Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data
With the always increasing amount of image data, it has become a necessity to automatically look for and process information in these images. As fashion is captured in images, the fashion sector provides the perfect foundation to be supported by the integration of a service or application that is bu...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-03-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/9/6/624 |
id |
doaj-409326b3461040c2b23564f3473b7111 |
---|---|
record_format |
Article |
spelling |
doaj-409326b3461040c2b23564f3473b71112021-03-17T00:00:24ZengMDPI AGMathematics2227-73902021-03-01962462410.3390/math9060624Image Classification for the Automatic Feature Extraction in Human Worn Fashion DataStefan Rohrmanstorfer0Mikhail Komarov1Felix Mödritscher2Department Computer Science, University of Applied Science Technikum Wien, 1200 Vienna, AustriaDepartment of Business Informatics, Graduate School of Business, National Research University Higher School of Economics, 101000 Moscow, RussiaDepartment Computer Science, University of Applied Science Technikum Wien, 1200 Vienna, AustriaWith the always increasing amount of image data, it has become a necessity to automatically look for and process information in these images. As fashion is captured in images, the fashion sector provides the perfect foundation to be supported by the integration of a service or application that is built on an image classification model. In this article, the state of the art for image classification is analyzed and discussed. Based on the elaborated knowledge, four different approaches will be implemented to successfully extract features out of fashion data. For this purpose, a human-worn fashion dataset with 2567 images was created, but it was significantly enlarged by the performed image operations. The results show that convolutional neural networks are the undisputed standard for classifying images, and that TensorFlow is the best library to build them. Moreover, through the introduction of dropout layers, data augmentation and transfer learning, model overfitting was successfully prevented, and it was possible to incrementally improve the validation accuracy of the created dataset from an initial 69% to a final validation accuracy of 84%. More distinct apparel like trousers, shoes and hats were better classified than other upper body clothes.https://www.mdpi.com/2227-7390/9/6/624image classificationneural networkconvolutionalmachine learningfashionapparel |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Stefan Rohrmanstorfer Mikhail Komarov Felix Mödritscher |
spellingShingle |
Stefan Rohrmanstorfer Mikhail Komarov Felix Mödritscher Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data Mathematics image classification neural network convolutional machine learning fashion apparel |
author_facet |
Stefan Rohrmanstorfer Mikhail Komarov Felix Mödritscher |
author_sort |
Stefan Rohrmanstorfer |
title |
Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data |
title_short |
Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data |
title_full |
Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data |
title_fullStr |
Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data |
title_full_unstemmed |
Image Classification for the Automatic Feature Extraction in Human Worn Fashion Data |
title_sort |
image classification for the automatic feature extraction in human worn fashion data |
publisher |
MDPI AG |
series |
Mathematics |
issn |
2227-7390 |
publishDate |
2021-03-01 |
description |
With the always increasing amount of image data, it has become a necessity to automatically look for and process information in these images. As fashion is captured in images, the fashion sector provides the perfect foundation to be supported by the integration of a service or application that is built on an image classification model. In this article, the state of the art for image classification is analyzed and discussed. Based on the elaborated knowledge, four different approaches will be implemented to successfully extract features out of fashion data. For this purpose, a human-worn fashion dataset with 2567 images was created, but it was significantly enlarged by the performed image operations. The results show that convolutional neural networks are the undisputed standard for classifying images, and that TensorFlow is the best library to build them. Moreover, through the introduction of dropout layers, data augmentation and transfer learning, model overfitting was successfully prevented, and it was possible to incrementally improve the validation accuracy of the created dataset from an initial 69% to a final validation accuracy of 84%. More distinct apparel like trousers, shoes and hats were better classified than other upper body clothes. |
topic |
image classification neural network convolutional machine learning fashion apparel |
url |
https://www.mdpi.com/2227-7390/9/6/624 |
work_keys_str_mv |
AT stefanrohrmanstorfer imageclassificationfortheautomaticfeatureextractioninhumanwornfashiondata AT mikhailkomarov imageclassificationfortheautomaticfeatureextractioninhumanwornfashiondata AT felixmodritscher imageclassificationfortheautomaticfeatureextractioninhumanwornfashiondata |
_version_ |
1724219191292067840 |