A Comparative Evaluation of Drone Detection Models on Aerial Imageryacross Varying Training Epochs

Astika Ayuningtyas; Imam Riadi; Anton Yudhana

doi:10.30595/juita.v13i3.26618

Authors

Astika Ayuningtyas Institut Teknologi Dirgantara Adisutjipto
Imam Riadi Universitas Ahmad Dahlan
Anton Yudhana Universitas Ahmad Dahlan

DOI:

https://doi.org/10.30595/juita.v13i3.26618

Keywords:

drone detection, deep learning, YOLOv5, performance analysis

Abstract

Drone detection in aerial imagery has become increasingly important in security, surveillance, and military applications. This study aims to evaluate the performance of a deep learning model in detecting drone images by varying the number of training epochs (10, 20, and 50 epochs). A drone image dataset was used to train and test the model, with performance evaluated using precision, recall, [email protected], and [email protected]:0.95 metrics. The experimental results indicate that increasing the number of epochs significantly enhances model performance. At 10 epochs, the model achieved a precision of 0.905, recall of 0.857, [email protected] of 0.904, and [email protected]:0.95 of 0.455. At 20 epochs, recall improved to 0.879, and [email protected]:0.95 increased to 0.476. The best performance was observed at 50 epochs, with a precision of 0.918, recall of 0.886, [email protected] of 0.920, and [email protected]:0.95 of 0.494. These findings demonstrate that increasing the number of training epochs not only improves detection accuracy but also enhances the model's generalization capability. The study concludes that training for 50 epochs is the optimal configuration for achieving the best performance in drone image detection, despite requiring longer training time. These results provide practical recommendations for implementing deep learning models in real-world drone detection applications.

References

[1] R. Clarke and L. Bennett Moses, “The regulation of civilian drones’ impacts on public safety,” Computer Law & Security Review, vol. 30, no. 3, pp. 263–285, Jun. 2014, doi: 10.1016/j.clsr.2014.03.007.

[2] B. Kotkova, “Airport defense systems against drones attacks,” in 2022 26th International Conference on Circuits, Systems, Communications and Computers (CSCC), IEEE, 2022, pp. 85–90, doi: 10.1109/CSCC55931.2022.00025.

[3] J.-P. Yaacoub, H. Noura, O. Salman, and A. Chehab, “Security analysis of drones systems: Attacks, limitations, and recommendations,” Internet of Things, vol. 11, p. 100218, 2020, doi: 10.1016/j.iot.2020.100218.

[4] X. Zhang and K. Chandramouli, “Critical Infrastructure Security Against Drone Attacks Using Visual Analytics,” 2019, pp. 713–722. doi: 10.1007/978-3-030-34995-0_65.

[5] S. A. H. Mohsan, N. Q. H. Othman, Y. Li, M. H. Alsharif, and M. A. Khan, “Unmanned aerial vehicles (UAVs): Practical aspects, applications, open challenges, security issues, and future trends,” Intell Serv Robot, vol. 16, no. 1, pp. 109–137, 2023, doi: 10.1007/s11370-022-00452-4.

[6] A. A. Laghari, A. K. Jumani, R. A. Laghari, H. Li, S. Karim, and A. A. Khan, “Unmanned aerial vehicles advances in object detection and communication security review,” Cognitive Robotics, 2024, doi: 10.1016/j.cogr.2024.07.002.

[7] Z. Liu, P. An, Y. Yang, S. Qiu, Q. Liu, and X. Xu, “Vision-Based Drone Detection in Complex Environments: A Survey,” Drones, vol. 8, no. 11, p. 643, 2024, doi: 10.3390/drones8110643.

[8] D. Reis, J. Kupec, J. Hong, and A. Daoudi, “Real-time flying object detection with YOLOv8,” arXiv preprint arXiv:2305.09972, 2023.

[9] R. Umar, I. Riadi, and M. Miladiah, “Sistem Identifikasi Keaslian Uang Kertas Rupiah Menggunakan Metode K-Means Clustering,” Techno. Com, vol. 17, no. 2, pp. 179–185, 2018, doi: 10.33633/tc.v17i2.1681.

