Barcode and QR Symbology Glare Mitigation via Multi-Pose Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Glare in captured images of marks, such as barcodes and QR codes, degrades the accuracy of processes that rely on these images, particularly in detecting photocopies or counterfeit marks, due to noise and specular reflections.
Innovation Solution
The method employs alignment-fusion techniques and neural networks to generate a single image with reduced glare from multiple poses of the mark, using processes like one-step and two-step alignment-fusion, as well as deep learning models like GANs and neural style transfers, to produce a glare-free image for accurate processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple images of the mark are captured to improve processing accuracy, then the reliability of mark detection is improved, but glare and noise in the images worsen the measurement precision
Solution Approach 1:
The patent combines multiple captured images of the mark into a single composite image through alignment and fusion processes. This merging approach integrates information from multiple poses while eliminating glare and noise, thereby simultaneously improving both the reliability of detection and the precision of measurement by leveraging redundant information across images.
Solution Approach 2:
The patent introduces an intermediary processing system that includes alignment modules, fusion modules, and glare mitigation algorithms. This intermediary layer processes the raw captured images to produce a cleaned, aligned composite image, mediating between the problematic raw images and the final mark detection process to preserve both reliability and precision.
2Manufacturing precision
If alignment-fusion techniques are applied to reduce glare, then the purity of the image data is improved, but the device complexity increases
Solution Approach 1:
The patent segments the image processing task into distinct modular components: alignment modules that handle pose correction, fusion modules that combine images, and glare mitigation algorithms that remove artifacts. This segmentation allows each component to be optimized independently while working together to achieve high image data purity without requiring a monolithic complex system.
Solution Approach 2:
The patent uses multiple captured images as copies of the same mark from different poses and conditions. By processing these copies through alignment and fusion, the system extracts a clean representative image without needing complex single-image processing, leveraging redundancy to simplify the overall approach while maintaining high purity.
3Adaptability or versatility
If multiple poses of the mark are captured to enhance detection accuracy, then the adaptability of the system is improved, but the loss of time for image capture and processing increases
Solution Approach 1:
The patent performs preliminary alignment of multiple captured images to a reference pose before fusion. This preliminary action organizes the data structure early in the process, enabling efficient subsequent processing. By pre-aligning images, the system captures adaptability to different orientations while minimizing the time required for later processing steps.
Solution Approach 2:
The patent dynamically selects and processes only the necessary subset of captured images based on their quality and pose characteristics. Rather than processing all captured images uniformly, the system adapts its processing pipeline to handle varying numbers and types of input images, reducing unnecessary computational overhead while maintaining comprehensive adaptability to different mark orientations.
Data Source
AI summary
Methods, systems, and apparatus, including medium-encoded computer program products, for glare mitigation techniques include: obtaining images containing a representation of a mark, the images comprising multiple poses of the mark and generating a single image from the images that contain a representation of the mark with reduced glare when compared to the images comprising multiple poses of the mark. The single image is provided for processing of the representation of the mark to identify information associated with the mark.