Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
7141930 | Sensors and Actuators B: Chemical | 2018 | 27 Pages |
Abstract
Knowledge of the viability status of seeds before sowing is important to farmers (for yield prediction) and to seed companies (for seed warrant determination). However, a diversity of factors collaborate to reduce or completely render seeds non-viable both during pre- and post-harvest operations. Many methods have been employed to detect seed viability, but perhaps one of the promising is hyperspectral imaging. This is because of its high speed and ability to non-destructively detect the internal conditions of seeds, making it the perfect solution especially for industrial sorting applications. This study was conducted to determine suitable classification model(s) for classifying corn seeds based on their viability using hyperspectral imaging. For this study, 600 corn samples were selected, and half of them treated using microwave heat treatment while the rest were kept as the control group. Hyperspectral imaging data from all the samples were then collected using a shortwave infrared hyperspectral camera with a range of 1000-2500Â nm. Three classification models, linear discriminant analysis (LDA), partial least squares discriminant analysis (PLS-DA), and support vector machines (SVM), coupled with some pre-processing methods, were tested to determine the most suitable among them. The SVM model resulted in the highest spectral classification of up to 100%, which is 5% better than the previous research PLS based method. The model also produced flawless classification images, suggesting that hyperspectral imaging can be used to accurately classify corn based on viability. In summary, the results of this study serve as a major step towards development of a fast and non-destructive large-scale hyperspectral-based sorting system for corn viability determination.
Related Topics
Physical Sciences and Engineering
Chemistry
Analytical Chemistry
Authors
Collins Wakholi, Lalit Mohan Kandpal, Hoonsoo Lee, Hyungjin Bae, Eunsoo Park, Moon S. Kim, Changyeun Mo, Wang-Hee Lee, Byoung-Kwan Cho,