Multi-Layer Image Reconstruction via Segmented Wave Partitioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
See-through sensing technologies, such as terahertz sensing, face challenges in reconstructing the three-dimensional structure of target objects due to complexity in image computation and the presence of undesirable artifacts, particularly in deeper layers, which are exacerbated by the shadow effect caused by non-uniform wave penetration.
Innovation Solution
The approach involves treating the target object as a multi-layer structure, partitioning the reflected wave into segments to define each layer, and using joint-layer hierarchical image recovery with hierarchical regularization to prevent sparsity increase and reduce artifacts, employing techniques like joint hierarchical regularization and total variation in the image reconstruction process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the target object is treated as a multi-layer structure and images of each layer are reconstructed individually or jointly, then the computational complexity is reduced and image quality is improved, but the device complexity increases due to the need for segmentation and multi-layer processing
Solution Approach 1:
The patent divides the three-dimensional target object into multiple two-dimensional layers along the propagation direction of the electromagnetic wave. Each layer is reconstructed separately or jointly using sparse optimization techniques. This segmentation approach reduces computational complexity by breaking down the 3D reconstruction problem into multiple simpler 2D problems, while also improving image quality by accounting for the layered structure of the target object.
2Measurement precision
If joint-layer hierarchical image recovery with hierarchical regularization is used to prevent sparsity increase, then image quality and artifact reduction are improved, but the computational complexity increases
Solution Approach 1:
The patent introduces hierarchical regularization parameters that control the sparsity of different layers in a hierarchical manner. The regularization parameters are optimized to prevent sparsity increase across layers, which reduces artifacts and improves image quality. This parameter optimization approach balances the trade-off between image quality and computational complexity by systematically adjusting the regularization strength at different hierarchical levels.
3Productivity
If the reflected wave is partitioned into segments to define multi-layer structure, then the sparsity of each layer can be considered and image reconstruction is simplified, but the measurement precision may be affected by segmentation accuracy
Solution Approach 1:
The patent performs preliminary segmentation of the reflected wave into multiple segments before image reconstruction. Each segment corresponds to a specific layer of the target object. This preliminary action of dividing the signal allows the subsequent reconstruction algorithm to process each layer separately, simplifying the computation while maintaining measurement precision through appropriate segment boundary selection.
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 improves the computational efficiency and image quality of reconstructed images by considering the sparsity of each layer and enforcing constraints to mitigate the shadow effect, resulting in clearer images with reduced artifacts, especially for deeper layers.
Implementation Method 1
an emitter configured to emit a wave in a direction of propagation to penetrate layers of a structure of a target object
Implementation Method 2
a receiver configured to measure intensities of the wave reflected by the layers of the target object
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A scanner includes an emitter configured to emit a wave in a direction of propagation to penetrate layers of a structure of a target object and a receiver configured to measure intensities of the wave reflected by the layers of the target object. The scanner also includes a hardware processor configured to partition the intensities of the reflected wave into a set of segments, such that each segment is the reflection from a corresponding layer of the target object, defining a multi-layered structure of the target object; and reconstruct images of the layers of the target object from corresponding segments using a joint-layer hierarchical image recovery that prevents an increase in sparsity of the layers of the target object in the direction of propagation of the wave. An output interface is configured to render the reconstructed images of layers of the target object.