Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: Hu’s moment invariants

Sortuj według:

Ogranicz wyniki do:

Exploring Hu’s moment invariants and Zernike moments for effective identification of two-row and six-row barley varieties

Szturo Karolina

Technical Sciences / University of Warmia and Mazury in Olsztyn

2024

Vol. 27

55--69

Barley variety identification is a complex and economically significant task. The identification of two-row and six-row grains is particularly important due to their different characteristics, such as protein and starch content, which have specific implications in different applications. This paper evaluates the effectiveness of discriminating between two-row and six-row barley grains using Hu moment invariants and Zernike moments in combination with various classifiers including linear and SVM classifiers with linear, RBF, polynomial, and sigmoid kernels. The application of Zernike moments and an SVM classifier using an RBF kernel achieved an accuracy level of 99.2%. In comparison, the application of Hu’s moment invariants resulted in an accuracy of 98.5%.