Recognition of isolated elements picture using backpropagation neural network / Melati Sabtu

Sabtu, Melati (2005) Recognition of isolated elements picture using backpropagation neural network / Melati Sabtu. Degree thesis, Universiti Teknologi MARA.

[img] Text
TD_MELATI SABTU CS 05_5 1.pdf

Download (63kB)

Abstract

This project is developed to train the computer programs to recognize objects in the pictures. The purpose of the project is basically to extract and identify each object elements in an image picture. The reviews about the project had been done through the study about image recognition and back-propagation neural network. Several methods that related to this project also been derived through discussion about the image extraction, image preprocessing and some techniques about image segmentation. Gaining information from some resources such as articles and journals contribute various information and knowledge in process of investigation and discussion in order to make this project work smoothly. The project used Back-propagation Neural Network for the algorithm to classified images. Images that capture using digital camera will perform through the algorithm to classified images. The methodology used in the development of this project is basically based on the eight major steps. There are problem assessment, data acquisition, cropping, pre-processing, design, training, testing and documentation. There are three main programs work together. The programs are back-propagation neural network program, training and performance program and recognition program. The momentum rate, learning rate, the number of nodes and layers are the important factors that affect the neural network performance. For overall, the back-propagation algorithm has been proved as a method that can be used for recognition areas.

Item Type: Thesis (Degree)
Creators:
CreatorsEmail
Sabtu, MelatiUNSPECIFIED
Subjects: 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 > Neural networks (Computer science)

T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Electronics > Pattern recognition systems
Divisions: Faculty of Information Technology and Quantitative Sciences
Item ID: 9397
Uncontrolled Keywords: Backpropagation neural network, isolated elements picture, image preprocessing
Last Modified: 25 Jan 2017 08:05
Depositing User: Staf Pendigitalan 2
URI: http://ir.uitm.edu.my/id/eprint/9397

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year