作者: Ebrahim Ebrahimi , Kaveh Mollazade , Sirwan Babaei
DOI: 10.1016/J.MEASUREMENT.2014.05.003
关键词:
摘要: Abstract Wheat product quality is closely related to wheat seed purity. Purity an important factor that has a considerable impact on prices in grain storage silos. The aim of this paper was introduce machine vision based approach as primarily step for fabricating automatic purity determination and grading device. Experimental data consists 52 color, morphology, texture characteristic parameters, extracted from images samples, including four local grades eight common weed seeds growing fields Iran, were used build the classification models. A new algorithm combines Imperialist Competitive Algorithm (ICA) Artificial Neural Networks (ANNs) been two purposes: find best parameters set create robust Based upon results obtained study, total rate ICA–ANN grains vs. non-wheat seeds, classes, classes 96.25%, 87.50%, 77.22%, respectively.