Inspection Method for PCB Distortion Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional shape measurement apparatuses struggle to accurately set inspection areas due to distortion in measurement targets like printed circuit boards, leading to incorrect location settings and measurement errors.
Innovation Solution
An inspection method that compensates for distortion by using reference and measurement data to select feature objects, extract feature variables, and apply a quantified conversion formula to set a corrected inspection area, considering changes in slope, size, and transformation degrees.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a predetermined inspection area is set based on theoretical location of circuit elements, then the inspection process is simple and fast, but the inspection area is not accurately positioned due to board distortion such as warp and twist
Solution Approach 1:
The patent applies preliminary action by pre-defining feature blocks with specific shape patterns at known locations on the board before inspection. These feature blocks serve as reference markers that are captured in images, allowing the system to pre-calculate transformation relationships and distortion compensation parameters before actual circuit element inspection begins. This preliminary setup enables accurate inspection area positioning without adding complex real-time processing during inspection.
Solution Approach 2:
The patent introduces feature blocks as intermediary reference objects between the camera system and the actual circuit elements. These feature blocks with distinct shape patterns act as mediators that bridge the theoretical design coordinates and the actual distorted board positions. By capturing and analyzing these intermediary feature blocks, the system can calculate transformation matrices that compensate for board distortion, thereby accurately positioning the inspection area without directly measuring the distorted circuit elements themselves.
2Measurement precision
If the inspection area is set without considering board distortion, then the measurement process is quick and straightforward, but measurement errors occur due to location mismatch between captured images and actual circuit element positions
Solution Approach 1:
The system performs preliminary capture of feature blocks and pre-calculates transformation relationships between design coordinates and actual positions. This preliminary action includes defining feature blocks with specific shape patterns, capturing their positions in the image, and computing distortion compensation parameters before actual inspection begins. This upfront preparation reduces measurement errors while minimizing time loss during the actual inspection process.
Solution Approach 2:
The patent creates a transformed copy of the design coordinates system that accounts for board distortion. By capturing feature blocks and calculating transformation matrices, the system generates a corrected coordinate mapping that copies the theoretical design positions to their actual distorted positions on the board. This copied transformed coordinate system is then used to accurately locate circuit elements without requiring repeated measurements.
3Measurement precision
If feature blocks with shape patterns are used to compensate for distortion, then accurate inspection area setting is achieved, but the process becomes more complex when similar patterns are adjacent on the board
Solution Approach 1:
The patent applies segmentation by dividing the board into distinct feature blocks with unique shape patterns. Each feature block is treated as an independent identification target with its own characteristic pattern. This segmentation allows the system to individually recognize and locate each feature block even when similar patterns exist adjacent to each other, by analyzing the specific shape characteristics and spatial relationships of each segmented feature block separately.
Solution Approach 2:
The patent employs asymmetry by designing feature blocks with non-symmetric or distinctive shape patterns that differ from adjacent feature blocks. By introducing asymmetric characteristics in the shape patterns (such as different orientations, unique geometric configurations, or varied pattern arrangements), the system can reliably distinguish between adjacent feature blocks with similar patterns. This asymmetric design prevents confusion during automatic identification and enables accurate location determination even in densely packed areas.
Data Source
AI summary
In order to set an inspection area, a measurement target is disposed onto a stage, a reference data of the measurement target is summoned, and a measurement data of the measurement target is acquired. Then, at least one feature object is selected in the measurement data and the reference data of the measurement target, and at least one feature variable for the selected feature object is extracted from each of the reference data and the measurement data. Thereafter, a change amount of the measurement target is produced by using the feature variable and a quantified conversion formula, and the produced change amount is compensated for to set an inspection area. Thus, the distortion of the measurement target is compensated for to correctly set an inspection area.


