Photomask Alignment Using Edge Count Accumulation for Complex Patterns
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional alignment methods struggle with high accuracy alignment of photomasks for patterns with complex shapes, such as wiring patterns on TFT substrates, due to difficulties in detecting reference positions on the edge portions of these patterns.
Innovation Solution
An alignment method that processes multiple images captured at regular intervals by an imaging device with light-receiving elements to detect brightness changes, accumulates edge count data, identifies positions of long sides, selects a position close to a target, and adjusts the photomask position to align with the object, enhancing alignment accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional alignment methods detect brightness changes at edge portions of patterns, then alignment is easy for simple rectangular shapes, but alignment accuracy deteriorates for complex shapes like wiring patterns
Solution Approach 1:
The patent segments the pattern recognition process into multiple components: edge detection through brightness change detection, edge count calculation for each position, and threshold-based selection of reliable edges. This segmentation allows the system to process complex patterns by breaking them down into detectable edge events along the conveying direction, thereby maintaining alignment accuracy for both simple and complex shapes.
Solution Approach 2:
The patent introduces a new dimension of analysis by calculating edge counts in the conveying direction (along the long side of the pattern) rather than only detecting edges perpendicular to the conveying direction. This dimensional shift allows reliable reference position detection for complex patterns by accumulating edge information along the length of the pattern, transforming a 2D edge detection problem into a 1D cumulative analysis along the conveying direction.
2Measurement precision
If reference position is detected on edge portion of complex pattern, then alignment can be performed, but detection accuracy deteriorates due to pattern complexity
Solution Approach 1:
The patent performs preliminary actions by detecting all brightness changes along the entire pattern before selecting the reference position. The system accumulates edge count data for all positions along the conveying direction, then applies threshold criteria to identify reliable reference edges. This preliminary comprehensive detection ensures that the selected reference position is based on complete information, improving detection accuracy for complex patterns.
Solution Approach 2:
The patent implements feedback through the edge count threshold mechanism. The system calculates edge counts for each position and compares them against a predetermined threshold, selecting reference positions only where the edge count exceeds the threshold. This feedback loop ensures that only reliable, clearly defined edges are selected as reference positions, filtering out ambiguous edges in complex patterns and thereby improving reference position detection accuracy.
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
This method allows for accurate alignment of photomasks with complex patterned objects, improving the precision of opening alignment with the object's pattern, even for complex shapes like TFT substrates with wiring patterns.
Implementation Method 1
an imaging device having a plurality of light-receiving elements arranged in a straight line in a direction perpendicular to a conveying direction of the object to be exposed captures an image of the object to be exposed
Data Source
AI summary
In the present invention, a number of times the brightness changes detected at the same position while a substrate conveys are added up in the conveying direction, thereby obtaining a plurality of edge count data, and then, a plurality of positions of long sides of patterns parallel to the conveying direction is identified based on the plurality of edge count data exceeding a predetermined threshold value, middle point positions of a plurality of proximity pairs are calculated, and a middle point position close to the target position preset in the imaging device is selected from the plurality of middle point positions of the proximity pairs, an amount of position displacement between the selected middle point position and the target position of imaging device is calculated, and the photomask in the direction substantially perpendicular to the conveying direction so that the amount of position displacement is a predetermined value.