แจ้งเอกสารไม่ครบถ้วน, ไม่ตรงกับชื่อเรื่อง หรือมีข้อผิดพลาดเกี่ยวกับเอกสาร ติดต่อที่นี่ ==>
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
Pintusorn Suttiponpisarn. An autonomous framework for real-time wrong-way driving vehicle detection from CCTV. Master's Degree(Artificial Intelligence and Internet of Things). Thammasat University. Thammasat University Library. : Thammasat University, 2021.
An autonomous framework for real-time wrong-way driving vehicle detection from CCTV
Name: Pintusorn Suttiponpisarn
keyword:
Image processing
;
FastMOT
Abstract:
In 2018, around 22,000 people were killed in road accidents in Thailand and more than 80% of road accidents involved motorcycles. There are several reasons for traffic death and one of them is motorcycles driving in the wrong direction. In our research, we proposed an autonomous framework, WrongWay-LVDC, that can detect vehicles that drive in the wrong direction from CCTV videos. The framework is composed of other additional features that allow the detection process to run smoothly and automatically. The features that are included in the framework are road lane detecting, correct driving direction validating, detecting the wrong-way driving vehicles, and evidence image capturing. In total, we have proposed three algorithms and a feature for our main contributions. The three algorithms are, first, the Road Lane Boundary Detection based on the CCTV camera angle (RLB-CCTV) algorithm. Second, is the Majority-Based Correct Direction Detection (MBCDD) algorithm. Third, is the Distance-based Direction Detection (DBDD) algorithm. The additional feature we developed is the Inside Boundary Image (IBI) capturing feature that captures the clearest shot of the wrong-way driving vehicle. Each algorithm and feature will be explained in terms of its methods, result, and discussion in this thesis. As a result, the framework can execute continuously and output a report for vehicles driving behaviors in each area. The accuracy of our approach is 95.23% as we tested with several CCTV videos using the DBDD algorithm. Moreover, the framework can be implemented on edge devices with real-time speed where it is proved that the framework is light-weighted and consume low resource while executing
Thammasat University. Thammasat University Library
Address:
BANGKOK
Email:
preserv@tu.ac.th
Name: Sasiporn Usanavasin
Role:
advisor
Created:
2021
Modified:
2023-04-19
Issued:
2023-04-19
วิทยานิพนธ์/Thesis
application/pdf
eng
DegreeName: Master of Engineering
Level: Master's Degree
Descipline: Artificial Intelligence and Internet of Things
Grantor: Thammasat University
©copyrights Thammasat University
RightsAccess:
ใช้เวลา
0.035212 วินาที
Pintusorn Suttiponpisarn
| Title | Contributor | Type |
|---|---|---|
| An autonomous framework for real-time wrong-way driving vehicle detection from CCTV
มหาวิทยาลัยธรรมศาสตร์ Pintusorn Suttiponpisarn | Sasiporn Usanavasin | วิทยานิพนธ์/Thesis |
Sasiporn Usanavasin