Virtual Scanline Decoder for Blurred Barcode Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing techniques for machine readable indicia, such as barcodes, fail in cases of excessive noise, high symbol resolution, and optical blur, leading to degraded or unsuccessful decoding.
Innovation Solution
An algorithm that generates virtual scanlines based on known optical system parameters and barcode structures, allowing for comparison with actual scanlines to improve decoding accuracy, even in challenging imaging conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional image processing techniques (edge detection, linear filtering) are used, then processing is effective for well-visible indicia, but decoding fails under excessive noise, high resolution, or optical blur
Solution Approach 1:
The patent pre-calculates and stores virtual scanlines for all possible codewords in a lookup table during system initialization. These virtual scanlines are generated by convolving ideal barcode patterns with the actual optical transfer function of the imaging system. During decoding, the system directly compares captured scanlines against these pre-computed virtual scanlines, eliminating the need for real-time complex processing and enabling reliable decoding under various degraded conditions.
Solution Approach 2:
The patent transforms the decoding approach by changing from direct image processing to parameter-based comparison. Instead of processing images to extract features, the system uses known optical parameters (OTF) and barcode structure parameters to generate virtual scanlines with specific characteristics. The comparison is based on correlation coefficients between virtual and actual scanlines, transforming the problem into a parameter-matching task that is more robust to noise and blur.
2Reliability
If improved optical systems are employed to read indicia under challenging conditions, then reading ability improves, but system cost increases and physical limitations are reached
Solution Approach 1:
The patent replaces physical/optical improvements with a computational approach. Instead of using better optics to capture clearer images, the system uses virtual scanline generation and correlation-based comparison to achieve reliable decoding. The optical system remains simple, but the processing methodology substitutes for optical complexity by compensating for optical deficiencies through algorithmic means.
Solution Approach 2:
The patent creates virtual copies of what the scanline should look like under ideal conditions, adjusted by the actual optical transfer function. These virtual scanlines serve as reference templates that represent expected patterns. By comparing actual captured scanlines against these virtual copies, the system can identify codewords even when the actual scanline is degraded by noise, blur, or high resolution, without needing improved optics.
3Measurement precision
If virtual scanlines are generated and compared with actual scanlines, then decoding accuracy improves under noise and blur, but processing complexity increases
Solution Approach 1:
The patent pre-calculates and stores virtual scanlines for all possible codewords in a lookup table during system initialization. This shifts the computational burden from real-time processing to offline preparation. During actual decoding, the system only needs to perform simple correlation comparisons between captured scanlines and pre-computed virtual scanlines, significantly reducing real-time processing complexity while maintaining high decoding accuracy.
Solution Approach 2:
The patent generates virtual scanlines for all possible codewords (excessive action) rather than trying to directly process the captured image to extract codewords. This exhaustive approach ensures that the correct codeword is always in the comparison set, and the correlation method provides a robust partial match even when the captured scanline is partially degraded. The lookup table approach trades memory usage for reduced processing complexity.
Data Source
AI summary
A system and method for processing imaged machine readable indicia may include capturing an image of a machine readable indicia including multiple codewords. A determination of an actual scanline of a codeword of the captured indicia may be made. A virtual scanline of the codeword of the imaged indicia for a type of symbology may be generated. In response to the virtual and actual scanlines matching, data representative of the codeword may be stored. Otherwise, repeat generating virtual scanlines of the codewords and compare the generated virtual scanlines with the actual scanlines until a match is identified or possible virtual codewords are exhausted. In response to a match, data representative of the matched scanline may be stored. The process may further repeatedly determine, generate, and compare until each codeword is determined. Data represented by the indicia based on the stored data representative of the matched codewords may be generated.