Open AccessImage Processing Identifacation for Indonesian Cake Cuisine using CNN Classification Technique
Author(s) -
Dian Ade Kurnia,
Andi Setiawan,
Dita Rizki Amalia,
Rita Wahyuni Arifin,
Didik Setiyadi
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1783/1/012047
Subject(s) - preprocessor , convolutional neural network , indonesian , artificial intelligence , computer science , pattern recognition (psychology) , visualization , data pre processing , kuala lumpur , deep learning , image processing , image (mathematics) , philosophy , linguistics , marketing , business
Indonesia is famous for its traditional food that is popular both domestically and abroad. A number of cakes are among the favorite traditional foods. There are types of cakes that can be processed in Indonesia, such as kue dadar gulung, kastangel, klepon, lapis, lumpur, putri salju, risoles and serabi. The most of types of cakes available, visually the human recognize are easy, however computer visiion requires a special technicality in identifying the object of the image to the type of cakes. Therefore, to recognize objects in the form of images from cakes as one of Indonesia’s traditional foods, deep learning algorithm techniques can be used, namely Convolutional Neural Network (CNN). In this paper, the CNN Algorithm technique will be applied to 1676 datasets consisting of 80% training data and 20% testing data in which there are images of traditional cakes from Indonesia. The stages are carried out through preprocessing, operational datasets, visualization datasets, modeling techniques, performance evaluations, errors analysis which finally result in the conclusion that performance evaluation reaches 65.00%.