UTAR Institutional Repository

Recognising Malaysia traffic sign with rain drop disturbance using deep learning

Pang, Wan Chee (2022) Recognising Malaysia traffic sign with rain drop disturbance using deep learning. Final Year Project, UTAR.

[img]
Preview
PDF
Download (628Kb) | Preview

    Abstract

    Traffic sign recognition is part of the driving assistance systems that can monitor driving by automatically recognising the traffic signs on the road and alerting drivers. In Malaysia, road accidents have remained one of the top principal causes of death in recent years. Other than the driver's carelessness, weather such as rain might challenge the visibility and recognition of the traffic signs. Hence, we develop a road sign recognition system based on a deep learning algorithm that can identify the Malaysia traffic sign with raindrop disturbance. In this study, a convolutional neural network (CNN) is applied as it is a deep neural network that is well-known for effective image recognition. The algorithms are developed in Python language using Anaconda Spyder as the software. There are 18 different combinations of the epochs, dropout rates, and colour of the image input in the first convolutional layer (either colour or greyscale version) used in this study. In conclusion, we proposed an image recognition system with a predictive accuracy of 99.52% for Malaysia traffic signs with raindrop disturbance.

    Item Type: Final Year Project / Dissertation / Thesis (Final Year Project)
    Subjects: H Social Sciences > H Social Sciences (General)
    H Social Sciences > HA Statistics
    Q Science > Q Science (General)
    T Technology > T Technology (General)
    T Technology > TE Highway engineering. Roads and pavements
    Divisions: Faculty of Science > Bachelor of Science (Honours) Statistical Computing and Operations Research
    Depositing User: ML Main Library
    Date Deposited: 05 Jan 2023 22:02
    Last Modified: 05 Jan 2023 22:02
    URI: http://eprints.utar.edu.my/id/eprint/4939

    Actions (login required)

    View Item