PERFORMANCE COMPARISON OF FASTER R-CONVOLUTIONAL NEURAL NETWORK (CNN) AND EFFICIENTNET FOR TRAIN DETECTION UNDER DIVERSE LIGHTING AND IMAGE QUALITY CONDITIONS

Suastika Yulia Riska; Achmad Noercholis

doi:10.52436/1.jutif.2024.5.6.3438

Suastika Yulia Riska Informatics Departement, Technology and Desain Faculty, Institut Teknologi dan Bisnis Asia, Malang, Indonesia
Achmad Noercholis Informatics Departement, Technology and Desain Faculty, Institut Teknologi dan Bisnis Asia, Malang, Indonesia

DOI: https://doi.org/10.52436/1.jutif.2024.5.6.3438

Keywords: Average Precision (AP), Average Recall (AR), EfficientNet, Faster R-CNN, Intersection over Union (IoU), Train Detection

Abstract

Object detection using computer vision has seen rapid advancements, especially with the advent of deep learning architectures such as Faster R-CNN and EfficientNet. This study compares the performance of the two models in detecting trains in various lighting conditions and noise disturbances. The dataset used consisted of 4500 train images which were divided into 70% for training, 20% for validation, and 10% for testing, reflecting real-world conditions. The evaluation was carried out using the Intersection over Union (IoU), Average Precision (AP), and Average Recall (AR) metrics. The results show that Faster R-CNN consistently excels in terms of detection accuracy, especially in less than ideal lighting conditions and under rain noise interference. In sufficient lighting conditions, Faster R-CNN showed a slightly superior AP score with a score of 0.844. As the lighting decreased, the difference between the two models became more pronounced, with Faster R-CNN recording an AP value of 0.810. In conditions with rain noise interference, the object detection performance of both models decreased more significantly, but the Faster R-CNN still excelled with an AP value of 0.798. Although EfficientNet is more efficient in terms of training speed, with a time of 5 hours and 37 minutes, and a smaller model size, Faster R-CNN shows higher reliability in complex environmental situations. This research provides important insights for the development of reliable and efficient train detection systems, taking into account the trade-off between resource efficiency and detection accuracy.

Downloads

Download data is not yet available.

References

T. Bai, J. Yang, G. Xu, and D. Yao, “An optimized railway fastener detection method based on modified Faster R-CNN,” Measurement, vol. 182, p. 109742, 2021.

G. Rachmad, Y. Yulianto, and M. Nurul, “PALANG PINTU PERLINTASAN KERETA API OTOMATIS YANG TERINTEGRASI BERBASIS DATA GPS,” Kohesi J. Sains dan Teknol., vol. 2, no. 9, pp. 1–10, 2024.

W. Li, “Analysis of object detection performance based on Faster R-CNN,” in Journal of Physics: Conference Series, 2021, vol. 1827, no. 1, p. 12085.

R. F. Alya, M. Wibowo, and P. Paradise, “Classification of Batik Motif Using Transfer Learning on Convolutional Neural Network (CNN),” J. Tek. Inform., vol. 4, no. 1, pp. 161–170, 2023, doi: 10.52436/1.jutif.2023.4.1.564.

A. Jiang, R. Noguchi, and T. Ahamed, “Tree trunk recognition in orchard autonomous operations under different light conditions using a thermal camera and faster R-CNN,” Sensors, vol. 22, no. 5, p. 2065, 2022.

K. Wang and M. Z. Liu, “Object recognition at night scene based on DCGAN and faster R-CNN,” IEEE Access, vol. 8, pp. 193168–193182, 2020.

S. Kalvankar, H. Pandit, and P. Parwate, “Galaxy morphology classification using efficientnet architectures,” arXiv Prepr. arXiv2008.13611, 2020.

L. T. Duong, T. B. Tran, N. H. Le, V. M. Ngo, and P. T. Nguyen, “Automatic detection of weeds: synergy between EfficientNet and transfer learning to enhance the prediction accuracy,” Soft Comput., vol. 28, no. 6, pp. 5029–5044, 2024.

M. Dian, L. Yudha, W. Setiawan, and Y. Wihardi, “Deteksi Sepeda Motor di Jalan Raya Menggunakan Faster R-CNN Berbasis VGG16 Detecting Motorcycle at Traffic Roads with Faster VGG16 based R-CNN,” JATIKOM J. Apl. dan Teor. Ilmu Komput., vol. 4, no. 2, pp. 10–13, 2021, [Online]. Available: https://ejournal.upi.edu/index.php/JATIKOM.

J. Pardede and H. Hardiansah, “Deteksi Objek Kereta Api menggunakan Metode Faster R-CNN dengan Arsitektur VGG 16,” MIND (Multimedia Artif. Intell. Netw. Database) J. MIND, vol. 7, no. 1, pp. 21–36, 2022.

