Aurélien Bourquard

Hessian-Based Norm Regularization for Image Restoration With Biomedical Applications

We present nonquadratic Hessian-based regularization methods that can be effectively used for image restoration problems in a variational framework. Motivated by the great success of the total-variation (TV) functional, we extend it to also include second-order differential operators. Specifically, we derive second-order regularizers that involve matrix norms of the Hessian operator. The definition of these functionals is based on an alternative interpretation of TV that relies on mixed norms of directional derivatives. We show that the resulting regularizers retain some of the most favorable properties of TV, i.e., convexity, homogeneity, rotation, and translation invariance, while dealing effectively with the staircase effect. We further develop an efficient minimization scheme for the corresponding objective functions. The proposed algorithm is of the iteratively reweighted least-square type and results from a majorization-minimization approach. It relies on a problem-specific preconditioned conjugate gradient method, which makes the overall minimization scheme very attractive since it can be applied effectively to large images in a reasonable computational time. We validate the overall proposed regularization framework through deblurring experiments under additive Gaussian noise on standard and biomedical images.

One-Bit Measurements With Adaptive Thresholds

We introduce a new method for adaptive one-bit quantization of linear measurements and propose an algorithm for the recovery of signals based on generalized approximate message passing (GAMP). Our method exploits the prior statistical information on the signal for estimating the minimum-mean-squared error solution from one-bit measurements. Our approach allows the one-bit quantizer to use thresholds on the real line. Given the previous measurements, each new threshold is selected so as to partition the consistent region along its centroid computed by GAMP. We demonstrate that the proposed adaptive-quantization scheme with GAMP reconstruction greatly improves the performance of signal and image recovery from one-bit measurements.

Optical imaging using binary sensors

This paper addresses the problem of reconstructing an image from 1-bit-quantized measurements, considering a simple but nonconventional optical acquisition model. Following a compressed-sensing design, a known pseudo-random phase-shifting mask is introduced at the aperture of the optical system. The associated reconstruction algorithm is tailored to this mask. Our results demonstrate the feasibility of the whole approach for reconstructing grayscale images.

A practical inverse-problem approach to digital holographic reconstruction

In this paper, we propose a new technique for high-quality reconstruction from single digital holographic acquisitions. The unknown complex object field is found as the solution of a nonlinear inverse problem that consists in the minimization of an energy functional. The latter includes total-variation (TV) regularization terms that constrain the spatial amplitude and phase distributions of the reconstructed data. The algorithm that we derive tolerates downsampling, which allows to acquire substantially fewer measurements for reconstruction compared to the state of the art. We demonstrate the effectiveness of our method through several experiments on simulated and real off-axis holograms.

Anisotropic Interpolation of Sparse Generalized Image Samples

Practical image-acquisition systems are often modeled as a continuous-domain prefilter followed by an ideal sampler, where generalized samples are obtained after convolution with the impulse response of the device. In this paper, our goal is to interpolate images from a given subset of such samples. We express our solution in the continuous domain, considering consistent resampling as a data-fidelity constraint. To make the problem well posed and ensure edge-preserving solutions, we develop an efficient anisotropic regularization approach that is based on an improved version of the edge-enhancing anisotropic diffusion equation. Following variational principles, our reconstruction algorithm minimizes successive quadratic cost functionals. To ensure fast convergence, we solve the corresponding sequence of linear problems by using multigrid iterations that are specifically tailored to their sparse structure. We conduct illustrative experiments and discuss the potential of our approach both in terms of algorithmic design and reconstruction quality. In particular, we present results that use as little as 2% of the image samples.

Binary Compressed Imaging

Compressed sensing can substantially reduce the number of samples required for conventional signal acquisition at the expense of an additional reconstruction procedure. It also provides robust reconstruction when using quantized measurements, including in the one-bit setting. In this paper, our goal is to design a framework for binary compressed sensing that is adapted to images. Accordingly, we propose an acquisition and reconstruction approach that complies with the high dimensionality of image data and that provides reconstructions of satisfactory visual quality. Our forward model describes data acquisition and follows physical principles. It entails a series of random convolutions performed optically followed by sampling and binary thresholding. The binary samples that are obtained can be either measured or ignored according to predefined functions. Based on these measurements, we then express our reconstruction problem as the minimization of a compound convex cost that enforces the consistency of the solution with the available binary data under total-variation regularization. Finally, we derive an efficient reconstruction algorithm relying on convex-optimization principles. We conduct several experiments on standard images and demonstrate the practical interest of our approach.

Autocalibrated signal reconstruction from linear measurements using adaptive GAMP

In this paper, we reconstruct signals from underdetermined linear measurements where the componentwise gains of the measurement system are unknown a priori. The reconstruction is performed through an adaptation of the messagepassing algorithm called adaptive GAMP that enables joint gain calibration and signal estimation. To evaluate our approach, we apply it to the problem of sparse recovery and compare it against an ℓ1-based approach. We numerically show that adaptive GAMP yields excellent results even for a moderate amount of data. It approaches the performance of oracle GAMP where the gains are perfectly known asymptotically.

Hessian-based regularization for 3-D microscopy image restoration

We investigate a non quadratic regularizer that is based on the Hessian operator for dealing with the restoration of 3-D images in a variational framework. We show that the regularizer under study is a valid extension of the total-variation (TV) functional, in the sense that it retains its favorable properties while following a similar underlying principle. We argue that the new functional is well suited for the restoration of 3-D biological images since it does not suffer from the well-known staircase effect of TV. Furthermore, we present an efficient 3-D algorithm for the minimization of the corresponding objective function. Finally, we validate the overall proposed regularization framework through image deblurring experiments on simulated and real biological data.

A second-order extension of TV regularization for image deblurring

In this paper, we propose a novel second-order regularizer based on the maximum response of the second-order directional derivative, assuming that the image under consideration belongs to the class of piecewise-linear signals. Compared to total-variation regularization that preserves edges but transforms piecewise-smooth regions into piecewise-constant regions, the proposed model is able to restore piecewise-linear regions and finer details. Deconvolution experiments demonstrate the performance of our approach in terms of the quality of reconstruction.

Genetic Algorithm optimization for a surgical ultrasonic transducer

An ultrasonic piezoelectric transducer to cut the human tissue or to remove the spinal disc is envisioned to improve efficiency and facilitate the surgeons work. Three genetic algorithms have been developed. The first one is based on conditional genetic operators, the second one is based on a specific crossover operator definition with a local search method, and the third is a combination of both the previous algorithms. The third algorithm is used to optimize the piezoelectric transducer. In less than 5000 simulations the optimizations give an amplitude of the cutting tip of about 4.8 microns.