Low Coherence Interferometry Signal Classification for Surface Topography
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for measuring the surface topography of objects with multiple structures using low coherence interferometry are inefficient due to incorrect phase shifting and thickness calculations, requiring complex and costly modifications to account for different materials and structures, and lack accurate identification of structures within the field of view.
Innovation Solution
A method and system that attribute interferometric signals to classes associated with typical structures, allowing for classification and analysis without prior knowledge of structure topology, using existing measurement devices with minimal modifications, and enabling corrected profile and thickness measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If theoretical reflectance models are constructed to correct measurements for different materials and structures, then measurement accuracy is improved, but device complexity and cost increase due to requiring multiple coloured lights or sensors
Solution Approach 1:
The invention segments the measurement process by classifying interferometric signals into different groups based on their characteristics. Each group corresponds to a specific structure type (e.g., transparent layer, metallic layer, absorptive layer). This segmentation allows the system to apply structure-specific analysis methods to each group, improving measurement accuracy without requiring complex multi-wavelength equipment. The classification step divides the field of view into regions with different structure types, enabling targeted correction approaches for each region.
Solution Approach 2:
The invention changes the approach from using multiple physical parameters (multiple wavelengths, colours) to using signal processing parameters. By analyzing the interferometric signal characteristics (phase, amplitude, envelope) and classifying structures based on these parameters, the system achieves material differentiation without adding complex optical components. The parameter change transforms the problem from optical physics to signal analysis, simplifying the device while maintaining measurement precision.
2Measurement precision
If multiple coloured lights or sensors are added to identify different structures, then structure identification accuracy is improved, but measuring time and equipment cost increase
Solution Approach 1:
The invention merges the structure identification function with the existing interferometric measurement process. Instead of adding separate identification steps or equipment, the system uses the interferometric signals already being captured to classify structures. The classification is performed on the same interferometric data used for topography measurement, combining two functions (measurement and identification) into a single integrated process, thereby avoiding additional measuring time and equipment cost.
Solution Approach 2:
The invention creates a virtual copy of the physical measurement process in the signal domain. By analyzing and classifying the interferometric signals computationally, the system replicates the structure identification capability that would otherwise require physical multi-wavelength sensors. This virtual copying approach achieves the same identification accuracy without the time and cost penalties of physical modifications.
3Device complexity
If three fixed wavelengths are used for spectral reflectance analysis, then equipment complexity is reduced, but measurement versatility deteriorates due to limitation in discriminating different material stacks
Solution Approach 1:
The invention introduces dynamic adaptability to the measurement system by enabling selection of different wavelength combinations based on the specific measurement requirements and material types. Rather than being fixed at three wavelengths, the system can dynamically adjust which wavelengths are used for analysis. This dynamic capability allows the simple three-wavelength equipment to adapt to different material stacks and measurement scenarios, maintaining versatility without increasing inherent device complexity.
Solution Approach 2:
The invention changes the approach from relying on fixed physical parameters (three specific wavelengths) to using variable analysis parameters (different wavelength combinations, different classification criteria). By changing the software-based analysis parameters rather than the hardware configuration, the system achieves enhanced material discrimination capability while keeping the equipment simple. The parameter changes occur in the signal processing domain, not the optical domain.
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 measurement of surface profiles and thicknesses of objects with multiple structures using low coherence optical interferometry, improving measurement accuracy and reducing the need for complex equipment modifications.
Implementation Method 1
By varying the optical paths of the reference or inspection rays, an interference signal (interferogram) is obtained. This is an intensity signal that is a function of the difference between the paths travelled by the rays.
Implementation Method 2
The wave reflected on, or emergent from, the surface of the object to be measured undergoes a phase shift which depends on the physical characteristics of the materials present which are passed through by the wave and/or on which it reflects.
Data Source
AI summary
A method and related system for measuring a surface of a substrate including at least one structure using low coherence optical interferometry, the method being implemented with a system having an interferometric device, a light source, an imaging sensor, and a processing module, the method including: - acquiring, with the imaging sensor, an interferometric signal formed by the interferometric device between a reference beam and a measurement beam reflected by the surface at a plurality of measurement points in a field of view; the following steps being carried out by the processing module: classifying, by a learning technique, the acquired interferometric signals according to a plurality of classes, each class being associated with a reference interferometric signal representative of a typical structure; and analysing the interferometric signals to derive information on the structure at the measurement points, as a function of the class of each interferometric signal.