G. I. Andaru, “Pengembangan Model Deteksi Untuk On-shelf Availability Produk menggunakan Yolov8 Pada Aplikasi Bergerak.” Universitas Islam Indonesia, 2024.

M. Harahap and A. M. Husein, “Peneraparan Efficient-Net Dalam Mengklasifikasi Kanker Kulit,” Publ. BUKU UNPRI Press ISBN, vol. 1, no. 1, 2024.

C. B. LIMBOING, “IMPLEMENTASI DEEP LEARNING UNTUK MENGANALISA BERBAGAI MACAM DAUN.” Universitas Mercu Buana Jakarta, 2022.

R. Firmansyah, “Implementasi deep learning menggunakan convolutional neural network untuk klasifikasi bunga.” Fakultas Sains dan Teknologi UIN Syarif Hidayatullah Jakarta, 2021.

X. Xu et al., “Crack detection and comparison study based on faster R-CNN and mask R-CNN,” Sensors, vol. 22, no. 3, p. 1215, 2022.

R. Budi, R. A. Harianto, and E. Setyati, “Segmentasi Citra Area Tumpukan Sampah Dengan Memanfaatkan Mask R-CNN,” INSYST J. Intell. Syst. Comput., vol. 5, no. 1, pp. 58–64, 2023.

R. A. Firmansah, H. Santoso, and A. Anwar, “Transfer Learning Implementation on Image Recognition of Indonesian Traditional Houses,” J. Tek. Inform., vol. 4, no. 6, pp. 1469–1478, 2023.

I. R. Fadhillah, M. M. Al Haromainy, and H. Maulana, “Implementasi Model Transfer Learning EfficientNet untuk Pendeteksian Bahasa Isyarat Indonesia (BISINDO) pada Perangkat Android,” JATI (Jurnal Mhs. Tek. Inform., vol. 8, no. 4, pp. 7816–7822, 2024.

F. Charli, H. Syaputra, M. Akbar, S. Sauda, and F. Panjaitan, “Implementasi Metode Faster Region Convolutional Neural Network (Faster R-CNN) Untuk Pengenalan Jenis Burung Lovebird,” J. Inf. Technol. Ampera, vol. 1, no. 3, pp. 185–197, 2020, doi: 10.51519/journalita.volume1.isssue3.year2020.page185-197.

Y. Mardianto, T. Dewi, and P. Risma, “Analisis Klasifikasi Kematangan Buah Tomat dengan Pendekatan Transfer Learning Model EfficientNet,” Techno Bahari, vol. 11, no. 1, pp. 20–25, 2024, doi: 10.52234/tb.v11i1.306.

D. P. Sidik, F. Utaminingrum, and L. Muflikhah, “Penggunaan Variasi Model pada Arsitektur EfficientNetV2 untuk Prediksi Sel Kanker Serviks,” … Teknol. Inf. dan Ilmu …, vol. 7, no. 5, pp. 2116–2121, 2023, [Online]. Available: https://j-ptiik.ub.ac.id/index.php/j-ptiik/article/view/12656.

A. Ganang Pradana, D. Rosal, I. M. Setiadi, and A. R. Muslikh, “Fine tuning model Convolutional Neural Network EfficientNet-B4 dengan augmentasi data untuk klasifikasi penyakit kakao Fine tuning Convolutional Neural Network model EfficientNet-B4 with data augmentation for cocoa disease classification,” J. Inf. Syst. Appl. Dev., vol. 2, no. 1, pp. 1–11, 2024, doi: 10.26905/jisad.v2i1.11899.

M. A. J. Asshiddiqie, B. Rahmat, and F. T. Anggraeny, “Deteksi tanaman tebu pada lahan pertanian menggunakan metode convolutional neural network,” J. Inform. dan Sist. Inf., vol. 1, no. 1, pp. 229–237, 2020.

R. S. Passa, S. Nurmaini, and D. P. Rini, “DETEKSI TUMOR OTAK PADA MAGNETIC RESONANCE IMAGING MENGGUNAKAN YOLOv7,” J. Ilm. MATRIK, vol. 25, no. 2, pp. 116–121, 2023.

Ü. Atila, M. Uçar, K. Akyol, and E. Uçar, “Plant leaf disease classification using EfficientNet deep learning model,” Ecol. Inform., vol. 61, p. 101182, 2021, doi: https://doi.org/10.1016/j.ecoinf.2020.101182.

D. G. Manurung et al., “Deteksi Dan Klasifikasi Hama Potato Beetle Pada Tanaman Kentang Menggunakan YOLOV8,” J. Teknol. Inf. dan Ilmu Komput., vol. 11, no. 4, pp. 723–734, 2024.

G. R. H. Galib, “Klasifikasi Area Tutupan Lahan Vegetasi Menggunakan Convolutional Neural Network.” Universitas Islam Negeri Maulana Malik Ibrahim, 2024.