Grading and sorting technique of dragon fruits using machine learning algorithms
Pallavi U. Patil, Sudhir B. Lande, Vinay J. Nagalkar, Sonal B. Nikam, G.C. Wakchaure
Abstract
Climate change-induced environmental stresses and limited agricultural land demanding intensification of sustainable agriculture over degraded land via crop diversification strategies. Dragon fruit is one of the potential options and popularising in resource-poor degraded lands apart from its several nutraceutical advantages. Hence, understanding of facts related to its consumer acceptability and maintaining high quality for marketing and processing is highly essential. Therefore in this study, we have developed grading and sorting techniques for dragon fruit using machine learning algorithms (CNN, ANN, and SVM) based on a thorough review of techniques or algorithms available to detect and classify fruit quality using various features of fruits and vegetables. Working of these algorithms is based on the, shape, size, weight, color, and diseases of dragon fruits. Raspberry functionality counts the total number of fruits that are available in the fruit bucket and these are separated by their maturity level using machine learning algorithms.