We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
The use of machine vision technique to classify cultivated rice seed variety and weedy rice seed variants for the seed industry.
- Authors
Aznan, A. A.; Rukunudin, I. H.; Shakaff, A. Y. M.; Ruslan, R.; Zakaria, A.; Saad, F. S. A.
- Abstract
Seed purity is a crucial seed quality parameter in the Malaysian rice seed standard. The use of high quality cultivated rice seed, free of any foreign seeds, is the prerequisite to sustaining high yield in rice production. The presence of foreign seeds such as weedy rice in the cultivated rice seeds used by the farmers can adversely affect growth and yield as it competes for space and nutrients with the cultivated rice varieties in the field. Being the most dominant and competitive element compared to the cultivated rice seeds, the Malaysian seed standard prescribed that the maximum allowable of weed seeds in a 20-kilogram certified rice seed bag produced by local rice seed processors is 10 weed seeds per kilogram. The current cleaning processes that rely mostly on the difference in physical traits do not guarantee effective separation of weedy rice seeds from the lots. Seed bags found to contain more than 10 weed seeds upon inspection by the enforcing agency will not be approved for distribution to farmers. The paper describes a study carried out to explore the use of machine vision approach to separate weedy rice seed from cultivated rice seeds as a potential cleaning technique for the rice seed industry. The mean classification accuracies levels of the extracted morphological feature model were achieved at 95.8% and 96.0% for training and testing data sets respectively.
- Subjects
RICE varieties; SEED quality; RICE yields; RICE seed industry; COMPUTER vision
- Publication
International Food Research Journal, 2016, Vol 23, pS31
- ISSN
1985-4668
- Publication type
Article