Exception Handling in Symbol Reader Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automated machine-readable symbol readers often fail to accurately identify objects due to misreads or non-reads, leading to exceptions such as damaged symbols, obscuration, misalignment, or mismatched visual characteristics, which can slow down scanning processes in applications like checkout or baggage handling.
Innovation Solution
An automated machine-readable symbol reader system that captures multiple images of an exception object from different views, extracts visual features, compares them with a database of known objects, and displays the most descriptive images to an operator to facilitate rapid identification and resolution of exceptions, using techniques like scale-invariant feature transform (SIFT) and geometric point features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automated machine-readable symbol readers are used to scan objects, then scanning speed and automation are improved, but misreads and non-reads occur due to damaged symbols, obscuration, misalignment, or visual mismatches
Solution Approach 1:
A camera captures images of exception objects and displays them on a display device as an intermediary between the automated reader and the operator. This allows the operator to visually identify objects that the automated system cannot read, resolving the contradiction by maintaining high automation speed while providing a visual mediation mechanism for handling read failures.
Solution Approach 2:
The system provides feedback to the operator by displaying images of objects that caused read errors. This feedback loop allows the operator to identify and correct exceptions, improving reliability without sacrificing the automated scanning speed for normal objects.
2Reliability
If operators manually rescan or input data for exception objects, then read accuracy is improved, but processing time increases
Solution Approach 1:
The camera automatically captures images of exception objects before the operator needs to identify them. This preliminary action prepares the visual information in advance, allowing the operator to quickly identify objects without manual rescan or data input, thus improving identification accuracy while minimizing time loss.
Solution Approach 2:
Instead of requiring the operator to physically handle or rescan the exception object, the system creates a visual copy (image) of the object and displays it. This copying approach allows rapid identification without the time-consuming manual processes, maintaining high accuracy while reducing handling time.
3Reliability
If multiple images of exception objects are captured and displayed, then identification accuracy is improved, but device complexity increases
Solution Approach 1:
The camera serves multiple functions: it captures images of exception objects, provides visual feedback to operators, and integrates with the existing automated reading system. This multi-functionality approach improves identification accuracy without requiring separate dedicated devices, thus limiting the increase in system complexity.
Data Source
AI summary
Systems and methods for exception handling in an automated machine-readable symbol reader system having a machine-readable symbol reader that captures machine-readable symbols within a view volume. One or more image capture devices obtain a plurality of images of an exception object in response to an exception generated in the view volume. A processor receives the images, identities visual object recognition features from each image, and compares the features to determine one or more descriptive measures indicative of a likelihood that an operator (e.g., store employee) will be able to identify the exception object by viewing the image. The processor displays at least one of the images (e.g., the most descriptive image) on a display device according to the descriptive measure so that an operator can rapidly identify the identity of the exception object and take steps to resolve the exception.