CogKnife: Food recognition from their cutting sounds

Takamichi Kojima, Takashi Ijiri, Jeremy White, Hidetomo Kataoka, Akira Hirabayashi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this study, we present 'CogKnife', a knife device which can identify food. For this, a small microphone is attached to a knife, which records the cutting sound of food. We extract spectrograms from the cutting sounds and use them as feature vectors to train a classifier. This study used the k-Nearest Neighbor method (k-NN), the support vector machine (SVM) and the convolutional neural network (CNN) to verify differences of the classification methods. To evaluate the accuracy of our technique, we performed classification experiments with six kinds of foods (apples, bananas, cabbages, leeks and peppers) in a laboratory environment. From 20-fold cross validation, we confirmed high recognition accuracies, such as 83% with k-NN, 95% with SVM and 89% with CNN.

LanguageEnglish
Title of host publication2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509015528
DOIs
StatePublished - 2016 Sep 22
Externally publishedYes
Event2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016 - Seattle, United States
Duration: 2016 Jul 112016 Jul 15

Other

Other2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016
CountryUnited States
CitySeattle
Period16/7/1116/7/15

Fingerprint

Acoustic waves
Support vector machines
Neural networks
Microphones
Classifiers
Experiments

Keywords

  • Cooking support
  • Food recognition
  • Machine learning
  • Pattern recognition
  • Sound recognition

ASJC Scopus subject areas

  • Signal Processing
  • Media Technology
  • Computer Vision and Pattern Recognition

Cite this

Kojima, T., Ijiri, T., White, J., Kataoka, H., & Hirabayashi, A. (2016). CogKnife: Food recognition from their cutting sounds. In 2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016 [7574741] Institute of Electrical and Electronics Engineers Inc.. DOI: 10.1109/ICMEW.2016.7574741

CogKnife : Food recognition from their cutting sounds. / Kojima, Takamichi; Ijiri, Takashi; White, Jeremy; Kataoka, Hidetomo; Hirabayashi, Akira.

2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016. Institute of Electrical and Electronics Engineers Inc., 2016. 7574741.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kojima, T, Ijiri, T, White, J, Kataoka, H & Hirabayashi, A 2016, CogKnife: Food recognition from their cutting sounds. in 2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016., 7574741, Institute of Electrical and Electronics Engineers Inc., 2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016, Seattle, United States, 16/7/11. DOI: 10.1109/ICMEW.2016.7574741
Kojima T, Ijiri T, White J, Kataoka H, Hirabayashi A. CogKnife: Food recognition from their cutting sounds. In 2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016. Institute of Electrical and Electronics Engineers Inc.2016. 7574741. Available from, DOI: 10.1109/ICMEW.2016.7574741
Kojima, Takamichi ; Ijiri, Takashi ; White, Jeremy ; Kataoka, Hidetomo ; Hirabayashi, Akira. / CogKnife : Food recognition from their cutting sounds. 2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{9526e9c231ec4560ae18f4f8681b7161,
title = "CogKnife: Food recognition from their cutting sounds",
abstract = "In this study, we present \{textquoteleft}CogKnife\{textquoteleft}, a knife device which can identify food. For this, a small microphone is attached to a knife, which records the cutting sound of food. We extract spectrograms from the cutting sounds and use them as feature vectors to train a classifier. This study used the k-Nearest Neighbor method (k-NN), the support vector machine (SVM) and the convolutional neural network (CNN) to verify differences of the classification methods. To evaluate the accuracy of our technique, we performed classification experiments with six kinds of foods (apples, bananas, cabbages, leeks and peppers) in a laboratory environment. From 20-fold cross validation, we confirmed high recognition accuracies, such as 83\{%} with k-NN, 95\{%} with SVM and 89\{%} with CNN.",
keywords = "Cooking support, Food recognition, Machine learning, Pattern recognition, Sound recognition",
author = "Takamichi Kojima and Takashi Ijiri and Jeremy White and Hidetomo Kataoka and Akira Hirabayashi",
year = "2016",
month = "9",
day = "22",
doi = "10.1109/ICMEW.2016.7574741",
language = "English",
booktitle = "2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - GEN

T1 - CogKnife

T2 - Food recognition from their cutting sounds

AU - Kojima,Takamichi

AU - Ijiri,Takashi

AU - White,Jeremy

AU - Kataoka,Hidetomo

AU - Hirabayashi,Akira

PY - 2016/9/22

Y1 - 2016/9/22

N2 - In this study, we present 'CogKnife', a knife device which can identify food. For this, a small microphone is attached to a knife, which records the cutting sound of food. We extract spectrograms from the cutting sounds and use them as feature vectors to train a classifier. This study used the k-Nearest Neighbor method (k-NN), the support vector machine (SVM) and the convolutional neural network (CNN) to verify differences of the classification methods. To evaluate the accuracy of our technique, we performed classification experiments with six kinds of foods (apples, bananas, cabbages, leeks and peppers) in a laboratory environment. From 20-fold cross validation, we confirmed high recognition accuracies, such as 83% with k-NN, 95% with SVM and 89% with CNN.

AB - In this study, we present 'CogKnife', a knife device which can identify food. For this, a small microphone is attached to a knife, which records the cutting sound of food. We extract spectrograms from the cutting sounds and use them as feature vectors to train a classifier. This study used the k-Nearest Neighbor method (k-NN), the support vector machine (SVM) and the convolutional neural network (CNN) to verify differences of the classification methods. To evaluate the accuracy of our technique, we performed classification experiments with six kinds of foods (apples, bananas, cabbages, leeks and peppers) in a laboratory environment. From 20-fold cross validation, we confirmed high recognition accuracies, such as 83% with k-NN, 95% with SVM and 89% with CNN.

KW - Cooking support

KW - Food recognition

KW - Machine learning

KW - Pattern recognition

KW - Sound recognition

UR - http://www.scopus.com/inward/record.url?scp=84992129694&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992129694&partnerID=8YFLogxK

U2 - 10.1109/ICMEW.2016.7574741

DO - 10.1109/ICMEW.2016.7574741

M3 - Conference contribution

BT - 2016 IEEE International Conference on Multimedia and Expo Workshop, ICMEW 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -