Klasifikasi Pakaian Menggunakan Convolutional Neural Networks CNN Berbasis Clothing Dataset

Abstract

The main problem in clothing classification is the large variety of shapes, types, and styles, which makes the identification and grouping of clothes a challenge in itself. In this era of global industry, clothing image recognition is important, especially for e-commerce and mobile applications. This research aims to develop a high-resolution clothing image classification model using the Convolutional Neural Network (CNN) method. In this study, we applied CNN architecture using a dataset of high-resolution clothing images with a variety of clothing types. We trained and tested this model using the Clothing Dataset (Full, High Resolution)-based CNN method to improve the accuracy of detection and classification of clothes in images. As a result, the model achieved an average accuracy of 93.25%, which shows that the system is effective in recognizing different types of clothing. This research is useful for image recognition applications in the fashion field, for example in automatic product cataloging and image search in e-commerce, which help users find products according to their visual preferences.

Keywords: Convolutional Neural Network (CNN), clothing classification, high-resolution datasets, image processing, e-commerce, object recognition, fashion industry, image search.