Discrimination of wheat grain varieties using image analysis: morphological features
2011
Zapotoczny, Piotr
This paper presents the results of a study on the discrimination of 11 wheat grain varieties in three successive years of cultivation and at the grain humidity of 12, 14 and 16%. Each grain was described with the use of 54 geometric variables which, after reduction of variables, left 20 for use in the main analysis. Variables calculated from linear dimensions had the greatest share in the group of discriminating variables, with shape-related indexes being of lesser importance. Seven methods of variables selection based on genetic algorithms (the Class Ranker and Class RankersSearch methods) were used in the study. The final discriminant analysis was performed with the use of stepwise progressive analysis and Meta MultiClass Classifier. The proposed statistical model classified varieties with 90–100% accuracy, depending on the experimental group. Grain images were acquired with a flat scanner, and grains were arranged with a specially designed matrix which enabled arranging 552 grains in rows and columns within several minutes; this makes the method usable in the cereal industry.
Mostrar más [+] Menos [-]Palabras clave de AGROVOC
Información bibliográfica
Este registro bibliográfico ha sido proporcionado por National Agricultural Library