[10] W. Y. Sulistyo, I. Riadi, and A. Yudhana, “Analisis Deteksi Keaslian Citra Menggunakan Teknik Error Level Analysis Dengan Forensicallybeta,” in Seminar Nasional Informatika (SEMNASIF), 2018.

[11] S. Liu, W. Wang, L. Deng, and H. Xu, “Cnn-trans model: A parallel dual-branch network for fundus image classification,” Biomed Signal Process Control, vol. 96, p. 106621, 2024, doi: 10.1016/j.bspc.2024.106621.

[12] M. T. Ahad, Y. Li, B. Song, and T. Bhuiyan, “Comparison of CNN-based deep learning architectures for rice diseases classification,” Artificial Intelligence in Agriculture, vol. 9, pp. 22–35, 2023, doi: 10.1016/j.aiia.2023.07.001.

[13] S. Sunardi, A. Yudhana, and A. R. W. Putri, “Mass classification of breast cancer using CNN and faster R-CNN model comparison,” Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control, pp. 243–250, 2022, doi: 10.22219/kinetik.v7i3.1462.

[14] V. K. Perumal, T. Supriyaa, P. R. Santhosh, and S. Dhanasekaran, "CNN based plant disease identification using PYNQ FPGA," Systems and Soft Computing, vol. 6, p. 200088, 2024, doi: 10.1016/j.sasc.2024.200088.

[15] N. Utami, “Analysis for Detecting Banana Leaf Disease Using the CNN Method,” 2025.

[16] J. T. Santoso, E. Sediyono, K. D. Hartomo, and I. Sembiring, “Optimizing Attendance System: Integrating Liveness Detection and Deep Learning for Reliable Face Recognition,” 2024, doi: 10.30595/juita.v12i2.21738.

[17] A. F. Akbar, P. D. W. Ayu, and D. P. Hostiadi, “Performance Analysis of Deep,” 2025.

[18] H. Zhang, L. Zhang, and Y. Jiang, “Overfitting and underfitting analysis for deep learning based end-to-end communication systems,” in 2019 11th international conference on wireless communications and signal processing (WCSP), IEEE, 2019, pp. 1–6, doi: 10.1109/WCSP.2019.8927876.

[19] G. A. Saputra and I. M. A. Agastya, “Betta Fish Identification System Based On Convolutional Neural Network,” Journal of Applied Informatics and Computing, vol. 8, no. 2, pp. 443–452, 2024, doi: 10.30871/jaic.v8i2.8449.

[20] R. Moradi, R. Berangi, and B. Minaei, “A survey of regularization strategies for deep models,” Artif Intell Rev, vol. 53, no. 6, pp. 3947–3986, 2020, doi: 10.1007/s10462-019-09784-7.

[21] M. A. Arshad, S. H. Khan, S. Qamar, M. W. Khan, I. Murtza, J. Gwak, and A. Khan, "Drone navigation using region and edge exploitation-based deep CNN," IEEE Access, vol. 10, pp. 95441–95450, 2022, doi: 10.1109/ACCESS.2022.3204876.

[22] I. Riadi, A. Fadlil, and I. J. D. E. P. Putra, “Batik pattern classification using naïve bayes method based on texture feature extraction,” Khazanah Informatika: Jurnal Ilmu Komputer dan Informatika, vol. 9, no. 1, 2023, doi: 10.23917/khif.v9i1.21207.

[23] V. N. Kristanto, I. Riadi, and Y. Prayudi, “Analisa Deteksi dan Pengenalan Wajah pada Citra dengan Permasalahan Visual,” JISKA (Jurnal Informatika Sunan Kalijaga), vol. 8, no. 1, pp. 78–89, 2023, doi: 10.14421/jiska.2023.8.1.78-89.

[24] A. Rahman, M. Salim, and I. Riadi, “Klasifikasi Citra Spesies Bunga di Indonesia Berbasis Convolutional Neural Network Menggunakan Teknik Transfer Learning,” Jurnal Software Engineering and Computational Intelligence, vol. 2, no. 02, pp. 92–100, 2024, doi: 10.36982/jseci.v2i02.4942.

