Icon Detection Using Adaptive Scanline Filtering for Overlapping Barcode Resolution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in accurately identifying and decoding graphical icons and encoded signals within images, particularly in point of sale systems, where conflicts arise from overlapping or multiple barcodes on products, leading to incorrect product identification and pricing issues.
Innovation Solution
The method involves generating thresholded scanlines, extracting features from candidate icons, evaluating element relationships and angles, and adaptively selecting image blocks to determine if they contain encoded signals, using a combination of filters to confirm the presence of specific icons like SmartLabel icons.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If multiple barcodes are decoded from an image, then more product information is obtained, but product identification accuracy deteriorates due to conflicts between overlapping codes
Solution Approach 1:
The system decodes multiple barcodes from the same image and uses feedback mechanisms to identify and resolve conflicts between them. By comparing decoded results and detecting inconsistencies, the system can identify the correct barcode and eliminate false readings from overlapping codes, thus maintaining both information completeness and identification accuracy.
Solution Approach 2:
The system introduces an intermediary conflict resolution process between image capture and final product identification. This intermediary layer analyzes multiple decoded barcodes, detects conflicts, and determines the valid code, acting as a mediator that reconciles the contradiction between obtaining multiple codes and ensuring accurate identification.
2Productivity
If conventional barcode decoding is used, then decoding speed is maintained, but reliability deteriorates in the presence of overlapping or multiple barcodes
Solution Approach 1:
The system performs preliminary actions by detecting and evaluating multiple potential barcodes before final identification. It pre-processes the image to identify candidate barcodes, evaluates their validity, and resolves conflicts before making the final product identification, thus maintaining speed while improving reliability through advance analysis.
Solution Approach 2:
The system applies partial action by not decoding every detected barcode equally, but rather selectively processing and validating specific candidates. It performs excessive action by initially detecting more barcodes than needed, then filtering down to the correct one, ensuring reliability without sacrificing overall decoding efficiency.
Data Source
AI summary
The present disclosure relates to signal decoding and icon (e.g., a logo, shape, icon, etc.) detection. In some implementations, a series of filters are applied to scanlines to determine whether an icon is present. Other aspects, combinations and implementations are described as well.


