Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/513063
Title: Study of Deep Learning Models for Vision Based Vehicle Detection
Researcher: Singhal Nikita (19ENG7CSE0008)
Guide(s): Prasad Lalji
Keywords: Computer Science
Computer Science Software Engineering
Engineering and Technology
University: SAGE University, Indore
Completed Date: 2023
Abstract: As the number of vehicles on the road increases, a severe problem of traffic congestion newlineand management arises. Many times, people face traffic jams, and due to this congestion, newlinepeople do not follow traffic rules and regulations, which results in personal injury, death, newlineand damage to one s vehicle or other property. The Intelligent Transport System (ITS) newlineplays an important role in handling common traffic issues such as accidents, congestion of newlinetraffic, vehicle robberies, traffic rule violation, and automatic toll collection, and so on. newlineThat s why ITS attracted lots of researchers in the last decade and became an important newlinearea of study. Vehicle detection is the heart of ITS, which is widely used in many newlineapplications like congestion prediction, future road infrastructure requirement prediction, newlineautomated parking, and security enforcement. Various vehicle detection systems have been newlinedeveloped based on innovative sensor-based technologies, machine learning, image newlineprocessing, deep learning, and wireless communication technologies; however, it is still a newlinechallenge to deal with certain realistic environments. newlineVehicle detection has received a great deal of attention in computer vision literature. It is newlinea method of locating vehicles in an image and classifying them into different categories, newlinesuch as cars, buses, trucks, and so on. Bounding boxes are drawn around the vehicles newlinepresent in the image, with the predicted vehicle class and confidence score associated with newlineeach bounding box. Vehicle detection methods based on sensors are basically divided into newlinetwo broad categories: intrusive and non-intrusive. Intrusive sensors are mounted directly newlineon the surface of the pavement, in saw-cuts or holes in the surface of the lane, by tunnelling newlinethrough the surface, or directly anchored to the surface of the pavement. There is a wide newlinevariety of intrusive sensors [1] used in vehicle detection and classification systems, such newlineas loop detectors, magnetometers, piezoelectric sensors, vibration sensors, accelerometers, newlineetc. The main downs
Pagination: 
URI: http://hdl.handle.net/10603/513063
Appears in Departments:Faculty of Engineering & Technology

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File488.02 kBAdobe PDFView/Open
02_prelim pages.pdf1.06 MBAdobe PDFView/Open
03_content.pdf484.94 kBAdobe PDFView/Open
04_abstract.pdf487.27 kBAdobe PDFView/Open
05_chapter 1.pdf1.19 MBAdobe PDFView/Open
06_chapter 2.pdf1.58 MBAdobe PDFView/Open
07_chapter 3.pdf723.88 kBAdobe PDFView/Open
08_chapter 4.pdf1.83 MBAdobe PDFView/Open
09_chapter 5_7.pdf3.35 MBAdobe PDFView/Open
10_annexures.pdf2.29 MBAdobe PDFView/Open
80_recommendation.pdf421.35 kBAdobe PDFView/Open
Show full item record


Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Altmetric Badge: