Snake detection system using convolutional neural network / Muhammad Danial Ahmad Tarmizi

Ahmad Tarmizi, Muhammad Danial (2020) Snake detection system using convolutional neural network / Muhammad Danial Ahmad Tarmizi. Degree thesis, Universiti Teknologi MARA, Cawangan Melaka.

Download

[thumbnail of 35680.pdf] Text
35680.pdf

Download (143kB)

Abstract

Snake known as most dangerous reptile as it threaten our live and can deal fatal wound for human. Mostly people cannot differentiate between venomous and non-venomous snake because most of non-venomous snake are likely to look like venomous one. This kind of problem can be solved using Artificial Intelligence approach. This paper aims to discuss about the project built for encountering that problems to detect and classify the snakes. Objectives of this project is to design the flow of the system, developed in Window application and test the functionality of the system and reliability for the predictive model. This project uses Convolutional Neural Network algorithms which is one the best algorithms for image processing. The algorithm is built using Tensorflow software. The development of the project is based on Waterfall methodology. Waterfall methodology consists of 6 phases starting from requirement analysis, system design, implementation, testing, deployment and maintenance. The predictive model success in detecting and classifying the snake. The model achieve 96.89 % of accuracy percentage in training set and 96% of accuracy from testing set. This project can be improved by employing in the less consumption hardware like Jetson Nano and using light sensitive camera to improve the image quality while detecting snake.

Metadata

Item Type: Thesis (Degree)
Creators:
Creators
Email
Ahmad Tarmizi, Muhammad Danial
2017995661
Contributors:
Contribution
Name
Email / ID Num.
Thesis advisor
Mahzan, Sulaiman
UNSPECIFIED
Subjects: Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science > Expert systems (Computer science). Fuzzy expert systems
Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science > Neural networks (Computer science)
Q Science > QA Mathematics > Instruments and machines > Electronic Computers. Computer Science > Artificial immune systems. Immunocomputers
Divisions: Universiti Teknologi MARA, Melaka > Jasin Campus > Faculty of Computer and Mathematical Sciences
Programme: Bachelor of Computer Science (Hons) (CS230)
Item ID: 35680
Uncontrolled Keywords: Snake detection system; Convolutional neural network; Artificial Intelligence approach
URI: https://ir.uitm.edu.my/id/eprint/35680

Fulltext

Fulltext is available at:
  • Library Terminal Workstation (Digital Format) - Accessible via UiTM Libraries
  • ID Number

    35680

    Indexing


    View in Google Scholar

    Edit Item
    Edit Item