• Save
 

Vision-Based Nutrtion Information Extraction from Product Packages on Smartphones

on

  • 514 views

 

Statistics

Views

Total Views
514
Views on SlideShare
439
Embed Views
75

Actions

Likes
0
Downloads
0
Comments
0

20 Embeds 75

http://vkedco.blogspot.com 29
http://www.vkedco.blogspot.com 13
http://vkedco.blogspot.in 6
http://vkedco.blogspot.mx 4
http://www.vkedco.blogspot.in 3
http://vkedco.blogspot.com.ar 2
http://vkedco.blogspot.com.br 2
http://vkedco.blogspot.com.es 2
http://vkedco.blogspot.jp 2
http://reader.aol.com 2
http://vkedco.blogspot.fr 1
http://vkedco.blogspot.de 1
http://www.vkedco.blogspot.co.uk 1
http://vkedco.blogspot.ru 1
http://vkedco.blogspot.kr 1
http://vkedco.blogspot.gr 1
http://vkedco.blogspot.co.uk 1
http://vkedco.blogspot.pt 1
http://cloud.feedly.com 1
http://vkedco.blogspot.ca 1
More...

Accessibility

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Vision-Based Nutrtion Information Extraction from Product Packages on Smartphones Vision-Based Nutrtion Information Extraction from Product Packages on Smartphones Presentation Transcript

  • Vision-Based Nutrition Information Extraction from Product Packages on Smartphones Vladimir Kulyukin Department of Computer Science Utah State University
  • Outline ● Introduction – Motivation – Critical Barriers – From Robotic Shopping Carts to PNUTS ● What information can be reliably extracted from product packages and how? – DOGs eat skewed barcodes – Lines & corners textchunk nutrition labels ● Preliminary Experiments & Conclusions
  • Introduction
  • Motivation ● U.S. Department of Agriculture estimates that U.S. residents have increased their caloric intake by 523 calories per day since 1970 ● Mismanaged diets are estimated to account for 30-35% of cancer and diabetes cases ● Many nutritionists consider proactive nutrition management to be a key factor in controlling cancer and diabetes ● Nutrition information needs to be available to low vision and blind individuals
  • Critical Barriers ● Lack of automated, real-time nutrition information analysis: Barcodes (especially skewed ones) and nutrition labels have characteristics that impede timely detection and/or adequate comprehension ● Lack of automated, real-time context-sensitive nutrition decision support: No coupling of purchasing decisions to patients' current locations, ODLs, and PHRs ● Lack of automated, real-time nutrition intake recording: manual nutrition intake recording is time- consuming and error-prone, especially on smartphones
  • RoboCart ShopTalk ShopMobile I ShopMobile II PNUTS From Robotic Shopping Carts to PNUTS 2003-05 2006-08 2008-10 2010-12 2013-Now
  • What Information Can Be Reliably Extracted from Product Packages and How?
  • Skewed Barcodes & Nutrition Labels in Persuasive NUTrition System PNUTS ● Product Names ● Nutrition Facts ● Caloric Content ● Ingredients
  • DOG: Dominant Orientation of Gradient 20 x 20 mask 50 x 50 mask
  • DOGs Eat Skewed Barcodes
  • Lines & Corners Textchunk Nutrition Labels
  • Preliminary Experiments & Results
  • Experiments ● The barcode localization performance of both algorithms was evaluated on a sample of 1,066 images of skewed UPC barcodes on bags, boxes, bottles, cans, books, and images with no barcodes ● NL Localization performance was evaluated on a sample of 45 images ● Evaluation was done Android 2.3.6 and 4.2
  • Results ● DOG performance on a sample of 1,066 images: precision (approximately 90%); true negatives (97%); false positives (2.5%) ● NL localization performance on a sample of 45 images: a mean error of 1% ● Textchunking performance on a sample of 45 images: precision (70%); recall (93%); specificity (89%); accuracy (85%)
  • References ● Kulyukin, V. and Zaman T. Vision-Based Localization of Skewed UPC Barcodes on Smartphones. In Proceedings of IPCV 2013, ISBN 1-60132-252-6, CSREA Press, Las Vegas, NV. ● Kulyukin, V., , Kutiyanawala, A., Zaman, T, & Clyde, S. Vision- Based Localization & Text Chunking of Nutrition Fact Tables on Android Smartphones. In Proceedings of IPCV 2013, ISBN 1-60132- 252-6, CSREA Press, Las Vegas, NV. ● Kulyukin, V. and Vanka, A. Skip Trie Matching for Real Time OCR Error Correction on Android Smartphones. In Proceedings of DICTAP 2013, ISBN 978-0-9891305-0-9, SDIWC Press, Ostrava, Czech Repulic. ● www.vkedco.blogspot.com