Beverage Detection Using Homography and Zone Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing object detection and tracking systems face challenges in real-time processing of multiple objects in busy environments, such as stores, due to computational intensity and the inability to determine physical locations of objects within images, especially when objects have non-uniform weights or are misplaced.

Innovation Solution

A tracking system that generates homographies to map camera pixels to physical locations, enables handoff of tracking information between sensors, detects shelf interactions, identifies non-uniform weight objects, and corrects homography errors using markers and distance measuring devices, allowing for efficient object assignment and re-identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional object detection and tracking systems are used in busy environments, then they can identify objects in images, but the computational intensity increases significantly and real-time processing becomes difficult

Engineering Contradiction:
Improveobject identification accuracyVSAvoidreal-time processing capability
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system divides the tracking space into multiple zones and uses multiple sensors to cover different areas. Each sensor tracks objects within its specific zone, reducing the computational burden on any single sensor while maintaining comprehensive coverage of the entire environment

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from 2D image pixel coordinates to 3D physical world coordinates by incorporating depth information and camera calibration data. This dimensional transformation enables more efficient spatial reasoning and reduces computational complexity in tracking multiple objects

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

2Loss of information

If conventional tracking systems are used, then objects can be detected in images, but the physical location of objects within the image cannot be determined

Engineering Contradiction:
Improvephysical location informationVSAvoidtracking system complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system introduces a homography matrix as an intermediary that maps between image pixel coordinates and physical world coordinates. This mathematical transformation layer enables accurate physical location determination without requiring complex additional hardware, bridging the gap between 2D images and 3D space

Inventive Principle:
Principle #24Intermediary (Mediator)

3Area of stationary object

If multiple sensors are deployed to track objects in large spaces, then coverage is improved, but the handoff of tracking information between sensors becomes complex

Engineering Contradiction:
Improvetracking coverage areaVSAvoidsensor handoff complexity
Core Design Contradiction:
Area of stationary objectVSDevice complexity

Solution Approach 1:

The system merges tracking information from multiple sensors by maintaining a unified object database that consolidates data from all sensor sources. This centralized approach simplifies the handoff process between sensors and enables seamless tracking across the entire monitored area without complex inter-sensor communication protocols

Inventive Principle:
Principle #5Merging (Combining)

4Productivity

If self-serve beverage detection is implemented, then checkout efficiency is improved, but the ability to handle non-uniform weight objects and misplaced items becomes challenging

Engineering Contradiction:
Improvecheckout efficiencyVSAvoidobject assignment accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system uses feedback from multiple sensors including weight sensors and image analysis to continuously verify object assignments. When discrepancies are detected (such as non-uniform weight objects or misplaced items), the system adjusts its tracking and assignment algorithms to maintain accuracy, ensuring reliable self-serve beverage detection

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11887372B2Image-based self-serve beverage detection and assignment
Publication Date: 2024.01.30 7-ELEVEN INC
  • US11887372B2 patent drawing
  • US11887372B2 patent drawing
  • US11887372B2 patent drawing

AI summary

An image sensor is positioned such that a field-of-view of the sensor encompasses portion of a beverage machine. The field-of-view includes a first zone associated with operating the beverage machine to dispense the beverage and a second zone in which a cup is placed to receive the beverage. A beverage assignment subsystem receives angled-view images from the image sensor. An event associated with an object entering one or both of the first zone and the second zone is detected. In image(s) associated with a start of the event, it is determined that both a hand of a person enters the first zone and the cup is placed in the second zone. If the cup remained in the second zone for at least a threshold time, the beverage is assigned to the person whose hand entered the first zone.