Video Surveillance for Frictionless Checkout Item Identification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In frictionless checkout systems, identifying purchasable items in a shopping cart can be challenging due to obscuration, as items may be hidden or partially visible, making it difficult to determine which items are selected for purchase.

Innovation Solution

A video surveillance system that analyzes multiple images over time to form a virtual shopping cart, using voting algorithms and deep learning to accurately identify items, even when they are obscured, and tracks items removed from or placed on shelves to determine selection for purchase.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If visual scanning of shopping cart items is performed, then item identification is achieved, but accuracy deteriorates due to item obscuration

Engineering Contradiction:
Improveitem identification accuracyVSAvoiditem visibility
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The system transitions from 2D image analysis to 3D spatial reasoning by incorporating depth information from multiple camera angles. By analyzing item positions across different spatial dimensions and temporal sequences, the system can infer obscured items' identities even when not directly visible in any single frame.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system performs preliminary actions by tracking items from shelf to cart before checkout. Items are identified and registered in the virtual cart during the shopping process, so by the time checkout occurs, all items are already identified regardless of their visibility in final cart images.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If traditional single-image analysis is used, then processing speed is maintained, but identification accuracy deteriorates due to obscured items

Engineering Contradiction:
Improveitem identification accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

Instead of continuous heavy processing, the system uses periodic action by analyzing images at key moments (item pickup, item placement in cart) rather than continuously processing every frame. This reduces computational load while maintaining accuracy through strategic sampling of the shopping process.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

Item identification is performed preliminarily during the shopping process rather than at checkout. This shifts the computational work to distributed time points throughout shopping, avoiding a single processing bottleneck and enabling real-time updates to the virtual cart.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If multiple images over time are analyzed, then identification accuracy improves by resolving obscuration, but data storage and computation requirements increase

Engineering Contradiction:
Improveitem identification accuracyVSAvoiddata storage requirements
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The system extracts only the essential information needed for identification (item features, positions, and relationships) rather than storing complete image data. By extracting and storing only relevant item attributes and spatial-temporal relationships, the system reduces storage requirements while maintaining identification accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system creates a virtual cart as a simplified digital copy of the physical cart contents. This virtual representation stores only necessary item information rather than actual image data, enabling accurate item tracking with minimal storage requirements while maintaining full identification capability.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS11790433B2Constructing shopper carts using video surveillance
Publication Date: 2023.10.17 NCR VOYIX CORP
  • US11790433B2 patent drawing
  • US11790433B2 patent drawing
  • US11790433B2 patent drawing

AI summary

A system can visually track which items in a store are selected for purchase by a shopper. The system can form a virtual shopping cart by analyzing multiple images, over time, to determine which purchasable items are located with the shopper, such as in a physical shopping cart, in a basket, or held by the shopper. By analyzing multiple images, over time, the system can account for items misidentified in one or more images, or fully or partially obscured in one or more images as the shopper traverses the store. Alternatively, the system can form a virtual shopping cart by analyzing instances in which a purchasable item is removed from a shelf or placed on a shelf. Items removed from, but not returned to, a shelf can be considered to be selected for purchase. The system can include a frictionless checkout that charges the shopper for the selected items.