Dietary assessment system and method

• US 8,363,913 B2
• Filed: 09/03/2009
• Issued: 01/29/2013
• Est. Priority Date: 09/05/2008
• Status: Active Grant

• Current Assignee: Purdue Research Foundation (The Trustees of Purdue University (d/b/a Purdue University))
• Original Assignee: Purdue Research Foundation (The Trustees of Purdue University (d/b/a Purdue University))
• Inventors: Boushey, Carol, Delp, Edward John, Ebert, David Scott, Lutes, Kyle DelMar, Kerr, Deborah
• Primary Examiner(s): PHAM, HOA Q
• Application Number:
  US12/553,597
• Publication Number:
  US 20100111383A1
• Time in Patent Office: 1,244 Days
• Field of Search: 382/110, 382/203, 382/286, 382/100, 382/128, 382/218, 600/593, 600/301, 600/300, 345/810, 707/709, 128921-923, 128897-898, 434/127, 434/247
• US Class Current: 382/128
• CPC Class Codes: A23L 33/30 Dietetic or nutritional methods, e.g. for losing weight A23L33/10 takes precedence
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06V 20/68 Food, e.g. fruit or vegetables

Abstract

The present system and method provides a more precise way to record food and beverage intake than traditional methods. The present disclosure provides custom software for use in mobile computing devices that include a digital camera. Photos captured by mobile digital devices are analyzed with image processing and comparisons to certain databases to allow a user to discretely record foods eaten. Specifically, the user captures images of the meal or snack before and after eating. The foods pictured are identified. Image processing software may identify the food or provide choices for the user. Once a food is identified and volume of the food is estimated, nutrient databases are used for calculating final portion sizes and nutrient totals.

Claims

1. A method of measuring food intake of a user, the method comprising:

providing a mobile device having a built-in camera;

capturing a first image of food to be eaten by the user using the mobile device;

calibrating a size of the food items in the image;

segmenting the first image into one or more segmented regions where different food items are located;

identifying the food of the segmented regions;

estimating the volume of food in the segmented regions of the first image.

2. The method of claim 1, further comprising calculating the nutritional information of the food items of the first image and communicating this information to the user before the user starts eating.

3. The method of claim 1, further comprising determining an amount of each food item consumed by the user by capturing a second image of any remaining food items after the user has eaten and repeating the calibrating, segmenting, identifying, and estimating steps for the food items in the second image, and then subtracting the remaining volume of each food item in the second image from the original volume of each food item in the first image to determine the amount of each food item eaten.

4. The method of claim 3, further comprising calculating nutritional information based on the amount and identity of each food item eaten.

5. The method of claim 1, further comprising the selection by the user of an option given by the mobile device for indicating that all food was consumed.

6. The method of claim 1 where in the mobile device is network-connected and allows for the transferring and receiving of information.

7. The method of claim 1 wherein the user labels the identity of the food on at least the first image.

8. The method of claim 1 wherein the step of identifying the food comprises comparing the image of the food taken by the user to images of known food items in a food database.

9. The method of claim 1 wherein the image includes a fiducial marker of a known size as a reference for size or volume calculations.

10. The method of claim 1 for measuring beverage intake rather than food intake.

11. The method of claim 1 wherein the user can enter food identity and volume estimation manually into the mobile device the when an image cannot be captured or is insufficient;

and wherein the nutritional content of food items identified can be calculated using known measurements for food items in a food database.