Identification of Pepper Image Using Convolutional Neural Network (CNN) Deep Learning Method

Abstract

The Convolutional Neural Network (CNN) method is one of the Deep Learning methods that is currently being developed. The Convolutional Neural Network (CNN) method is one of the Deep Learning methods that is currently being developed. The purpose of this study is to identify the pepper image by applying the Convolutional Neural Network (CNN) Deep Learning method. There are several stages in this technique, the first stage is to prepare the required pepper image data set. The next stage is the preprocessing and sorting of pepper images. Then, the formation of models and system training, the last is to do for system testing. This research uses CNN to recognize pepper images and determine the accuracy value. In this study, 19 pepper images were used as testing data from 76 pepper images used in the training dataset. Pepper testing produces an average value of data testing accuracy of 0, 97756064.