Deconvolution Process Research Articles

An Efficient Method on ISAR Image Reconstruction via Norm Regularization

Inverse synthetic aperture radar (ISAR) imaging technology is a powerful tool to distinguish between targets with different types. The high-resolution radar image acquisition is an important basis for further automatic target recognition. In the high-frequency radar band, since the spatially collected received signals have strong sparsity in the image domain (Fourier domain), they can be downsampled and restored by corresponding norm regularization framework. When the target is electrically large, its contributions are equal to several point scatterers [also called scattering centers (SCs)]. So that the imaging process can be regarded as the summation of the point scatterers impulse response, the convolution process, and deconvolution method, the reverse process is adopted to extract the SCs from the image, where the number of SCs reflects the sparseness of ISAR image. As for the reconstruction method, while L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> regularization is widely used, the L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">q</sub> (0 <; q <; 1) regularization has proved to be a more sparse regularization framework and achieves better performance in reconstructing a signal with a lower down-sampling rate. This article implements L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> regularization into the ISAR image reconstruction 2 under down-sampling. The ISAR image is generated by patch integration formula from the physical optics method. The sparsity of the ISAR image is estimated through the appropriate deconvolution process CLEAN. Given the sparsity estimated, and compared with the L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> regularization, the experiment of two complex geometry targets validates the reconstruction precision and convergence and lossless recovery probability of L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1/2</sub> regularization.

Color lensless in-line holographic microscope with sunlight illumination for weakly-scattered amplitude objects

We propose a method for acquiring color images with a lensless in-line holographic microscope (LIHM) using sunlight illumination, which is suitable for weakly scattered amplitude objects. In the color LIHM, the sample is illuminated by sunlight, and the in-line hologram is recorded by a color CMOS imaging sensor located behind the sample. For weakly-scattered amplitude object, we show that the hologram can be described as the convolution of object transmission and a point spread function (PSF) that depends on the spectral distribution of the light. The captured color hologram is first separated into the red, green, and blue components, and then the sub-holograms of each color are used to reconstruct the corresponding color components of the sample by a deconvolution process. We proved that the deconvolution process was able to improve the imaging resolution, which was deteriorated because of insufficient temporal coherence of the light. The resolution enhancement capability of our color LIHM was demonstrated by numerical simulations and imaging experiments with the U. S. Air Force target as the sample. We also imaged a stained root of the herb Saposhnikovia divaricata to further demonstrate the capability of our method for color imaging applications. Our proposed color LIHM method provides a way to realize color holographic imaging with white light sources and thus reduces the cost and complexity of the lensless color microscope.

RNA interference screen identifies NAA10 as a regulator of PXR transcription

The pregnane X receptor (PXR) is a principal xenobiotic receptor crucial in the detection, detoxification, and clearance of toxic substances from the body. PXR plays a vital role in the metabolism and disposition of drugs, and elevated PXR levels contribute to cancer drug resistance. Therefore, to modulate PXR activity and mitigate drug resistance, it is imperative to fully understand its regulation. To this end, we screened a transcription factor siRNA library in pancreatic cancer cells that express high levels of PXR. Through a comprehensive deconvolution process, we identified N-alpha-acetyltransferase 10 (NAA10) as a factor in the transcriptional machinery regulating PXR transcription. Because no one single factor has 100% operational control of PXR transcriptional regulation, our results together with other previous findings suggest that the transcriptional regulation of PXR is complex and that multiple factors contribute to the process including NAA10.

Real-time, multiframe, blind deconvolution of solar images

The quality of images of the Sun obtained from the ground are severely limited by the perturbing effect of the Earth’s turbulent atmosphere. The post-facto correction of the images to compensate for the presence of the atmosphere require the combination of high-order adaptive optics techniques, fast measurements to freeze the turbulent atmosphere, and very time-consuming blind deconvolution algorithms. Under mild seeing conditions, blind deconvolution algorithms can produce images of astonishing quality. They can be very competitive with those obtained from space, with the huge advantage of the flexibility of the instrumentation thanks to the direct access to the telescope. In this contribution we make use of deep learning techniques to significantly accelerate the blind deconvolution process and produce corrected images at a peak rate of ∼100 images per second. We present two different architectures that produce excellent image corrections with noise suppression while maintaining the photometric properties of the images. As a consequence, polarimetric signals can be obtained with standard polarimetric modulation without any significant artifact. With the expected improvements in computer hardware and algorithms, we anticipate that on-site real-time correction of solar images will be possible in the near future.

