Image Inspection Equipment Position Displacement Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques for inspecting semiconductor circuits face challenges in maintaining accuracy due to position displacements between design data and captured images, which can lead to decreased inspection sensitivity and increased variations in probability distributions.
Innovation Solution
An image inspection equipment and method that includes a training processing unit to estimate and reflect position displacements in the probability distribution model, ensuring accurate alignment and reducing variations caused by image distortions and non-linear displacements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pre-alignment is performed on design data and captured image, then training accuracy of probability distribution is improved, but position displacement due to image distortion and non-linear displacement cannot be corrected
Solution Approach 1:
The patent applies preliminary action by performing pre-alignment before training to establish an initial accurate correspondence between design data and captured image. This preliminary alignment creates a foundation for subsequent training while the system later compensates for residual distortions through the probability distribution modeling approach.
Solution Approach 2:
The patent changes parameters by modeling position displacement as a variation in the probability distribution parameters (mean and standard deviation) rather than attempting to correct the displacement geometrically. This allows the system to accommodate non-linear and local position displacements by adjusting statistical parameters of the probability distribution.
2Adaptability or versatility
If captured image with image distortion is used for training, then manufacturing margin is modeled as variation in probability distribution, but position displacement is modeled as manufacturing margin causing increased variation
Solution Approach 1:
The patent applies local quality by allowing different regions of the probability distribution to have different mean and standard deviation values. This enables the model to capture local variations in pixel values that correspond to actual manufacturing variations rather than global position displacement effects.
Solution Approach 2:
The patent introduces an intermediary approach by using the probability distribution as a mediator between the captured image and the design data. Instead of directly comparing pixel values, the system models the relationship through probability distributions that capture manufacturing variations while being robust to position displacements.
3Ease of operation
If position displacement is modeled as manufacturing margin, then inspection can proceed with displaced images, but variation in probability distribution increases reducing inspection sensitivity
Solution Approach 1:
The patent applies dynamics by making the probability distribution parameters (mean and standard deviation) variable rather than fixed. This allows the model to dynamically adapt to different position displacements and manufacturing variations, maintaining inspection sensitivity across a range of conditions.
Solution Approach 2:
The patent performs preliminary alignment before training to establish an initial accurate correspondence between design data and captured image. This preliminary alignment minimizes position displacement effects, allowing the system to focus on modeling manufacturing variations without being overwhelmed by geometric misalignments.
Data Source
AI summary
Provided is an image inspection device which can prevent a degradation in the accuracy of an estimation value of a probability distribution caused by position displacement between design data and a captured image in image processing in which a model that estimates a pixel value probability distribution of a captured image is trained by using design data and the captured image of a sample. The image inspection device inspects a captured image by using design data and the capture image of a sample, the device being characterized by comprising: a training processing unit which trains a probability distribution estimation model that estimates, from the design data, a pixel value probability distribution of the captured image; and an inspection processing unit which inspects a captured image for inspection by using the probability distribution estimation model created in the training processing unit, design data for inspection, and the captured image for inspection, wherein the training processing unit includes: a probability distribution estimation unit which estimates, from sample design data for training, a pixel value probability distribution of a sample captured image for training; a position displacement amount estimation unit which estimates a position displacement amount between a probability distribution during training estimated by the probability distribution estimation unit, and the captured image for training; a position displacement reflection unit which reflects, to the probability distribution during training, the estimation position displacement amount estimated by the position displacement amount estimation unit; and a model update unit which evaluates the probability distribution estimation model of the probability distribution estimation unit by using the captured image for training and the probability distribution during training to which the position displacement calculated by the position displacement reflection unit has been reflected, and updates parameters of the probability distribution estimation model according to the evaluation values.