Open AccessDevelopment of a machine vision system for rice seed inspection system
Author(s) -
Rafiza Ruslan,
Siti Khairunniza-Bejo,
Ibni Hajar Rukunudin,
Mahirah Jahari,
M. M. Ibrahim
Publication year - 2020
Publication title -
food research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.218
H-Index - 7
ISSN - 2550-2166
DOI - 10.26656/fr.2017.4(s6).049
Subject(s) - machine vision , image processing , agricultural engineering , software , protocol (science) , rice plant , process (computing) , engineering , computer science , artificial intelligence , image (mathematics) , agronomy , biology , operating system , medicine , alternative medicine , pathology
Rice seed production in Malaysia is greatly dependent on the purity of the cultivatedpaddy seed produced through the government certified paddy seed program. The seeds tobe marketed by the seed processors must undergo quality control protocol where the seedlots are sampled from the seed farms and seed processing plants for purity analysis by theenforcing agency at the Seed Testing Laboratory of the Department of Agriculture (DoA).The current inspection conducted by the laboratory is based on a manual process, which islaborious and time-consuming. Therefore, a prototype (Patent ID: PI2018500018) of amachine vision-based rice seed inspection system (RiSe-IViS) was developed to explorethe possibility of replacing the existing manual method in distinguishing the weedy riceand cultivated rice seeds under the Standard Jabatan Pertanian Malaysia (SJPM) standardprotocol with a modern, effective and efficient technique using an image processingapproach. The developed RiSe-IViS prototype consists of two parts i) hardwareconfiguration and ii) software development. This paper discussed the criteria to beestablished, challenges and limitation encountered in developing the hardware prototypeinvolving the image acquisition setup, lighting configuration and seed plate design. Theimportance of each criterion to ensure its reproducibility are also discussed. A softwareprogramme was developed to assist the user for image acquisition and analysis. The imageprocessing steps undertaken in the programme are also discussed. The RiSe-IViS isexpected to classify major rice seed varieties available in Malaysia against the weedy ricevariants with superior accuracy.