KLASIFIKASI JENIS BUAH DENGAN CONVOLUTIONAL NEURAL NETWORK (CNN) MENGGUNAKAN ARSITEKTUR XCEPTION

Abstract

Deep learning is a subfield of machine learning that focuses on training artificial neural networks to learn and make intelligent decisions without being explicitly programmed. It is inspired by the structure and the functioning of the human brain. One of the most popular types of artificial neural networks is Convolutional Neural Network (CNN), CNN is known for its excellence in analyzing visual data such as image recognition, object detection, and image classification. In this research, CNN architecture exception is being used for classifying image of fruit variance. Exception is a type of CNN architecture, and it is an extension of the Inception architecture, which itself was designed to improve the efficiency and effectiveness of CNN. This research aims to know the performance of CNN using exception architecture in classify image of different types of fruits, such as Starfruit, Durian, Mangosteen, Rambutan, and Snake fruit. The result of this research shows that CNN models using exception architecture can predict image input very well. Using confusion matrix as an evaluation method, the model got 0.98 precision and 0.98 accuracy, therefore the model can accurately classify the image input to the fruit type classes that have been assigned.