Single-Camera 3D Reconstruction Using Multi-Angle Illumination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semiconductor inspection methods struggle to accurately and efficiently detect 3D defects due to the increased complexity and cost of current techniques, which often require multiple camera positions or expensive equipment, leading to delayed production and reduced accuracy.
Innovation Solution
A system utilizing multiple light sources at different angles and an AI model to generate a 3D height map of a workpiece, allowing for efficient detection of 3D defects like BGA ball heights without the need for extensive high-accuracy data collection, by training and refining the AI model with a combination of low and high-accuracy data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If expensive dedicated measurement equipment is used for 3D defect detection, then measurement precision is improved, but device complexity and production time increase
Solution Approach 1:
The patent uses AI models to create a virtual copy of the complex 3D measurement process. Instead of using expensive dedicated measurement equipment, the system captures 2D images from multiple angles and uses AI to reconstruct 3D information, effectively copying the functionality of complex measurement equipment through software intelligence rather than hardware complexity
Solution Approach 2:
The patent replaces mechanical 3D measurement systems with an AI-based computational approach. Rather than using physical measurement equipment with complex mechanical structures, the system uses image processing and AI algorithms to derive 3D defect information from 2D images, substituting mechanical measurement with intelligent computation
2Measurement precision
If multiple camera positions are used for 3D defect detection, then measurement precision is improved, but productivity decreases due to longer inspection time
Solution Approach 1:
The patent performs preliminary actions by capturing multiple 2D images from different angles during the normal production flow, then uses AI models to process these pre-captured images and reconstruct 3D defect information. This eliminates the need for slow, time-consuming 3D measurement equipment while maintaining accuracy
Solution Approach 2:
The system copies the 3D measurement function through AI processing of 2D images. Instead of physically moving multiple cameras or using slow 3D scanning, the AI model computationally reconstructs 3D information from standard 2D images, achieving both speed and accuracy
3Device complexity
If depth from focus algorithms are used, then hardware cost is reduced, but measurement precision deteriorates due to accuracy reduction in various lighting and geometry scenarios
Solution Approach 1:
The patent employs AI models that have been trained with feedback from multiple 2D images captured at different illumination angles. The AI system processes these images with different lighting conditions and geometry scenarios, providing feedback-adjusted 3D reconstructions that maintain high accuracy across varying conditions without requiring expensive hardware
Solution Approach 2:
The system changes the approach by using multiple 2D images with different illumination parameters rather than relying on a single 3D measurement method. By varying lighting angles and processing these through AI models, the system achieves accurate 3D height measurements without the hardware cost penalties of dedicated measurement equipment
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate and cost-effective in-line inspection of 3D defects, reducing production delays and hardware costs while maintaining high detection accuracy.
Implementation Method 1
a plurality of light sources each provided at a different illumination angle and configured to generate a beam of light... a detector configured to capture a plurality of images of the workpiece based on the beam of light reflected from the workpiece from each illumination angle
Data Source
AI summary
The system includes a plurality of light sources each provided at a different illumination angle and configured to generate a beam of light, a stage configured to hold a workpiece in the path of the beam of light from each illumination angle, a detector configured to capture a plurality of images of the workpiece based on the beam of light reflected from the workpiece at each illumination angle, a processor in electronic communication with the detector, and an electronic data storage unit in electronic communication with the processor and storing an AI model. The processor is configured to generate a 3D height map of the workpiece based on the plurality of images of the workpiece received from the detector using the AI model. The 3D height map includes a peak height of a 3D surface feature defined on a surface of the workpiece relative to the surface of the workpiece.