Presentation 2010-07-23
Classification of Food Images with the Visual Words, Constructed from Multiple Image Features
Seiichi KONYA, Katsuyoshi TANABE, Tadasu UCHIYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recently, health guidance support via meal images are available and network communications about cooking has becoming popular. Extracting information from food images has been required. We propose the food image classification algorithms using visual words. Previously, visual words are constructed from single feature, such as intensity edge. In contrast, our algorithms integrate multiple features, such as average color, lightness texture and color tone, efficiently. We construct the dataset, which consists of 13 food categories, about 900 images, and evaluate our algorithms on it. In the experiments, the classification accuracies, those contain correct answers in first, second and third ranks, are 58.7, 74.4 and 83.3 percent respectively.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) image classification / local features / visual words
Paper # LOIS2010-16
Date of Issue

Conference Information
Committee LOIS
Conference Date 2010/7/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Life Intelligence and Office Information Systems (LOIS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Classification of Food Images with the Visual Words, Constructed from Multiple Image Features
Sub Title (in English)
Keyword(1) image classification
Keyword(2) local features
Keyword(3) visual words
1st Author's Name Seiichi KONYA
1st Author's Affiliation NTT Cyber Solutions Laboratories()
2nd Author's Name Katsuyoshi TANABE
2nd Author's Affiliation NTT Cyber Solutions Laboratories
3rd Author's Name Tadasu UCHIYAMA
3rd Author's Affiliation NTT Cyber Solutions Laboratories
Date 2010-07-23
Paper # LOIS2010-16
Volume (vol) vol.110
Number (no) 141
Page pp.pp.-
#Pages 5
Date of Issue