Classification of Sweet Potato Variety using Convolutional Neural Network

Sweet potato (Ipomoea batatas) from being 'a poor man's crop,' sweet potatoes it is now viewed as a good with high business assets also significance for a multitude of reasons. Due to its versatility and high nutritional value, it is now considered a cash crop, making it one of the fastest growing commodities on the market. However, variety classification on this commodity is considerably significant to ensure the quality of products before reaching production. This study aims to present a sweet potato classification using convolutional neural network method. The results of the study show that the convolutional neural network achieves 96.33 percent accuracy in classifying sweet potato variety. Thus, a convolutional neural network is appropriate in the classification of sweet potatoes. This paper could bring forward an approach of sweet potato classification. Practical implications and research directions are presented.