Characterization of the Cerenkov scatter function: a convolution kernel for Cerenkov light dosimetry.

Cerenkov light is created in clinical applications involving high-energy radiation such as in radiation therapy. There is considerable interest in using Cerenkov light as a means to perform in vivo dosimetry during radiation therapy; however, a better understanding of the light-to-dose relationship is needed. One such method to solve this relationship is that of a deconvolution formulation, which relies on the Cerenkov scatter function (CSF). The CSF describes the creation of Cerenkov photons by a pencil beam of high-energy radiation, and the subsequent scattering that occurs before emission from the irradiated medium surface. This study investigated the dependence of the CSF on common radiation beam parameters (beam energy and incident angle) and the type of irradiated medium. An analytical equation with fitting coefficients of the CSF was obtained for common beam energies in a stratified skin model and optical phantom. Perturbation analysis was performed to investigate the dependence of the deconvolved Cerenkov images on the full-width at half-maximum and amplitude of the CSF. The irradiated material and beam angle had a large impact on the deconvolution process, whereas the beam energy had little effect.

Marchenko imaging by unidimensional deconvolution

ABSTRACTObtaining an accurate image of the subsurface still remains a great challenge for the seismic method. Migration algorithms aim mainly on positioning seismic events in complex geological contexts. Multiple reflections are typically not accounted for in this process, which can lead to the emergence of artefacts. In Marchenko imaging, we retrieve the complete up‐ and downgoing wavefields in the subsurface to construct an image without such artefacts. The quality of this image depends on the type of imaging condition that is applied. In this paper, we propose an imaging condition that is based on stabilized unidimensional deconvolution. This condition is computationally much cheaper than multidimensional deconvolution, which has been proposed for Marchenko imaging earlier. Two specific approaches are considered. In the first approach, we use the full up‐ and downgoing wavefields for deconvolution. Although this leads to balanced and relatively accurate amplitudes, the crosstalk is not completely removed. The second approach is to incorporate the initial focussing function in the deconvolution process, in such a way that the retrieval of crosstalk is avoided. We compare images with the results of the classical cross‐correlation imaging condition, which we apply to reverse‐time migrated wavefields and to the up‐ and downgoing wavefields that are retrieved by the Marchenko method.

Evaluation of seismic reliability of gravity dam-reservoirinhomogeneous foundation coupled system

The seismic performance of gravity dam-reservoir-foundation coupled system is investigated utilizing probabilistic approach. In this research, the uncertainties associated with modeling parameters are incorporated in nonlinear response history simulations to realistically quantify their effects on the seismic performance of the system. The methodology is applied to Pine Flat gravity dam and the foundation is considered to be inhomogeneous assuming a constant spatial geometry but with various rock material properties. The sources of uncertainty are taken into account in the reliability analysis using Latin Hypercube Sampling procedure. The effects of the deconvolution process, number of samples, and foundation inhomogeneity are investigated.

Chemical genetic-based phenotypic screen reveals novel regulators of gluconeogenesis in human primary hepatocytes

Insulin resistance is a pathophysiological hallmark of type 2 diabetes and nonalcoholic fatty liver disease. Under the condition of fat accumulation in the liver, suppression of hepatic glucose production by insulin is diminished. In order to gain deeper understanding of dysregulation of glucose production in metabolic diseases, in the present study, we performed an unbiased phenotypic screening in primary human hepatocytes to discover novel mechanisms that regulate gluconeogenesis in the presence of insulin. To optimize phenotypic screening process, we used a chemical genetic screening approach by building a small-molecule library with prior knowledge of activity-based protein profiling. The “positive hits” result from the screen will be small molecules with known protein targets. This makes downstream deconvolution process (i.e., determining the relevant biological targets) less time-consuming. To unbiasedly decipher the molecular targets, we developed a novel statistical method and discovered a set of genes, including DDR3 and CACNA1E that suppressed gluconeogenesis in human hepatocytes. Further investigation, including transcriptional profiling and gene network analysis, was performed to understand the molecular functions of DRD3 and CACNA1E in human hepatocytes.