[25] M. F. Dzulqarnain, A. Fadlil, and I. Riadi, “Improving the Accuracy of Batik Classification using Deep Convolutional Auto Encoder,” Compiler, vol. 13, no. 2, pp. 123–130, 2024, doi: 10.28989/compiler.v13i2.2649.

[26] A. Ayuningtyas, I. Riadi, and A. Yudhana, “Comparison of Drone and Helicopter Image Classification Accuracy Using Naïve Bayes Based on Mean Red-Green-Blue (RGB) Values and First-Order Statistics,” International Journal of Informatics and Computation (IJICOM), vol. 7, no. 2, 2025, doi: https://doi.org/10.35842/ijicom.v7i2.144.

[27] C. Shorten and T. M. Khoshgoftaar, “A survey on image data augmentation for deep learning,” J Big Data, vol. 6, no. 1, pp. 1–48, 2019, doi: 10.1186/s40537-019-0197-0.

[28] S. Sapakova, A. Sapakov, and Y. Yilibule, “A YOLOv5-Based Model for Real-Time Mask Detection in Challenging Environments,” Procedia Comput Sci, vol. 231, pp. 267–274, 2024, doi: 10.1016/j.procs.2023.12.202.

[29] B. Cheng, B. Wang, Z. Liu, Q. Wang, T. Shen, and Y. Gu, “Airplane symmetry-based convolutional neural network for airplane detection in remote sensing image,” 2023, doi: 10.1049/icp.2024.1726.

[30] X. Tang, C. Ruan, X. Li, B. Li, and C. Fu, “MSC-YOLO: Improved YOLOv7 Based on Multi-Scale Spatial Context for Small Object Detection in UAV-View,” Computers, Materials and Continua, vol. 79, no. 1, pp. 983–1003, 2024, doi: 10.32604/cmc.2024.047541.

[31] C. Feng, C. Wang, D. Zhang, R. Kou, and Q. Fu, “Enhancing Dense Small Object Detection in UAV Images Based on Hybrid Transformer,” Computers, Materials and Continua, vol. 78, no. 3, pp. 3993–4013, 2024, doi: 10.32604/cmc.2024.048351.

[32] J. Redmon and A. Farhadi, “Yolov3: An incremental improvement,” arXiv preprint arXiv:1804.02767, 2018.

[33] L. Zhang, R. Xu, H. Ye, K. Wang, B. Xu, and D. Zhang, “High definition images transmission through single multimode fiber using deep learning and simulation speckles,” Opt Lasers Eng, vol. 140, p. 106531, 2021, doi: 10.1016/j.optlaseng.2021.106531.

[34] Z. Ren, H. Zhang, and Z. Li, “Improved YOLOv5 network for real-time object detection in vehicle-mounted camera capture scenarios,” Sensors, vol. 23, no. 10, p. 4589, 2023, doi: 10.3390/s23104589.

[35] T. Saidani, “Deep learning approach: YOLOv5-based custom object detection,” Engineering, Technology & Applied Science Research, vol. 13, no. 6, pp. 12158–12163, 2023, doi: 10.48084/etasr.6397.

[36] A. A. Alsuwaylimi, R. Alanazi, S. M. Alanazi, S. M. Alenezi, T. Saidani, and R. Ghodhbani, “Improved and efficient object detection algorithm based on yolov5,” Engineering, Technology & Applied Science Research, vol. 14, no. 3, pp. 14380–14386, 2024, doi: 10.48084/etasr.7386.

[37] C. Dewi and H. Juli Christanto, “Combination of deep cross-stage partial network and spatial pyramid pooling for automatic hand detection,” Big Data and Cognitive Computing, vol. 6, no. 3, p. 85, 2022, doi: 10.3390/bdcc6030085.

[38] B. Taha and A. Shoufan, “Machine learning-based drone detection and classification: State-of-the-art in research,” IEEE access, vol. 7, pp. 138669–138682, 2019, doi: 10.1109/ACCESS.2019.2942944.

[39] A. Balasundaram, A. Mohanty, A. Shaik, K. Pradeep, K. P. Vijayakumar, and M. S. Kavitha, “Zero-DCE++ Inspired Object Detection in Less Illuminated Environment Using Improved YOLOv5,” Computers, Materials and Continua, vol. 77, no. 3, pp. 2751–2769, 2023, doi: 10.32604/cmc.2023.044374.

