FERDINAND, ANDRE ABEL (2024) ANALYSIS OF YOLO V8 MODEL AND FUZZY ALGORITHM IN DETECTING VEHICLES, ASSESSING ROAD CONDITIONS AND CONTROL TRAFFIC LIGHTS. S1 thesis, UNIVERSITAS KATOLIK SOEGIJAPRANATA.
|
Text
21.K3.0001_ANDRE ABEL FERDINAND_COVER.pdf Download (526kB) | Preview |
|
![[img]](http://repository.unika.ac.id/style/images/fileicons/text.png) |
Text
21.K3.0001_ANDRE ABEL FERDINAND_BAB I.pdf Restricted to Registered users only Download (413kB) |
|
|
Text
21.K3.0001_ANDRE ABEL FERDINAND_BAB II.pdf Restricted to Registered users only Download (415kB) |
|
|
Text
21.K3.0001_ANDRE ABEL FERDINAND_BAB III.pdf Restricted to Registered users only Download (1MB) |
|
|
Text
21.K3.0001_ANDRE ABEL FERDINAND_BAB IV.pdf Restricted to Registered users only Download (1MB) |
|
|
Text
21.K3.0001_ANDRE ABEL FERDINAND_BAB V.pdf Restricted to Registered users only Download (530kB) |
|
|
Text
21.K3.0001_ANDRE ABEL FERDINAND_DAPUS.pdf Download (527kB) | Preview |
|
|
Text
21.K3.0001_ANDRE ABEL FERDINAND_LAMP.pdf Restricted to Registered users only Download (683kB) |
Abstract
This research presents an analysis of integrating object detection with traffic light control through YOLO and fuzzy logic systems to address increasing traffic congestion. The YOLO (You Only Look Once) algorithm is utilized to detect and classify vehicles in real-time, while fuzzy logic, particularly the Tsukamoto model, assesses traffic conditions and adjusts traffic light durations based on vehicle counts. The dataset from the Roboflow Universe titled "Top-View Vehicle Detection Image Dataset" was refined through image augmentation techniques such as grayscaling, contrast adjustment, and noise addition to train the YOLOv8 model. Among the YOLO variants tested, YOLOvL outperformed others with an accuracy of 0.7972, precision of 0.8736, F1-score of 0.8563, and recall of 0.8398, while YOLOvM showed stable performance. The integration of these technologies demonstrates a positive impact on traffic flow and urban congestion management. This study validates the effectiveness of combining YOLO and fuzzy logic for optimizing traffic light responses to varying traffic conditions.
| Item Type: | Thesis (S1) |
|---|---|
| Subjects: | 000 Computer Science, Information and General Works > 004 Data processing & computer science |
| Divisions: | Faculty of Computer Science > Department of Informatics Engineering |
| Depositing User: | ms. Wiwien Vieragustin |
| Date Deposited: | 11 Jul 2025 01:36 |
| Last Modified: | 11 Jul 2025 01:36 |
| URI: | http://repository.unika.ac.id/id/eprint/37423 |
| Keywords: | UNSPECIFIED |
Actions (login required)
 |
View Item |