Abstract—The objective of this research is to develop a computer system, which can recognize a plant seed image. The system is called “Plant seed image recognition system (PSIRS)”. The system consists of 5 processing modules, namely: 1) image acquisition, 2) image preprocessing, 3) feature extraction, 4) image recognition, and 5) display result. The experiment was conducted on more than 1,000 seed images by employing the Euclidean distance technique to recognize them. The precision rates of the system were 95.1 percent for correct matching in the training data set and 64.0 percent for unknown in the untrained data set. The average access time was 8.79 seconds per image.
Index Terms—Plant seed, seed recognition, image processing.
Benjamaporn Lurstwut is graduated from the Faculty of Information and Communication Technology, Mahidol University, Bangkok 10400, Thailand (e-mail: g5237682@student.mahidol.ac.th).
Chomtip Pornpanomchai is an Assistant Professor in the Faculty of Information and Communication Technology, Mahidol University, Bangkok 10400, Thailand (e-mail: itcpp@mahidol.ac.th).
[PDF]
Cite:Benjamaporn Lurstwut and Chomtip Pornpanomchai, "Plant Seed Image Recognition System (PSIRS)," International Journal of Engineering and Technology vol. 3, no. 6, pp. 600-605, 2011.
