Abstract
AbstractThe study was aimed at the evaluation of the usefulness of textures of the outer surface from the images of apple skin and flesh for discrimination of different cultivars. The texture parameters were calculated from color channels: R, G, B, L, a, b, U, V, H, S, I, X, Y, Z. In the case of cultivar discrimination performed for the apple skin, the highest accuracies were obtained for textures from channels R, a and X. In the case of channels R and a, the apples were classified with the total accuracy of up to 93%. For channel X, the highest total accuracy was 90%. For discrimination based on the textures selected from images of a longitudinal section of apples, the total accuracy reached 100% for channels G, b and U. In the case of the cross-section images, the total accuracies were also satisfactory and reached 93% for channel G, 97% for channels b and U. The obtained results proved that the use of image processing based on textures can allow the discrimination of apple cultivars with a high probability of up to 100% in the case of textures selected from images of a longitudinal section. The results can be applied in practice for cultivar discrimination and detection of the falsification of apple cultivars. The obtained results revealed that texture features can allow for cultivar identification of apples with a very high probability in an inexpensive, objective, and fast way.
Graphic abstract
Publisher
Springer Science and Business Media LLC
Subject
Industrial and Manufacturing Engineering,Biochemistry,General Chemistry,Food Science,Biotechnology
Reference30 articles.
1. Nabi SU, Baranwal VK, Yadav MK, Rao GP (2020) Association of Apple necrotic mosaic virus (ApNMV) with mosaic disease in commercially grown cultivars of apple (Malus domestica Borkh) in India. 3 Biotech 10(122):1–9
2. Luo W, Huan S, Fu H, Wen G, Cheng H, Zhou J, Wu H, Shen G, Yu R (2011) Preliminary study on the application of near infrared spectroscopy and pattern recognition methods to classify different types of apple samples. Food Chemistr 128:555–561
3. Cortés V, Cubero S, Blasco J, Aleixos N, Talens P (2019) In-line application of visible and near-infrared diffuse reflectance spectroscopy to identify apple varieties. Food Bioprocess Technol 12(6):1021–1030
4. Ronald M, Evans M (2016) Classification of selected apple fruit varieties using naive Bayes. Indian J Comput Sci Eng (IJCSE) 7(1):13–19
5. Wu X, Wu B, Sun J, Yang N (2016) Classification of apple varieties using near infrared reflectance spectroscopy and fuzzy discriminant C-means clustering model. J Food Process Eng 40(2):e123551–e123557