Blind deconvolution based on cyclostationarity maximization and its application to fault identification

Blind deconvolution algorithms prove to be effective tools for fault identification, being able to extract excitation sources from noisy observations only. In this scenario, the present paper introduces a novel blind deconvolution method based on the generalized Rayleigh quotient and solved by means of an iterative eigenvalue decomposition algorithm. This approach not only is characterized by a weighting matrix that drives the deconvolution process, but can also be easily adapted to arbitrary criteria. Based on this framework, a novel criterion rooted on the maximization of the cyclostationarity of the excitation – as typically encountered with machine faults – is proposed and compared with other blind deconvolution methods existing in the literature. The comparisons involve both synthesized and real vibration signals, taking into account a gear tooth spall and an outer race bearing fault. The results reveal superior capability to recover impulsive cyclostationary sources with respect to other blind deconvolution methods, even in the presence of impulsive noise or under non-constant speed.

Continuous Pain Intensity Estimation from Autonomic Signals with Recurrent Neural Networks.

Pain is usually measured by patient's self-report, which requires patient collaboration. Hence, the development of an objective automatic pain detection method would be useful in many clinical applications and patient populations. Previous studies have explored the feasibility of using physiological autonomic signals to detect the presence of pain. In this study, we focused on continuously estimating experimental heat pain intensity with high temporal resolution from autonomic signals. Specifically, we employed skin conductance deconvolution and point process heart rate variability analysis to continuously evaluate time-varying autonomic parameters, and presented a regression algorithm based on recurrent neural networks.

PyUnfold: A Python package for iterative unfolding

PyUnfold is a Python package for incorporating imperfections of the measurement process into a data analysis pipeline. In an ideal world, we would have access to the perfect detector: an apparatus that makes no error in measuring a desired quantity. However, in real life, detectors have finite resolutions, characteristic biases that cannot be eliminated, less than full detection efficiencies, and statistical and systematic uncertainties. By building a matrix that encodes a detector's smearing of the desired true quantity into the measured observable(s), a deconvolution can be performed that provides an estimate of the true variable. This deconvolution process is known as unfolding. The unfolding method implemented in PyUnfold accomplishes this deconvolution via an iterative procedure, providing results based on physical expectations of the desired quantity. Furthermore, tedious book-keeping for both statistical and systematic errors produces precise final uncertainty estimates.

Sparsity‐enhanced wavelet deconvolution

ABSTRACTWe propose a three‐step bandwidth enhancing wavelet deconvolution process, combining linear inverse filtering and non‐linear reflectivity construction based on a sparseness assumption. The first step is conventional Wiener deconvolution. The second step consists of further spectral whitening outside the spectral bandwidth of the residual wavelet after Wiener deconvolution, i.e., the wavelet resulting from application of the Wiener deconvolution filter to the original wavelet, which usually is not a perfect spike due to band limitations of the original wavelet. We specifically propose a zero‐phase filtered sparse‐spike deconvolution as the second step to recover the reflectivity dominantly outside of the bandwidth of the residual wavelet after Wiener deconvolution. The filter applied to the sparse‐spike deconvolution result is proportional to the deviation of the amplitude spectrum of the residual wavelet from unity, i.e., it is of higher amplitude; the closer the amplitude spectrum of the residual wavelet is to zero, but of very low amplitude, the closer it is to unity. The third step consists of summation of the data from the two first steps, basically adding gradually the contribution from the sparse‐spike deconvolution result at those frequencies at which the residual wavelet after Wiener deconvolution has small amplitudes. We propose to call this technique “sparsity‐enhanced wavelet deconvolution”. We demonstrate the technique on real data with the deconvolution of the (normal‐incidence) source side sea‐surface ghost of marine towed streamer data. We also present the extension of the proposed technique to time‐varying wavelet deconvolution.

Surface potential extraction from electrostatic and Kelvin-probe force microscopy images

A comprehensive comparison study of electrostatic force microscopy (EFM) and Kelvin probe force microscopy (KPFM) is conducted in this manuscript. First, it is theoretically demonstrated that for metallic or semiconductor samples, both the EFM and KPFM signals are a convolution of the sample surface potential with their respective transfer functions. Then, an equivalent point-mass model describing cantilever deflection under distributed loads is developed to reevaluate the cantilever influence on detection signals, and it is shown that the cantilever has no influence on the EFM signal, while it will affect the KPFM signal intensity but not change the resolution. Finally, EFM and KPFM experiments are carried out, and the surface potential is extracted from the EFM and KPFM images by deconvolution processing, respectively. The extracted potential intensity is well consistent with each other and the detection resolution also complies with the theoretical analysis. Our work is helpful to perform a quantitative analysis of EFM and KPFM signals, and the developed point-mass model can also be used for other cantilever beam deflection problems.

