Klasifikasi Jenis Kendaraan Darat di Indonesia Menggunakan Pendekatan Deep Learning
Submission Date: 2024-02-15 22:33:30
Accepted Date: 2024-11-25 12:12:18
Abstract
Proses klasifikasi jenis kendaraan dengan menggunakan pendekatan Deep Learning dalam hal ini menggunakan metode Convolutional Neural Networks (CNN), ada beberapa masalah yang muncul yaitu ketergantungan pada data pelatihan yang besar dan generalisasi data, CNN memerlukan dataset pelatihan yang besar dan bervariasi untuk melakukan pelatihan yang efektif. Jumlah data pelatihan yang terbatas dapat menyebabkan overfitting, di mana model menjadi terlalu beradaptasi dengan data pelatihan dan tidak dapat melakukan generalisasi dengan baik pada data yang belum pernah dilihat sebelumnya, sehingga membuat proses klasifikasi jenis kendaraan tidak maksimal. Penelitian ini bertujuan untuk melakukan klasifikasi jenis kendaraan darat di Indonesia ke dalam 5 kelas yaitu Bus, Minibus, Sedan, SUV, dan Truk menggunakan 5 (lima) jenis model arsitektur CNN yaitu ResNet50V2, MobileNetV2, InceptionV3, Xception, dan InceptionResNetV2, dimana dataset yang digunakan bersumber dari internet sebanyak 750 citra kendaraan melalui pengumpulan dataset dengan menggunakan teknik web scraping. Hasil eksperimen menunjukkan bahwa model terbaik didapatkan oleh ResNet50V2 pada skenario uji coba K-Fold dengan Augmentasi diperoleh accuracy sebesar 0,9906, precision sebesar 0,9916, recall sebesar 0,9907, dan F1-score sebesar 0,9911. Model terbaik diuji cobakan pada data baru berupa rekaman video yang diperoleh dari CCTV Gate Institut Teknologi Sepuluh Nopember (ITS). Hasil akhir akurasi yang didapatkan oleh model ResNet50V2 adalah sebesar 0,65.
Keywords
CNN; Deep Learning; Jenis Kendaraan Darat; Klasifikasi; YOLOv3-Tiny
CC Licencing
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
Refbacks
- There are currently no refbacks.
Jurnal Teknik ITS by Direktorat Riset dan Pengabdian Masyarakat (DRPM) ITS is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Based on a work at https://ejurnal.its.ac.id/index.php/teknik.
Statistik Pengunjung