[40] Z. Li and S. Arora, “An exponential learning rate schedule for deep learning,” arXiv preprint arXiv:1910.07454, 2019.

[41] M. M. H. Shuvo, S. K. Islam, J. Cheng, and B. I. Morshed, “Efficient acceleration of deep learning inference on resource-constrained edge devices: A review,” Proceedings of the IEEE, vol. 111, no. 1, pp. 42–91, 2022, doi: 10.1109/JPROC.2022.3226481.

[42] M. Reyad, A. M. Sarhan, and M. Arafa, “A modified Adam algorithm for deep neural network optimization,” Neural Comput Appl, vol. 35, no. 23, pp. 17095–17112, 2023, doi: 10.1007/s00521-023-08568-z.

[43] M. K. Anam, S. Defit, H. Haviluddin, L. Efrizoni, and M. B. Firdaus, “Early Stopping on CNN-LSTM Development to Improve Classification Performance,” Journal of Applied Data Sciences, vol. 5, no. 3, pp. 1175–1188, 2024, doi: 10.47738/jads.v5i3.312.

[44] H. Li, G. K. Rajbahadur, D. Lin, C.-P. Bezemer, and Z. M. Jiang, “Keeping deep learning models in check: A history-based approach to mitigate overfitting,” IEEE Access, 2024, doi: 10.1109/ACCESS.2024.3402543.

[45] M. Ş. Gündüz and G. Işık, “A new YOLO-based method for real-time crowd detection from video and performance analysis of YOLO models,” J Real Time Image Process, vol. 20, no. 1, p. 5, 2023, doi: 10.1007/s11554-023-01276-w.

[46] T. Diwan, G. Anirudh, and J. V Tembhurne, “Object detection using YOLO: Challenges, architectural successors, datasets and applications,” Multimed Tools Appl, vol. 82, no. 6, pp. 9243–9275, 2023, doi: 10.1007/s11042-022-13644-y.

[47] J. Yang, K. Zhou, Y. Li, and Z. Liu, “Generalized out-of-distribution detection: A survey,” Int J Comput Vis, vol. 132, no. 12, pp. 5635–5662, 2024, doi: 10.1007/s11263-024-02117-4.

[48] S. Lakshmi and C. P. Maheswaran, “Effective deep learning based grade prediction system using gated recurrent unit (GRU) with feature optimization using analysis of variance (ANOVA),” Automatika: časopis za automatiku, mjerenje, elektroniku, računarstvo i komunikacije, vol. 65, no. 2, pp. 425–440, 2024, doi: 10.1080/00051144.2023.2296790.

[49] C. E. Agbangba, E. S. Aide, H. Honfo, and R. G. Kakai, “On the use of post-hoc tests in environmental and biological sciences: A critical review,” Heliyon, vol. 10, no. 3, 2024, doi: 10.1016/j.heliyon.2024.e25131.

[50] Y. Liu, A. Medlar, and D. Glowacka, “Statistically significant detection of semantic shifts using contextual word embeddings,” arXiv preprint arXiv:2104.03776, 2021, doi: 10.18653/v1/2021.eval4nlp-1.11.

[51] A.-L. Heikkinen, “Evaluation of cognitive complaints and neuropsychological performance in early-onset cognitive impairment and dementia,” 2024.

[52] Q. Wang, Z. Chen, Y. Wang, and H. Qu, “A survey on ML4VIS: Applying machine learning advances to data visualization,” IEEE Trans Vis Comput Graph, vol. 28, no. 12, pp. 5134–5153, 2021, doi: 10.1109/TVCG.2021.3106142.

[53] A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “Yolov4: Optimal speed and accuracy of object detection,” arXiv preprint arXiv:2004.10934, 2020.

[54] A. AyunYudhana, I. Riadi, and A. Yudhana, “Analisis Sistem Deteksi Citra untuk Optimalisasi Pengawasan Lalu Lintas Udara Menggunakan Algoritma YOLOv5,” 2025, doi: 10.14421/jiska.2025.10.3.364-376.