Unsupervised Learning for Monaural Source Separation Using Maximization⁻Minimization Algorithm with Time⁻Frequency Deconvolution.

This paper presents an unsupervised learning algorithm for sparse nonnegative matrix factor time–frequency deconvolution with optimized fractional -divergence. The -divergence is a group of cost functions parametrized by a single parameter . The Itakura–Saito divergence, Kullback–Leibler divergence and Least Square distance are special cases that correspond to , respectively. This paper presents a generalized algorithm that uses a flexible range of that includes fractional values. It describes a maximization–minimization (MM) algorithm leading to the development of a fast convergence multiplicative update algorithm with guaranteed convergence. The proposed model operates in the time–frequency domain and decomposes an information-bearing matrix into two-dimensional deconvolution of factor matrices that represent the spectral dictionary and temporal codes. The deconvolution process has been optimized to yield sparse temporal codes through maximizing the likelihood of the observations. The paper also presents a method to estimate the fractional value. The method is demonstrated on separating audio mixtures recorded from a single channel. The paper shows that the extraction of the spectral dictionary and temporal codes is significantly more efficient by using the proposed algorithm and subsequently leads to better source separation performance. Experimental tests and comparisons with other factorization methods have been conducted to verify its efficacy.

Towards monitoring the englacial fracture state using virtual-reflector seismology

In seismology, coda wave interferometry (CWI) is an effective tool to monitor time-lapse changes using later arriving, multiply scattered coda waves. Typically, CWI relies on an estimate of the medium's impulse response. The latter is retrieved through simple time-averaging of receiver-receiver cross-correlations of the ambient field, that is, seismic interferometry (SI). In general, the coda is induced by heterogeneities in the Earth. Being comparatively homogeneous, however, ice bodies such as glaciers and ice sheets exhibit little scattering. In addition, the temporal stability of the time-averaged cross-correlations suffers from temporal variations in the distribution and amplitude of the passive seismic sources. Consequently, application of CWI to ice bodies is currently limited. Nevertheless, fracturing and changes in the englacial macroscopic water content alter the bulk elastic properties of ice bodies, which can be monitored with cryoseismological measurements. To overcome the current limited applicability of CWI to ice bodies, we therefore introduce virtual-reflector seismology (VRS). VRS relies on a so-called multidimensional deconvolution (MDD) process of the time-averaged crosscorrelations. The technique results in the retrieval of a medium response that includes virtual reflections from a contour of receivers enclosing the region of interest (i.e. the region to be monitored). The virtual reflections can be interpreted as artificial coda replacing the (lacking) natural scattered coda. Hence, this artificial coda might be exploited for the purpose of CWI. From an implementation point of view, VRS is similar to SI by MDD, which, as its name suggests, also relies on a multidimensional deconvolution process. SI by MDD, however, does not generate additional virtual reflections. Advantageously, both techniques mitigate spurious coda changes associated with temporal variations in the distribution and amplitude of the passive seismic sources. In this work, we apply SI by MDD and VRS to synthetic and active seismic surface-wave data. The active seismic data were acquired on Glacier de la Plaine Morte, Switzerland. We successfully retrieve virtual reflections through the application of VRS to this active seismic data. In application to both synthetic and active seismic data, we show the potential of VRS to monitor time-lapse changes. In addition, we find that SI by MDD allows for a more accurate determination of phase velocity.

Permuted Coordinate-wise Optimizations Applied to Lp-regularized Image Deconvolution.

Image deconvolution is an ill-posed problem that usually requires prior knowledge for regularizing the feasible solutions. In literature, iterative methods estimate an intrinsic image, minimizing a cost function regularized by specific prior information. However, it is difficult to directly minimize the constrained cost function, if a nondifferentiable regularization (e.g., the sparsity constraint) is employed. In this paper, we propose a nonderivative image deconvolution algorithm that solves the under-constrained problem (i.e., a non-blind image deconvolution) by successively solving the permuted subproblems. The subproblems, arranged in permuted sequences, directly minimize the nondifferentiable cost functions. Various Lp-regularized (0 < p ≤ 1, p = 2) objective functions are utilized to demonstrate the pixel-wise optimization, in which the projection operator generates simplified, low-dimensional subproblems for estimating each pixel. The subproblems, after projection, are dealt with in the corresponding hyperplanes containing the adjacent pixels of each image coordinate. Furthermore, successively solving the subproblems can accelerate the deconvolution process with a linear speed-up, by parallelizing the subproblem sequences. The image deconvolution results with various regularization functionals are presented and the linear speed-up is also demonstrated with a parallelized version of the proposed algorithm. Experimental results demonstrate that the proposed method outperforms the conventional methods in terms of the improved-signal-to-noise ratio and structural similarity index measure.

