UPM Institutional Repository

Immature paddy detection using image processing technique


Citation

Sudin, Nor Munira and Bejo, Siti Khairunniza (2012) Immature paddy detection using image processing technique. In: International Conference on Agricultural and Food Engineering for Life (Cafei2012), 26-28 Nov. 2012, Palm Garden Hotel, Putrajaya. (pp. 799-806).

Abstract

Maturity of paddy contributes a very high impact on the production of rice quality. Immature paddy will produce chalky rice which reduced the rice quality. The use of image processing technique in paddy grading can increase the commercial value of paddy. In this study, six different techniques of an automatic image thresholding had been studied to segment the paddy image. The best technique was later used in detecting immature paddy. Results taken from the experiment has shown that median (non-iterative) technique perform the best among the others. It gave significant different between mature and immature paddy, acceptable time taken to process the image and narrow overlapping area between mature and immature paddy. The paddy was later classified using its mean intensity. Immature paddy give lower value of mean intensity compared to mature paddy. Based on the validation results using 130 samples of paddy taken from 100% mature, 100% immature and 10% mature, it has been shown that the pure samples give higher percentage accuracy of detection. Contacting paddy on the sample need to be detached to produce high accuracy result.


Download File

[img] PDF
33582.pdf
Restricted to Repository staff only

Download (540kB)

Additional Metadata

Item Type: Conference or Workshop Item (Poster)
Divisions: Faculty of Engineering
Publisher: Faculty of Engineering, Universiti Putra Malaysia
Keywords: Paddy; Image processing; Automatic thresholding; Mean intensity; Detection
Depositing User: Erni Suraya Abdul Aziz
Date Deposited: 22 Jul 2015 07:16
Last Modified: 03 Feb 2017 09:29
URI: http://psasir.upm.edu.my/id/eprint/33582
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item