Self-Checkout Item Recognition Using Motion-Changed Pixels

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Checkout systems struggle to accurately identify multiple items in a crowded area of interest, leading to misidentification or failure to recognize items, and require lengthy processing times.

Innovation Solution

An imaging system that captures and compares image datasets before and after a motion event to identify a group of pixels associated with the detected motion, using techniques such as segmenting and analyzing the segmented image dataset to verify the object based on pixel characteristics, and optionally employing an artificial neural network to confirm the object matches a decoded barcode.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional checkout systems analyze all pixels in the area of interest to identify multiple items, then item identification may be achieved, but computational intensity and processing time increase significantly

Engineering Contradiction:

Improveitem identification accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the area of interest into multiple zones based on spatial relationships between items. By dividing the complex scene into smaller manageable segments, the system reduces computational intensity while maintaining accurate item identification. Each segment can be processed independently, significantly reducing overall processing time compared to analyzing all pixels uniformly.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts and focuses computational resources on specific regions containing items of interest rather than processing the entire area of interest uniformly. By identifying and extracting relevant item regions from the crowded scene, the system reduces the pixel analysis scope while maintaining identification accuracy, thereby reducing processing time.

Inventive Principle:

Principle #2Taking out (Extraction)

2Measurement precision

If traditional checkout systems analyze all pixels in the area of interest to identify multiple items, then item identification may be achieved, but computational intensity increases significantly

Engineering Contradiction:

Improveitem identification accuracyVSAvoidcomputational intensity

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by analyzing only the necessary portions of the image data required for item identification rather than processing all pixels. By focusing computational resources on relevant item regions and using motion detection to identify changed areas, the system reduces computational intensity while maintaining sufficient identification accuracy.

Inventive Principle:

Principle #16Partial or excessive action

3Productivity

If checkout systems use motion detection to identify changed pixels, then processing time and computational intensity are reduced, but item identification accuracy may be compromised in crowded areas

Engineering Contradiction:

Improveprocessing efficiencyVSAvoiditem identification accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent employs feedback mechanisms where the system continuously monitors the bagging area, compares current state with previous states, and adjusts its analysis based on detected changes. Motion detection triggers targeted analysis of changed regions, and the system refines its identification by comparing against known item characteristics, thereby maintaining accuracy while improving processing efficiency.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent performs preliminary actions by pre-processing image data to identify motion and changed regions before conducting detailed item analysis. By detecting motion first and then focusing analysis on specific changed pixel groups, the system prepares the data in advance, reducing the scope of detailed analysis required and maintaining accuracy while improving efficiency.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20260004271A1Method to Recognize Items During Self-Checkout

Publication Date: 2026.01.01 ZEBRA TECHNOLOGIES CORP

AI summary

Systems and methods to recognize items during self-checkout are disclosed herein. An example system includes: one or more processors; one or more sensors; one or more image acquisition assemblies; and one or more memories including computer-executable instructions stored thereon that cause the system to: detect, via the one or more sensors, a motion within an area of interest of one or more areas of interest; obtain, from the one or more image acquisition assemblies, a first image dataset of the area of interest captured prior to the detected motion; obtain, from the one or more image acquisition assemblies, a second image dataset of the area of interest captured after the detected motion; compare the first image dataset and the second image dataset to identify a group of pixels associated with the detected motion; and perform, based on the group of pixels identified, one or more actions.