Nonlinear dynamic response of concrete gravity dams considering the deconvolution process

In order to model any physical system, including concrete dams, one needs to apply simplifications to the real system to make the modeling feasible. To study the dynamic response of concrete dams, relevant assumptions are made regarding foundation modeling, input motion mechanism, dam-reservoir interaction, and material behavior. Some of these simplifications may lead to more conservative and uneconomical outcomes. In this paper, nonlinear time-domain dynamic analyses are conducted to evaluate the effect of the input motion mechanism on the concrete gravity dam response. An ideal model of a dam-reservoir-foundation system, considering the inertia of the foundation, appropriate boundary conditions and precise deconvolved base motions, is selected as the reference model. The process of deconvolution of seismic waves through the frequency- and time-domain approaches is discussed, and suggestions for the selection of an appropriate damping model for the rock foundation are presented. The results are compared to those of the standard model of a massless foundation system as recommended by the US Army Corps of Engineers. The numerical results indicate that, generally, the nonlinear response of the two dam models follow similar patterns, but with a larger amplitude for the massless system. It is noted that the consequence of neglecting the dynamic effect of the foundation can be significant. Additionally, the results demonstrate that the degree of overestimation varies dramatically for the various seismic excitations. The average overestimation of the massless system for the crest displacement, crest acceleration, and the contact opening and sliding are 57%, 45%, 152%, and 90%, respectively. It is further shown that the use of the deconvolved input excitation at the base of the finite foundation derived from the frequency-domain approach can yield discrepancies between the target and convolved surface ground motions, which can also have a considerable effect on the dam response.

Novel identification of ultraviolet/visible Cr6+/Cr3+ optical transitions in borate glasses

Chromium oxide-doped borate glasses with composition xP2O5-(70-x) B2O3- 29.5Li2O-0.5 Cr2O3 (x = 0.0, 0.5, 1.0, 1.2, 1.5 and 2.0 mol %) were prepared following the melt quenching technique. A detailed deconvolution process, applied to their optical absorption spectra, revealed distinct optical transitions from Cr6+/Cr3+ in the ultraviolet/visible spectral regimes, which are simultaneously tuned through slight P2O5 additives. This tuability is mainly derived by the ligand field experienced by the optically-active Cr6+/Cr3+ ions. Interestingly, the intensity reduction of Cr6+/Cr3+ transitions with P2O5 contents allowed smooth variations of the glass color and unravel the, otherwise masked, optical band gap of the homogenous glass host. The coexistence of Cr6+ and Cr3+ was, further, confirmed by mapping its oxidation states via electron spin resonance spectroscopy, while structural rearrangements, produced by P2O5, and Cr6+ vibrational signal were inferred from infrared spectroscopy. The rich optical transitions provided by the coexistence of Cr6+ and Cr3+ within the same host and their simultaneous tuability should find applications in multifunctional optical devices operating in the ultraviolet and visible spectral regimes.

Robust non-destructive measurement system for extraction of ultrasonic wave parameters using the prism technique

Abstract In this paper we propose a measurement system to evaluate the elastic constants of materials using only one experiment. This measurement system is based on the ultrasonic prism technique for the nondestructive measurement of both compressional and shear waves, and on the matching pursuit combined with genetic algorithms (MP-GA) to estimate these two backscattered echo waves. Under this framework, samples are realized in prism-shaped forms using a special molding box. Then, the MP-GA method is used to derive the time of flight of the measured data by a deconvolution process. The backscattered echoes are constructed using a parametric model-based vector for which the corresponding parameters are closely related to the physical properties of the ultrasonic signal propagating through the investigated material. The effectiveness of the measurement system is approved by estimating the elastic properties of homogeneous and non-homogeneous solid materials. Further, several tests have shown that this measurement system works well even for low SNR levels.

Blind Multichannel Deconvolution with Cascaded Kalman Filters

Blind multichannel deconvolution is a demanding technology in modern signal processing field. The proposed algorithm is based on two cascaded Kalman filter. The first Kalman filter performs the blind identification of the multichannel impulse response using the cross-correlation condition; the second Kalman filter carries out the deconvolution process. The two Kalman filters are unified into a common iteration period, so the blind multichannel deconvolution is adaptively executed. Simulation shows that the proposed algorithm works well.