Synthetic Golden Template for Multi-Layer Vision Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine vision systems face challenges in inspecting multi-layered printed or patterned surfaces due to difficulties in resolving misalignment and overlap between layers, which are not easily addressed by standard golden template comparison processes.
Innovation Solution
A system and method that generates a synthetic golden template image using acquired images and binary mask images to register and align each layer independently, accounting for translation, angle, scale, and distortion, and combines these to create a customized golden template for defect detection in multi-layered patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If standard golden template comparison is used for multi-layer inspection, then the inspection process is simple, but it cannot resolve misalignment and overlap between layers
Solution Approach 1:
The patent segments the multi-layer pattern into individual layer masks, where each mask represents a specific print layer. By separating the layers and processing them independently through registration and warping operations, the system can accurately handle misalignment and overlap between layers while maintaining reliable defect detection.
2Ease of manufacture
If a single golden template is used for multi-layer patterns, then the template generation is straightforward, but it cannot account for varying alignment among layers
Solution Approach 1:
The patent implements dynamic layer registration where each layer mask is independently registered and warped according to its specific alignment characteristics. The system calculates registration transforms for each layer based on detected features, allowing the template to adapt to varying alignments dynamically rather than using a fixed single-template approach.
Solution Approach 2:
The patent introduces layer masks as intermediary representations between the raw multi-layer image and the final golden template. These masks serve as mediators that capture the ideal geometry of each layer, which are then registered and combined to create the final template, enabling handling of alignment variations.
3Measurement precision
If layer-specific registration is performed for each printed layer, then alignment accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the complex registration problem into simpler sub-problems by creating separate layer masks for each printed layer. Each mask is registered independently using simplified feature matching, which reduces the overall computational complexity compared to attempting to register all layers simultaneously as a single complex pattern.
Data Source
AI summary
A system and method for generating golden template images in a vision system to inspect an acquired runtime image of an object with a multi-layer printed pattern is provided. The system and method performs a registration process on runtime images using registration models each trained on respective canonical layer mask images, and outputting poses. Based upon the poses, warped layer masks are generated. Combination masks are computed based upon differing combinations of the warped layer masks. Intensity values for pixels of the foreground regions for the combination masks are estimated. The estimated intensity values are then blended associated with the combination masks to generate a golden template image. This golden template image can be used to compare with a runtime image. An exemplary application of this system and method is in print inspection on flat and non-flat surfaces.