleh employingcnnmobilenetv2andensemblemodelsinclassifyingdronesforestfiredetectionimages AT wolframhardt employingcnnmobilenetv2andensemblemodelsinclassifyingdronesforestfiredetectionimages AT yazanalzain employingcnnmobilenetv2andensemblemodelsinclassifyingdronesforestfiredetectionimages |