Thesis Details
Automatic Trafic Scene Analysis Using Image Processing
This thesis deals with the issue of scene analysis using computer vision methods. The aim of this work is to create a system capable of automatically detecting anomalies found in video recordings. The present thesis discusses object-detection and object-tracking systems. It also pursues graphical user interface and violation-detecting algorithm of user-defined rules. As a result of the present thesis, a web application is created that allows users to manage their videos, to define rules for scenes, to start the anomaly detection as well as display the results of the analysis. The system operates in real-time, notifies users when the operation is finished and stores the analysis results for further processing.
Image Recognition, Machine Learning, Python, Flask, YOLO, DeepSort, Web Application, Object detection, Object tracking, Computer Vision
Bařina David, Ing., Ph.D. (DCGM FIT BUT), člen
Burget Radek, doc. Ing., Ph.D. (DIFS FIT BUT), člen
Holík Lukáš, doc. Mgr., Ph.D. (DITS FIT BUT), člen
Jaroš Jiří, doc. Ing., Ph.D. (DCSY FIT BUT), člen
@bachelorsthesis{FITBT20774, author = "Luk\'{a}\v{s} V\'{a}lek", type = "Bachelor's thesis", title = "Automatic Trafic Scene Analysis Using Image Processing", school = "Brno University of Technology, Faculty of Information Technology", year = 2020, location = "Brno, CZ", language = "english", url = "https://www.fit.vut.cz/study/thesis/20774/" }