Machine Vision Object Tallying to Reduce Checkout Handling Time
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current retail checkout systems are inefficient, causing physical stress and potential repetitive motion injuries due to the need for frequent object handling and orientation, and often require complex equipment, limiting the size and type of objects that can be processed.
Innovation Solution
A data reading system that uses a machine vision system with a field of view to capture images of objects on a conveyor belt or counter, allowing for automatic recognition and tallying without the need for manual scanning, using multiple imagers and image processing techniques to identify optical codes and track object positions, with visual feedback to customers and cashiers for unrecognized items.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual scanning with optical scanner is used, then object identification can be achieved, but processing time increases and physical stress on cashiers increases
Solution Approach 1:
The patent replaces the mechanical manual scanning process with an automated machine vision system. Multiple cameras capture images of objects on the conveyor belt, and image processing algorithms automatically identify and tally items without requiring cashiers to manually grasp, orient, and scan each object, thereby eliminating the time loss and physical stress associated with manual operation
Solution Approach 2:
The system enables self-service automated identification where the machine vision system independently performs object recognition and tallying without human intervention. The system automatically processes objects as they pass through the field of view, identifying and counting items without requiring cashier actions, thus improving productivity while reducing time loss
2Extent of automation
If tunnel-based scanning systems are used, then automated processing is achieved, but the physical size limits what objects can be processed
Solution Approach 1:
The patent transitions from a confined tunnel-based scanning approach to an open field-of-view imaging system. By using multiple cameras positioned to capture a wide area, the system can process objects of various sizes and shapes without the spatial constraints of a tunnel, thereby maintaining automation while significantly improving adaptability to different object dimensions
Solution Approach 2:
The machine vision system with multiple imagers is designed to handle a universal range of objects regardless of size, shape, or orientation. The system can process everything from small items to large packages within its field of view, making it versatile for different retail environments and object types while maintaining automated processing
3Extent of automation
If item-by-item processing through tunnels is used, then automated tallying is achieved, but processing speed is reduced
Solution Approach 1:
The patent implements continuous imaging and processing as objects move through the field of view. Multiple cameras continuously capture images, and the system processes multiple objects simultaneously in parallel rather than sequentially, maintaining continuous automated tallying while significantly improving processing speed by eliminating the bottlenecks of item-by-item tunnel processing
4Extent of automation
If complex machine recognition systems are used, then automated identification is achieved, but system complexity and cost increase
Solution Approach 1:
The patent uses multiple camera copies to capture images of objects from different positions simultaneously. Instead of using a single complex recognition system, the approach uses simpler camera units that capture visual information, which is then processed through image analysis algorithms to achieve automated identification, reducing overall system complexity while maintaining automation
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for tallying objects (40) presented for purchase preferably images the objects (40) with a machine vision system (5) while the objects (40) are still, or substantially still. Images of the objects may be used to recognize the objects (40) and to collect information about each object (40), such as the price. A pre-tally list may be generated and displayed to a customer showing the customer the cost of the recognized objects. A prompt on a customer display (45) may be given to urge a customer to re-orient unrecognized objects (40) to assist the machine vision system with recognizing such unrecognized objects (40). A tallying event, such as removing a recognized object (40) from the machine vision system's (5) field of view, preferably automatically tallies recognized objects (40) so it is not necessary for a cashier to scan or otherwise input object information into a point of sale system.