Browse Prior Art Database

Image identification of foods to calculate nutrition/calories leveraging machine learning capabilities

IP.com Disclosure Number: IPCOM000242182D
Publication Date: 2015-Jun-23
Document File: 2 page(s) / 89K

Publishing Venue

The IP.com Prior Art Database

Abstract

A system and method for utilizing artificial intelligence for image recognition of foods to calculate nutrition and calories is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Image identification of foods to calculate nutrition/calories leveraging machine learning capabilities

Disclosed is a system and method for utilizing artificial intelligence for image recognition of foods to calculate nutrition and calories.

Currently, the ability to accurately determine the calories/nutrition values of food is limited to knowing the exact food, amount, and servings per recipe. Knowing all that information is sometimes very difficult to impossible to obtain (especially in mixed dishes). The ability to gather most if not all that information via a photo of the food

would be of immense help to people wanting to regulate nutrition/caloric intake or even

just for knowing the ingredients for food allergies.

To assist with this determination, a system and method is disclosed that utilizes machine learning or artificial intelligence to gather as much information as possible about the foods via many sources. These source may include a photo, location, social crowd sourcing, food isolation and identification techniques (for example, a photo of an actual recipe), and etc. By combining all the available data and applying machine learning techniques to ask questions and refine accuracy and confidence, the user assists the system to build up a fairly easy and accurate way to measure and understand the calories/nutrition values of the user's daily food intake.

The following are examples of aspects of the disclosed system and method applied to an image used to determine more accurate calories/nutrition values of foods:

Sectioning - Ability to section off one food from another on a plate of foods. Detection - Recognition of standard foods and identifying differences of amounts of each food.

Selection - Allowing the user to select the type of food (for example, system identified a food as a green bean and user provides feedback to identify the food as asparagus).

Refinement - Knowing if the food rich or light, steamed (for example, broccoli vs buttered broccoli and user provides feedback to adjust details of how the food

was prepare...