THz Feature Localization via Wavelength-Based Signal Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current THz imaging techniques face challenges in feature localization due to ambiguity in pixel size definition caused by broadband pulses, leading to decreased detection and localization of features based on their size and location, especially when features are similar in size to or smaller than the focal spot sizes, and misalignment with the pixel array.
Innovation Solution
A method that utilizes a transform to convert broadband returns into wavelength-based returns, grouping them into distinct focal diameter categories, calculating intra- and inter-return probabilities, and establishing a refined pixel grid for enhanced feature localization by aligning the scanned features with wavelength groups and overlap regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If broadband pulses are used for imaging, then the imaging capability is improved, but the pixel size definition becomes ambiguous leading to decreased feature localization precision
Solution Approach 1:
The broadband pulse signal is segmented into multiple wavelength components, each associated with a specific focal spot size. By processing each wavelength component separately and then combining the results, the patent resolves the ambiguity of pixel size definition while maintaining the benefits of broadband imaging.
Solution Approach 2:
The patent dynamically adjusts the effective pixel size based on the wavelength content of the signal. Different wavelength ranges are assigned to different pixel sizes according to their focal spot characteristics, creating a dynamic rather than static pixel definition that adapts to the imaging requirements.
2Area of stationary object
If pixel size is defined to match the largest focal spot size, then the coverage area is improved, but the image resolution decreases
Solution Approach 1:
The imaging process is segmented into multiple wavelength-based processing streams, each with its own optimized pixel size. Shorter wavelengths use smaller pixels for high resolution, while longer wavelengths use larger pixels for broader coverage, and the results are combined to achieve both goals simultaneously.
3Manufacturing precision
If pixel size is defined to match the smallest focal spot size, then the image resolution is improved, but the measured area coverage decreases
Solution Approach 1:
Multiple wavelength components with different focal spot sizes are merged into a single composite image. Each wavelength contributes to different spatial frequencies, and their combination provides both high resolution (from short wavelengths) and broad coverage (from long wavelengths) in the final image.
4Difficulty of detecting and measuring
If features are similar in size to or smaller than focal spot sizes, then the detection capability is challenged, but the localization accuracy decreases
Solution Approach 1:
The patent introduces a wavelength dimension to the imaging process, transforming the problem from a single spatial dimension to a combined wavelength-spatial domain. By analyzing features across multiple wavelength components, small features can be detected and localized with higher accuracy than would be possible with a single focal spot size.
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 approach dynamically defines pixel size and location, improving image resolution and feature localization by separating the signal from each pulse into component spot sizes, allowing for precise spatial analysis and accurate representation of feature positions.
Implementation Method 1
When light passes through an aperture, or lens, diffraction occurs. Equation (1) describes the intensity of light in the focal plane... The pattern resulting from this diffraction is referred to as the Airy pattern.
Implementation Method 2
The position of the pulses defines the location of each pixel... The size of each pixel is defined by the focal spot size of the pulse... Equation (3), which describes this diameter [focal spot diameter]
Data Source
AI summary
A method for increasing localization utilizing overlapped broadband pulses includes using a transform to convert broadband returns into wavelength based returns. The wavelength based returns are grouped into at least two wavelength group returns for each location having different focal diameters. Intra-return probabilities of object location are computed from the group returns. Inter-return probabilities are computed for overlapping regions of the pulse returns. A pixel grid is established for displaying the calculated object location probabilities. By further processing, the pixel grid can be refined to show finer details.