Chart Classification Using Siamese CNN

In recovering information from the chart image, the first step should be chart type classification. Throughout history, many approaches have been used, and some of them achieve results better than others. The latest articles are using a Support Vector Machine (SVM) in combination with a Convolutiona...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Filip Bajić, Josip Job
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	chart classification chart image processing data visualization Siamese neural network image processing and computer vision Photography TR1-1050 Computer applications to medicine. Medical informatics R858-859.7 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/7631a6f4f2ba46cfa6b2a9ece9490a9d
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:7631a6f4f2ba46cfa6b2a9ece9490a9d
record_format	dspace
spelling	oai:doaj.org-article:7631a6f4f2ba46cfa6b2a9ece9490a9d2021-11-25T18:03:23ZChart Classification Using Siamese CNN10.3390/jimaging71102202313-433Xhttps://doaj.org/article/7631a6f4f2ba46cfa6b2a9ece9490a9d2021-10-01T00:00:00Zhttps://www.mdpi.com/2313-433X/7/11/220https://doaj.org/toc/2313-433XIn recovering information from the chart image, the first step should be chart type classification. Throughout history, many approaches have been used, and some of them achieve results better than others. The latest articles are using a Support Vector Machine (SVM) in combination with a Convolutional Neural Network (CNN), which achieve almost perfect results with the datasets of few thousand images per class. The datasets containing chart images are primarily synthetic and lack real-world examples. To overcome the problem of small datasets, to our knowledge, this is the first report of using Siamese CNN architecture for chart type classification. Multiple network architectures are tested, and the results of different dataset sizes are compared. The network verification is conducted using Few-shot learning (FSL). Many of described advantages of Siamese CNNs are shown in examples. In the end, we show that the Siamese CNN can work with one image per class, and a 100% average classification accuracy is achieved with 50 images per class, where the CNN achieves only average classification accuracy of 43% for the same dataset.Filip BajićJosip JobMDPI AGarticlechart classificationchart image processingdata visualizationSiamese neural networkimage processing and computer visionPhotographyTR1-1050Computer applications to medicine. Medical informaticsR858-859.7Electronic computers. Computer scienceQA75.5-76.95ENJournal of Imaging, Vol 7, Iss 220, p 220 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	chart classification chart image processing data visualization Siamese neural network image processing and computer vision Photography TR1-1050 Computer applications to medicine. Medical informatics R858-859.7 Electronic computers. Computer science QA75.5-76.95
spellingShingle	chart classification chart image processing data visualization Siamese neural network image processing and computer vision Photography TR1-1050 Computer applications to medicine. Medical informatics R858-859.7 Electronic computers. Computer science QA75.5-76.95 Filip Bajić Josip Job Chart Classification Using Siamese CNN
description	In recovering information from the chart image, the first step should be chart type classification. Throughout history, many approaches have been used, and some of them achieve results better than others. The latest articles are using a Support Vector Machine (SVM) in combination with a Convolutional Neural Network (CNN), which achieve almost perfect results with the datasets of few thousand images per class. The datasets containing chart images are primarily synthetic and lack real-world examples. To overcome the problem of small datasets, to our knowledge, this is the first report of using Siamese CNN architecture for chart type classification. Multiple network architectures are tested, and the results of different dataset sizes are compared. The network verification is conducted using Few-shot learning (FSL). Many of described advantages of Siamese CNNs are shown in examples. In the end, we show that the Siamese CNN can work with one image per class, and a 100% average classification accuracy is achieved with 50 images per class, where the CNN achieves only average classification accuracy of 43% for the same dataset.
format	article
author	Filip Bajić Josip Job
author_facet	Filip Bajić Josip Job
author_sort	Filip Bajić
title	Chart Classification Using Siamese CNN
title_short	Chart Classification Using Siamese CNN
title_full	Chart Classification Using Siamese CNN
title_fullStr	Chart Classification Using Siamese CNN
title_full_unstemmed	Chart Classification Using Siamese CNN
title_sort	chart classification using siamese cnn
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/7631a6f4f2ba46cfa6b2a9ece9490a9d
work_keys_str_mv	AT filipbajic chartclassificationusingsiamesecnn AT josipjob chartclassificationusingsiamesecnn
_version_	1718411686978781184

Chart Classification Using Siamese CNN

Ejemplares similares