Object recognition is a development of the world of AI (Artificial Intelligence) which is used to recognize objects in digital images. Various kinds of recognition can be done with Clothing Image Recognition. In the global industrial era, development of fashion type and variant makes it difficult for people to recognize, for example topwear has variants ranging from sweatshirts, kurtas, waistcoats, etc. This makes users hard to distinguish categories from types of clothing. In fact, users often forget or didn't know the name of a clothing type that they want to find because they just remember the shape.Clothing image recognition uses the CNN (Convolutional Neural Network) method with the VGG16 model. The application of CNN with the VGG16 model will be carried out on Python and Keras as the library to speed up the research process on training data. Application of the program to mobile devices using Android Studio with the Kotlin programming language, with the help of the Tensorflow Lite library as machine learning on mobile devices. The results of this thesis show an accuracy of 81.83% for training data in epoch 1 and 82%-94% for training data in epoch 10 on clothing image recognition.
Copyrights © 2022