Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/75220
Title: Vehicle detection and tracking from road side camera
Authors: Tan, Cedric Kuang Huo
Keywords: DRNTU::Engineering
Issue Date: 2018
Abstract: With deep learning becoming more useful in the industry, there have been several research studies concerning vehicle detection. Many static cameras have been installed along roadways in order to capture the amount and different types of vehicles for research purposes. This technique can help people to do analysis and extraction of essential information from the images or videos accurately and proficiently. This study aims to develop an algorithm using deep learning method to detect vehicles from a static camera which is mounted on the school’s compound street lamppost. By doing pre-training, it will compare with the datasets provided to classify the class accordingly. Vehicle detection using video algorithms can bring many benefits to the road users allowing them to save time and to ease the flow of the traffic. While the normal algorithms being implemented is not as efficient in terms of speed, resources needed and portability. With the help of MXNET framework, we can do training with different types of datasets with different parameters so that we can achieve results for the detection by the single shot detector. To conclude, comparison will be made with respect to the results of the detection and findings have shown that the fine-tuned pre- trained Neural Network can help to boost the detection performance.
URI: http://hdl.handle.net/10356/75220
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP-Cedric_Tan_Kuang_Huo_U1522780C.pdf
  Restricted Access
Vehicle Detection and Tracking from Road Side Camera3.24 MBAdobe PDFView/Open

Page view(s) 50

89
checked on Oct 19, 2020

Download(s) 50

44
checked on Oct 19, 2020

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.