[Paper Review] Compressive Diffraction Tomography for Weakly Scattering

LianLin Li, Wenji Zhang+1 more|ArXiv.org|Apr 17, 2009

Sparse and Compressive Sensing Techniques11 references3 citations

TL;DR

This paper proposes a compressive diffraction tomography framework for weakly scattering objects using random media to enable super-resolution reconstruction from highly sparse data. By leveraging wavefield incoherence and a modified sparse Bayesian algorithm, it achieves exact K-sparse recovery with measurements comparable to standard compressive sensing, significantly reducing data requirements while maintaining resolution trade-offs.

ABSTRACT

An appealing requirement from the well-known diffraction tomography (DT) exists for success reconstruction from few-view and limited-angle data. Inspired by the well-known compressive sensing (CS), the accurate super-resolution reconstruction from highly sparse data for the weakly scatters has been investigated in this paper. To realize the compressive data measurement, in particular, to obtain the super-resolution reconstruction with highly sparse data, the compressive system which is realized by surrounding the probed obstacles by the random media has been proposed and empirically studied. Several interesting conclusions have been drawn: (a) if the desired resolution is within the range from to, the K-sparse N-unknowns imaging can be obtained exactly bymeasurements, which is comparable to the required number of measurement by the Gaussian random matrix in the literatures of compressive sensing. (b) With incorporating the random media which is used to enforce the multi-path effect of wave propagation, the resulting measurement matrix is incoherence with wavelet matrix, in other words, when the probed obstacles are sparse with the framework of wavelet, the required number of measurements for successful reconstruction is similar as above. (c) If the expected resolution is lower than, the required number of measurements of proposed compressive system is almost identical to the case of free space. (d) There is also a requirement to make the tradeoff between the imaging resolutions and the number of measurements. In addition, by the introduction of complex Gaussian variable the kind of fast sparse Bayesian algorithm has been slightly modified to deal with the complex-valued optimization with sparse constraints.

Motivation & Objective

• To address the challenge of reconstructing high-resolution images from few-view and limited-angle diffraction tomography data.
• To enable super-resolution imaging using highly sparse measurements in diffraction tomography.
• To reduce the number of required measurements while maintaining image quality through compressive sensing principles.
• To investigate the role of random media in enhancing measurement incoherence and improving reconstruction performance.
• To adapt sparse Bayesian algorithms for complex-valued, wavelet-sparse imaging problems.

Proposed method

• The compressive system uses random media surrounding the object to induce multi-path wave propagation, enhancing incoherence with the wavelet sparsity basis.
• The resulting measurement matrix is incoherent with the wavelet transform, satisfying a key condition for successful compressive sensing recovery.
• A modified sparse Bayesian algorithm is developed to handle complex-valued optimization under sparsity constraints.
• The method enables exact K-sparse reconstruction of N unknowns using M ≈ K measurements, matching theoretical bounds from compressive sensing.
• The framework is validated through empirical studies on weakly scattering objects under varying resolution and measurement conditions.

Experimental results

Research questions

• RQ1 Can super-resolution reconstruction be achieved from highly sparse data in diffraction tomography using compressive sensing principles?
• RQ2 How does the use of random media affect measurement incoherence and reconstruction accuracy?
• RQ3 What is the minimal number of measurements required for successful K-sparse recovery in this compressive system?
• RQ4 How does the imaging resolution trade-off with the number of measurements in the proposed compressive setup?
• RQ5 To what extent does the modified sparse Bayesian algorithm improve complex-valued sparse reconstruction in this context?

Key findings

• K-sparse N-unknowns imaging can be exactly recovered using M ≈ K measurements, matching the theoretical minimum from compressive sensing.
• The random media-induced multi-path effect ensures incoherence with the wavelet basis, enabling stable and accurate reconstruction.
• When the desired resolution is below a certain threshold, the required measurements are nearly identical to those in free-space configurations.
• A trade-off exists between imaging resolution and the number of measurements, with higher resolution demanding more data.
• The modified sparse Bayesian algorithm effectively handles complex-valued, wavelet-sparse optimization, enabling robust reconstruction.