APPLICATION OF THE ARTIFICIAL NEURAL NETWORK ON TOMATO COLOR IDENTIFICATION

The 2nd National Intelligent Systems And Information Technology Symposium (ISITS’07), Oct 30-31, 2007, ITMA -UPM, Malaysia. pg314

APPLICATION OF THE ARTIFICIAL NEURAL NETWORK ON
TOMATO COLOR IDENTIFICATION

Aziz Mamat, a,d Wan Ishak Wan Ismail, c Napsiah Ismail,
b Abdul Rahman Ramli and a Khairuddin Abdul Rahman.

Abstract

Recently, vision technology has become more and more important especially in color detection and identification for several sectors such as in food and agricultural sectors. One of the techniques in identifying color is using Artificial Neural Network (ANN). The neural network has the intelligent in making their clarification or decision. This paper describes how neural network produce the result in evaluation of two colors. The neural network in this paper was based on back propagation technique. The result will be presented in percent of error. Number of epoch in the result shows the training made by neural network in matching of the colors. The training analysis was made through small iteration such as pixel to pixel. In this experiment, 10 by 10 pixel of the color plates was used for color identification. A small square of pixel was used to reduce the computation time. The result shows that the same color plate will be recognized significantly. The color plates under shadow or mixed pixel value color plates were difficult to recognize in a single neural network analysis.

.

a Robotic And Automation Laboratory Department of Biology And Agricultural Engineering Faculty of Engineering Universiti Putra Malaysia 43400 UPM Serdang, Selangor, Malaysia
b Intelligent Systems and Robotics Laboratory Institute of Advanced Technology Universiti Putra Malaysia 43400 UPM Serdang, Selangor, Malaysia
c Department of Mechanical Engineering Faculty of Engineering Universiti Putra Malaysia 43400 UPM Serdang, Selangor, Malaysia
d Department of Biology And Agricultural Engineering Faculty of Engineering Universiti Putra Malaysia 43400 UPM Serdang, Selangor, Malaysia

.

Paper 36, pg 314-319Aziz Mamat, Wan Ishak Wan Ismail, Napsiah Ismail, Abdul Rahman Ramli, Khairuddin Abdul Rahman(UPM)APPLICATION OF THE NEURAL NETWORK ON TOMATO COLOR IDENTIFCATION