Klasifikasi Jenis Buah Pisang Menggunakan Algoritma Probabilitas Neural Network

Abstract

In the current technological era, very rapid developments make many components of life that can be carried out efficiently. Especially for the identification of types of bananas Kepok, Ambon and Barangan. Where in Indonesia there are various types of Kepok, Ambon and Barangan bananas which have quite identical texture qualities, so that it makes it difficult for people to recognize them. However, this problem can be solved through the development of existing technology. A researcher proposes an application that classifies three types of bananas Kepok, Ambon and Barangan using python language. The process in this research is through a 3-stage original image process, namely Resize, Median Filter and Hough transform, segmentation in edge detection, the algorithm used is the Neural Network Probability and is carried out before this application is implemented in the form of the Google Colabs application. The results of this study are the application of the classification of the types of kepok bananas with an accuracy of 76% for the types of Kepok bananas, Barangan bananas have an accuracy of 87%, and Ambon bananas 88%.