RAMADHAN, FAHREZA SYAHRUL (2026) IMPLEMENTING CONVOLUTIONAL NEURAL NETWORK BASED ON SOBEL EDGE DETECTION IN CAR TIRE WEAR DETECTION. Tugas Akhir thesis, University of Technology Yogyakarta.
![[img]](http://eprints.uty.ac.id/style/images/fileicons/text.png) |
Text
5220411407_Fahreza Syahrul Ramadhan_Abstrak.pdf Download (142kB) |
Abstract
ABSTRACT Car tires are a critical component affecting the safety of drivers and passengers. Undetected tire wear can increase the risk of accidents due to vehicle instability. The process of assessing tire wear levels, typically performed through visual inspection, is often inconsistent and may yield inaccurate results. Therefore, this study aims to enhance an automatic detection system capable of objectively and accurately calculating the level of car tire wear using Convolutional Neural Network (CNN) technology based on edge detection methods. The research methodology includes data collection, labeling tire image datasets (Safety/Warning), preprocessing (grayscale conversion, noise reduction, resizing), and Sobel edge extraction; training a CNN model; model evaluation using random and stratified sampling with a confusion matrix; integration of the model into the DigiTire web application; and finally, Black Box functional testing and User Acceptance Testing (UAT) to assess usability and user satisfaction. The results of training and testing the CNN model demonstrated strong performance with the random sampling method: training accuracy reached 99.15%, and test accuracy was 86.62% using 90% for training and 10% for testing. Functional testing confirmed that all features operated as intended, while the UAT yielded a score of 86.0%, considered excellent. This research is expected to contribute to the development of advanced detection systems applicable in the automotive industry to improve vehicle maintenance and road safety. Keywords: Tire Wear Detection, CNN, Edge Detection
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > T Technology (General) > T201 Patents. Trademarks |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Informatika |
| Depositing User: | Kaprodi S1 Informatika UTY |
| Date Deposited: | 07 May 2026 07:27 |
| Last Modified: | 07 May 2026 07:27 |
| URI: | http://eprints.uty.ac.id/id/eprint/19842 |
Actions (login required)
 |
View Item |
