Kajian Pengaruh Dataset dan Bias Dataset terhadap Performa Akurasi Deteksi Objek

Main Article Content

Ridho Iman Tiyar
Dhomas Hatta Fudholi

Abstract

Deteksi objek merupakan kemampuan sistem yang dapat mengenali objek tertentu yang berada dalam suatu gambar atau video. Dalam proses deteksi objek, sistem bisa memberikan hasil yang tidak sesuai atau tidak dapat mendeteksi suatu objek yang disebabkan oleh dataset yang tidak optimal. Penelitian ini bertujuan mengkaji proses pembuatan dataset dan bias yang muncul. Pencarian dan analisis dilakukan terhadap literatur yang berkaitan dengan dataset deteksi objek. Proses pencarian literatur dilakukan pada Google Scholar, Science Direct, dan DSpace Repository dengan memasukkan kata kunci utama “data centric”, “object detection dataset”, dan “dataset bias”. Hasil analisis literatur meliputi dataset dan bias dataset. Pada penelitian sebelumnya terdapat kekurangan seperti belum adanya peningkatan performa sistem deteksi objek melalui pengoptimalan dataset. Dari kajian literatur, pembuatan dataset yang baik dapat dilakukan dengan cara menyesuaikan kondisi pengambilan gambar saat pengumpulan data dan pengujian di lapangan. Selain itu, untuk dapat menambah kemampuan generalisasi sistem dengan cara menambahkan variasi gambar dalam dataset melalui teknik augmentasi. Selanjutnya, dalam proses pembuatan dataset pasti akan selalu ada bias dalam data sehingga mempengaruhi kemampuan deteksi objek. Oleh karena itu, dalam proses pembuatan sistem deteksi objek, data memiliki pengaruh yang cukup besar terhadap performa akurasi deteksi objek.

Downloads

Download data is not yet available.

Article Details

How to Cite
Tiyar, R., & Fudholi, D. (2021). Kajian Pengaruh Dataset dan Bias Dataset terhadap Performa Akurasi Deteksi Objek. Petir, 14(2), 258 - 268. https://doi.org/10.33322/petir.v14i2.1350
Section
Articles

References

[1] R. A. Putro Eko Cahyo, Awangga Maulana Rolly, Tutorial Object Detection People With Faster region-Based Convolutional Neural Network(Faster R-CNN). Kreatif, 2020.
[2] E. C. Rahmad, S. T. M. Kom, D. Rawansyah, M. Pd, and T. K. Rochastu, “Mendeteksi Objek Sekitar untuk Penyandang,” pp. 81–88, 2018.
[3] S. Shao et al., “Objects365: A large-scale, high-quality dataset for object detection,” Proceedings of the IEEE International Conference on Computer Vision, vol. 2019-Octob, pp. 8429–8438, 2019, doi: 10.1109/ICCV.2019.00852.
[4] A. Khosla, T. Zhou, T. Malisiewicz, A. A. Efros, and A. Torralba, “Undoing the damage of dataset bias,” Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7572 LNCS, no. PART 1, pp. 158–171, 2012, doi: 10.1007/978-3-642-33718-5_12.
[5] S. Zheng, M. Hadi Kiapour, F. Yang, and R. Piramuthu, “ModaNet: A large-scale street fashion dataset with polygon annotations,” MM 2018 - Proceedings of the 2018 ACM Multimedia Conference, pp. 1670–1678, 2018, doi: 10.1145/3240508.3240652.
[6] P. de Tournemire, D. Nitti, E. Perot, D. Migliore, and A. Sironi, “A Large Scale Event-based Detection Dataset for Automotive,” 2020, [Online]. Available: http://arxiv.org/abs/2001.08499.
[7] G. Xia et al., “DOTA: A Large-Scale Dataset for Object Detection in Aerial Images | Enhanced Reader,” pp. 3974–3983, [Online]. Available: moz-extension://bd0c5e16-8147-45c5-ae8b-8401aaca2c83/enhanced-reader.html?openApp&pdf=http%3A%2F%2Fopenaccess.thecvf.com%2Fcontent_cvpr_2018%2Fpapers%2FXia_DOTA_A_Large-Scale_CVPR_2018_paper.pdf.
[8] A. Ziller, J. Hansjakob, V. Rusinov, D. Zügner, P. Vogel, and S. Günnemann, “Oktoberfest Food Dataset,” pp. 1–4, 2019, [Online]. Available: http://arxiv.org/abs/1912.05007.
[9] P. Spagnolo, F. Filieri, C. Distante, P. L. Mazzeo, and P. D’Ambrosio, “A new annotated dataset for boat detection and re-identification,” 2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2019, no. September, 2019, doi: 10.1109/AVSS.2019.8909831.
[10] A. Tourani, S. Soroori, A. Shahbahrami, and A. Akoushideh, “Iranis: A Large-scale Dataset of Farsi License Plate Characters,” 2021, [Online]. Available: http://arxiv.org/abs/2101.00295.
[11] H. Maeda, Y. Sekimoto, T. Seto, T. Kashiyama, and H. Omata, “Road Damage Detection Using Deep Neural Networks with Images Captured Through a Smartphone,” pp. 4–6, 2018, doi: 10.1111/mice.12387.
[12] P. S. Statistika, F. Matematika, D. A. N. Ilmu, P. Alam, and U. I. Indonesia, “IMPLEMENTASI DEEP LEARNING OBJECT DETECTION RAMBU K3 PADA VIDEO MENGGUNAKAN METODE CONVOLUTIONAL NEURAL NETWORK ( CNN ) DENGAN,” 2020.
[13] A. Tri Nurolan, “Deteksi Dan Klasifikasi Jenis Kendaraan Berbasis Pengolahan Citra Dengan Metode Convolutional Neural Network (Cnn),” 2019.
[14] R. Huang, J. Pedoeem, and C. Chen, “YOLO-LITE: A Real-Time Object Detection Algorithm Optimized for Non-GPU Computers,” Proceedings - 2018 IEEE International Conference on Big Data, Big Data 2018, pp. 2503–2510, 2019, doi: 10.1109/BigData.2018.8621865.
[15] A. Gupta, A. Murali, D. Gandhi, and L. Pinto, “Robot learning in homes: Improving generalization and reducing dataset bias,” Advances in Neural Information Processing Systems, vol. 2018-Decem, pp. 9094–9104, 2018.
[16] A. Kortylewski, B. Egger, A. Schneider, T. Gerig, A. Morel-Forster, and T. Vetter, “Empirically analyzing the effect of dataset biases on deep face recognition systems,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, vol. 2018-June, pp. 2174–2183, 2018, doi: 10.1109/CVPRW.2018.00283.
[17] T. Tommasi, N. Patricia, B. Caputo, and T. Tuytelaars, “A deeper look at dataset bias,” Advances in Computer Vision and Pattern Recognition, no. 9783319583464, pp. 37–55, 2017, doi: 10.1007/978-3-319-58347-1_2.
[18] A. Kortylewski et al., “Can Synthetic Faces Undo the Damage of Dataset Bias to Face Recognition and Facial Landmark Detection?,” 2018, [Online]. Available: http://arxiv.org/abs/1811.08565.
[19] E. Derman, “Dataset Bias Mitigation Through Analysis of CNN Training Scores.”

Most read articles by the same author(s)

Obs.: This plugin requires at least one statistics/report plugin to be enabled. If your statistics plugins provide more than one metric then please also select a main metric on the admin's site settings page and/or on the journal manager's settings pages.