The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Quando as pessoas aprendem um artesanato com conteúdos instrucionais como livros, vídeos e páginas da web, muitas delas desistem no meio do caminho porque o conteúdo nem sempre garante como fazê-lo. Este estudo tem como objetivo fornecer aos alunos de origami, principalmente iniciantes, feedbacks sobre suas operações de dobramento. É proposta uma abordagem para reconhecer o estado do aluno usando uma única câmera de visão superior e apontar os erros cometidos durante a operação de dobramento do origami. Primeiro, é definido um modelo de instrução que armazena operações de dobramento fáceis de seguir. Em segundo lugar, é proposto um método para reconhecer o estado da folha de papel de origami do aluno. Terceiro, um método para detectar erros cometidos pelo aluno por meio da detecção de anomalias usando um classificador de máquina de vetor de suporte de uma classe (SVM de uma classe) (usando o progresso de dobramento e a diferença entre a forma de origami do aluno e a forma correta) é proposto. Como existem ruídos nas imagens da câmera devido a sombras e oclusões causadas pelas mãos do aluno, as formas da folha de origami nem sempre são extraídas com precisão. Para treinar o classificador SVM de uma classe com alta precisão, é proposto um método de limpeza de dados que separa automaticamente os quadros de vídeo com ruídos. Além disso, a utilização das estatísticas de características extraídas dos frames de uma janela deslizante permite reduzir a influência dos ruídos. O método proposto foi experimentalmente demonstrado ser suficientemente preciso e robusto contra ruídos, e sua taxa de falsos alarmes (taxa de falsos positivos) pode ser reduzida a zero. Exigindo apenas uma única câmera e papel de origami comum, o método proposto permite monitorar erros cometidos por alunos de origami e apoiar sua autoaprendizagem.
Hiroshi SHIMANUKI
Nagoya Industrial Science Research Institute
Toyohide WATANABE
Daido University
Koichi ASAKURA
Daido University
Hideki SATO
Kyushu University
Taketoshi USHIAMA
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Hiroshi SHIMANUKI, Toyohide WATANABE, Koichi ASAKURA, Hideki SATO, Taketoshi USHIAMA, "Anomaly Detection of Folding Operations for Origami Instruction with Single Camera" in IEICE TRANSACTIONS on Information,
vol. E103-D, no. 5, pp. 1088-1098, May 2020, doi: 10.1587/transinf.2019EDP7242.
Abstract: When people learn a handicraft with instructional contents such as books, videos, and web pages, many of them often give up halfway because the contents do not always assure how to make it. This study aims to provide origami learners, especially beginners, with feedbacks on their folding operations. An approach for recognizing the state of the learner by using a single top-view camera, and pointing out the mistakes made during the origami folding operation is proposed. First, an instruction model that stores easy-to-follow folding operations is defined. Second, a method for recognizing the state of the learner's origami paper sheet is proposed. Third, a method for detecting mistakes made by the learner by means of anomaly detection using a one-class support vector machine (one-class SVM) classifier (using the folding progress and the difference between the learner's origami shape and the correct shape) is proposed. Because noises exist in the camera images due to shadows and occlusions caused by the learner's hands, the shapes of the origami sheet are not always extracted accurately. To train the one-class SVM classifier with high accuracy, a data cleansing method that automatically sifts out video frames with noises is proposed. Moreover, using the statistics of features extracted from the frames in a sliding window makes it possible to reduce the influence by the noises. The proposed method was experimentally demonstrated to be sufficiently accurate and robust against noises, and its false alarm rate (false positive rate) can be reduced to zero. Requiring only a single camera and common origami paper, the proposed method makes it possible to monitor mistakes made by origami learners and support their self-learning.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019EDP7242/_p
Copiar
@ARTICLE{e103-d_5_1088,
author={Hiroshi SHIMANUKI, Toyohide WATANABE, Koichi ASAKURA, Hideki SATO, Taketoshi USHIAMA, },
journal={IEICE TRANSACTIONS on Information},
title={Anomaly Detection of Folding Operations for Origami Instruction with Single Camera},
year={2020},
volume={E103-D},
number={5},
pages={1088-1098},
abstract={When people learn a handicraft with instructional contents such as books, videos, and web pages, many of them often give up halfway because the contents do not always assure how to make it. This study aims to provide origami learners, especially beginners, with feedbacks on their folding operations. An approach for recognizing the state of the learner by using a single top-view camera, and pointing out the mistakes made during the origami folding operation is proposed. First, an instruction model that stores easy-to-follow folding operations is defined. Second, a method for recognizing the state of the learner's origami paper sheet is proposed. Third, a method for detecting mistakes made by the learner by means of anomaly detection using a one-class support vector machine (one-class SVM) classifier (using the folding progress and the difference between the learner's origami shape and the correct shape) is proposed. Because noises exist in the camera images due to shadows and occlusions caused by the learner's hands, the shapes of the origami sheet are not always extracted accurately. To train the one-class SVM classifier with high accuracy, a data cleansing method that automatically sifts out video frames with noises is proposed. Moreover, using the statistics of features extracted from the frames in a sliding window makes it possible to reduce the influence by the noises. The proposed method was experimentally demonstrated to be sufficiently accurate and robust against noises, and its false alarm rate (false positive rate) can be reduced to zero. Requiring only a single camera and common origami paper, the proposed method makes it possible to monitor mistakes made by origami learners and support their self-learning.},
keywords={},
doi={10.1587/transinf.2019EDP7242},
ISSN={1745-1361},
month={May},}
Copiar
TY - JOUR
TI - Anomaly Detection of Folding Operations for Origami Instruction with Single Camera
T2 - IEICE TRANSACTIONS on Information
SP - 1088
EP - 1098
AU - Hiroshi SHIMANUKI
AU - Toyohide WATANABE
AU - Koichi ASAKURA
AU - Hideki SATO
AU - Taketoshi USHIAMA
PY - 2020
DO - 10.1587/transinf.2019EDP7242
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2020
AB - When people learn a handicraft with instructional contents such as books, videos, and web pages, many of them often give up halfway because the contents do not always assure how to make it. This study aims to provide origami learners, especially beginners, with feedbacks on their folding operations. An approach for recognizing the state of the learner by using a single top-view camera, and pointing out the mistakes made during the origami folding operation is proposed. First, an instruction model that stores easy-to-follow folding operations is defined. Second, a method for recognizing the state of the learner's origami paper sheet is proposed. Third, a method for detecting mistakes made by the learner by means of anomaly detection using a one-class support vector machine (one-class SVM) classifier (using the folding progress and the difference between the learner's origami shape and the correct shape) is proposed. Because noises exist in the camera images due to shadows and occlusions caused by the learner's hands, the shapes of the origami sheet are not always extracted accurately. To train the one-class SVM classifier with high accuracy, a data cleansing method that automatically sifts out video frames with noises is proposed. Moreover, using the statistics of features extracted from the frames in a sliding window makes it possible to reduce the influence by the noises. The proposed method was experimentally demonstrated to be sufficiently accurate and robust against noises, and its false alarm rate (false positive rate) can be reduced to zero. Requiring only a single camera and common origami paper, the proposed method makes it possible to monitor mistakes made by origami learners and support their self-learning.
ER -