No Detail Found!

Machine vision approach for identification of four variant pakistani rice using multi-features dataset

Author(s)
Hafiz Muhammad Ijaz Department of Computer Science & IT, Institute of Southern Punjab (ISP-M) Multan, Multan Punjab, Pakistan
Shehzad khan Department of Computer Science & IT, Institute of Southern Punjab (ISP-M) Multan, Multan Punjab, Pakistan
Mubasher Hussain Department of Computer Science & IT, Institute of Southern Punjab (ISP-M) Multan, Multan Punjab, Pakistan
Tanveer Aslam Department of Information Technologies, Islamia University of Bahawalpur, 63100 Bahawalpur Punjab, Pakistan
Nazir Ahmad Department of Information Technologies, Islamia University of Bahawalpur, 63100 Bahawalpur Punjab, Pakistan
Muzammil ur Rehman Department of Information Technologies, Islamia University of Bahawalpur, 63100 Bahawalpur Punjab, Pakistan
Syed Ali Nawaz Department of Information Technologies, Islamia University of Bahawalpur, 63100 Bahawalpur Punjab, Pakistan
Abdul Razzaq Department of Computer Science, Muhammad Nawaz Shareef University of Agriculture Multan, Multan Punjab, Pakistan
Salman Qadri Department of Computer Science, Muhammad Nawaz Shareef University of Agriculture Multan, Multan Punjab, Pakistan
Abstract
Crops are the most essential and beneficial food source in Pakistan. The demand for food has increased in Pakistan due to population growth. Pakistan produced 7,410 million tons of rice according to the financial year survey 2020 (FYS-2020). Pakistani rice has been cultivated in 3,304 hectares of agricultural land and exported worldwide. Rice is also increased by 0.6% of Pakistan's Gross Domestic Product (GDP) (FYS-2020). The old and manual process of rice classification is more expensive and time-consuming. In this study, we describe a machine vision approach for rice identification. We use four different rice varieties for the experimental process such as Pakei_Kaynat, Kaynat_Kauchei, Kauchei_Super_Banaspati and Tootaa_Kauchei (P1, P2, P3, and P4). The 100 images dataset have been used for practical work and the total calculated 400 (4 x 100) rice image. The different process has been deployed on available datasets, such as introduction, preprocessing methodology, and result discussion. A quality enhancement technique has been implementedto clarifythe difference between rice color and shape sampling, and it is also converted color imagestoa gray scale level. Every image has employed six different non-overlapping regions of interest (ROI's) and calculated 2400 (6 x 400) ROI's. Binary (B), Histogram (H), and Texture (T) features have been implemented and extracted 43 features on each ROI's and total calculated 103,200 (2400 x 43) machine learning (ML) features. Best First Search (BFS) Algorithm was used for feature optimization. Different ML classifiers are implemented for the experimental process: Function Multi-Layer-Perception, Function SMO, Random Tree, J48 Tree, Meta Classifier via Regression and Meta Bagging. The Function Multi-Layer-Perception overall accuracy (OA) has describeda better accuracy result is 99.8333%.
Publication Details
Page(s) 32-43
Published Journal: Lahore Garrison University Research Journal of Computer Science and Information Technology, Volume: 7, Issue: 1, Year: 2023

Request the full text from PASTIC

Keywords
Feature Extraction, Rice Grains, Machine Learning Classifier, Image Preprocessing
0

Citations

0

Downloads

47

Views