Noise Robust Decoder for Imaging Sensor Array Multiplexing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional compressed sensing (CS) approaches for detector array multiplexing struggle in noisy environments, particularly in low Signal-to-Noise Ratio (SNR) conditions, limiting their practical application in imaging systems.
Innovation Solution
A three-step process involving preliminary image computation using compressed sensing, support estimation, and final image reconstruction via maximum likelihood estimation with a support constraint, effectively addressing noise challenges and reducing the number of readout channels in imaging systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional compressed sensing approaches are used for detector array multiplexing, then the number of readout channels is reduced, but noise robustness deteriorates
Solution Approach 1:
The patent applies preliminary action by first computing a preliminary image from the raw multiplexed data using compressed sensing algorithms before final reconstruction. This preliminary image serves as an initial estimate that guides subsequent support estimation and refinement steps, enabling the system to handle noisy data more effectively while maintaining channel reduction
Solution Approach 2:
The patent introduces an intermediary support estimation step between the preliminary compressed sensing reconstruction and the final image reconstruction. By estimating the support (locations of non-zero elements) of the preliminary image and using it as a constraint in the final reconstruction, the system mediates between the reduced channel data and the desired high-quality image, improving noise robustness
2Measurement precision
If the number of detector elements is increased, then imaging resolution is improved, but the number of readout channels and system complexity increases
Solution Approach 1:
The patent merges multiple detector element signals into fewer readout channels through multiplexing schemes. By combining the outputs of multiple detector elements into a smaller number of readout channels while using compressed sensing and support estimation to recover the full-resolution image, the system achieves high imaging resolution without proportionally increasing the number of readout channels
Solution Approach 2:
The patent changes the parameter of readout channel count by applying compressed sensing techniques that allow reconstruction of high-resolution images from undersampled data. The support estimation step identifies which spatial locations contain signal information, enabling efficient parameter allocation where readout channels are concentrated on informative regions rather than uniformly distributed
3Ease of manufacture
If compressed sensing is applied to reduce readout channels, then cost is reduced, but performance in noisy conditions deteriorates
Solution Approach 1:
The patent performs preliminary compressed sensing reconstruction to obtain an initial image estimate from the reduced channel data. This preliminary action enables subsequent support estimation that identifies meaningful signal locations, allowing the system to maintain cost-effectiveness while improving noise performance through the two-stage reconstruction approach
Solution Approach 2:
The support estimation acts as an intermediary that bridges the cost-effective compressed sensing reconstruction and the final high-performance image reconstruction. By using the preliminary image to estimate support and then applying this support constraint in the final reconstruction, the system achieves both cost reduction through channel multiplexing and improved noise performance
Data Source
AI summary
Compressed sensing (CS) estimation approaches rely on a priori sparsity to significantly reduce the number of samples needed to provide high sampling fidelity, relative to the normal Shannon-Nyquist limit. Accordingly, CS approaches are of considerable interest for detector multiplexing in applications which have inherently sparse signals (e.g., the two correlated photon detection events in PET imaging). However, CS approaches also tend to fare poorly in the presence of noise, which has limited their applicability in practice. In this work, we show that CS estimation can be used to provide an estimate of the support of an image. This estimated support is then used as a constraint for maximum likelihood image reconstruction. This approach has robust noise performance and provides high reconstruction fidelity.