Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Comparison of two or more photographs of products to produce a shopping list from the differences

IP.com Disclosure Number: IPCOM000238537D
Publication Date: 2014-Sep-03
Document File: 2 page(s) / 91K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method for comparing at least two images and creating a shopping list from the differences. Textual data is extracted from product labels and a database lookup system as well as user interaction. This method is based on photographs capturing a current state and is therefore easier to use than text based apps. Gamification is added to encourage a more effective result of the image comparison.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

Page 01 of 2

Comparison of two or more photographs of products to produce a shopping list from the differences

Disclosed is a method to generate a shopping list by comparing images of a storeroom from two different dates. This method can be used in general stock rooms (pantry, medicines, laundry or paper products, office stockrooms) where Radio-frequency identification systems are too costly.

Procedure:

1. Take picture of full pantry for "golden photo"

Note: multiple pictures can be used, with all results being stored in a "golden" database.

Pull out thumbnails of individual items (boxes, cans, jars, bags), creating a golden list of images. Extract identifying text, like brand name and size of contents, from the photo.

Classify each image/item (cereal, canned meat, juice, crackers)

This classification is done using a customizable database to map general household products/foods with a standard subset of classes (cereal, canned meat, etc)

    New items unknown to the database are added thru a user interface that asks the user to classify the item. The tool is interactive allowing the user to
review the items,

prioritize them,
comment on them (e.g. indicate brand loyalty or not),
set a purchase threshold (when the flour bin is only 20% full or user indication that they are down to the last box of crackers)


2. Before grocery shopping day, take a second picture of an emptier pantry for "hungry photo"

Note: multiple pictures can be used, with all results being stored in a "hungry" database. Repeat step 1 above with the new picture

Pull out thumbnails of individual items (boxes, cans, jars, bags) creating a hungry list of images Identify text like brand name and size from the photo
Classify each image/item (cereal, canned meat, juice, crackers)

Allow the user to review.

3. Compare golden and hungry lists of images by comparing the associated databases. Eliminate identical matches in the golden vs hungry data...