Iterative Deconvolution Method Research Articles

Multi-scale CLEAN for Fourier-based hard X-ray solar imaging

Abstract Multi-scale deconvolution is an ill-posed inverse problem in imaging, with applications ranging from microscopy, through medical imaging, to astronomical remote sensing. In the case of high-energy space telescopes, multi-scale deconvolution algorithms need to account for the peculiar property of native measurements, which are sparse samples of the Fourier transform of the incoming radiation. The present paper proposes a multi-scale version of CLEAN, which is the most popular iterative deconvolution method in Fourier-based astronomical imaging. Using synthetic data generated according to a simulated but realistic source configuration, we show that this multi-scale version of CLEAN performs better than the original one in terms of accuracy, photometry, and regularization. Further, the application to a data set measured by the NASA Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI) shows the ability of multi-scale CLEAN to reconstruct rather complex flaring topographies.

Unbiased CLEAN for STIX in Solar Orbiter

CLEAN is an iterative deconvolution method for radio and hard-X-ray solar imaging. In a specific step of its pipeline, CLEAN requires the convolution between an idealized version of the instrumental point-spread function (PSF), and a map collecting point sources located at positions from where most of the flaring radiation is emitted. This step has highly heuristic motivations and the shape of the idealized PSF, which depends on the user’s choice, impacts the shape of the reconstruction. This study introduces a user-independent release of CLEAN for image reconstruction from observations recorded by the Spectrometer/Telescope for Imaging X-rays (STIX) on board Solar Orbiter. Specifically, we show here that this unbiased release of CLEAN outperforms the standard version of the algorithm, with reconstructions in line with the ones offered by other imaging methods developed in the STIX framework.

Moho depths beneath the European Alps: a homogeneously processed map and receiver functions database

Abstract. We use seismic waveform data from the AlpArray Seismic Network and three other temporary seismic networks, to perform receiver function (RF) calculations and time-to-depth migration to update the knowledge of the Moho discontinuity beneath the broader European Alps. In particular, we set up a homogeneous processing scheme to compute RFs using the time-domain iterative deconvolution method and apply consistent quality control to yield 112 205 high-quality RFs. We then perform time-to-depth migration in a newly implemented 3D spherical coordinate system using a European-scale reference P and S wave velocity model. This approach, together with the dense data coverage, provide us with a 3D migrated volume, from which we present migrated profiles that reflect the first-order crustal thickness structure. We create a detailed Moho map by manually picking the discontinuity in a set of orthogonal profiles covering the entire area. We make the RF dataset, the software for the entire processing workflow, as well as the Moho map, openly available; these open-access datasets and results will allow other researchers to build on the current study.

Fast Resolution Enhancement for Real Beam Mapping Using the Parallel Iterative Deconvolution Method

Super-resolution methods for real beam mapping (RBM) imagery play a significant role in many microwave remote sensing applications. However, the existing super-resolution methods require high-dimensional matrix operations in the case of wide-field imaging, which makes it difficult to satisfy the requirements of real-time signal processing. To solve this problem, this paper introduces an improved Poisson distribution-based maximum likelihood (IPML) method by adding an adaptive iterative acceleration factor to effectively improve the algorithm convergence speed without introducing high-dimensional matrix operations. Furthermore, a GPU-based parallel processing architecture is proposed through the multithreading characteristics of the computing platform, and a cooperative CPU–GPU working model is constructed. This can realize the parallel optimization of the echo reception, preprocessing, and super-resolution processing. We verify that the proposed parallel super-resolution method can significantly improve the computational efficiency without sacrificing performance, using a real dataset.

Research on Transient Temperature Rise Measurement Method for Semiconductor Devices Based on Photothermal Reflection

In this article, a transient temperature rise measurement method for semiconductor devices based on photothermal reflection is proposed. The relationship between the reflectivity of a 360 nm laser beam and the device temperature is studied. Through the appropriate design of the optical circuit and the data acquisition system, measurement of the transient temperature rise characteristics of GaN-based microwave power devices is realized. In combination with resistor-capacitor network impulse response theory, the transient temperature rise curve is processed using the Bayesian iterative deconvolution method. By extracting the thermal time constant, nondestructive characterization of the temperature increase and thermal resistance characteristics of each material layer in the heat flow path is achieved, and the complementary relationship between the temperature change in the active region during heating and the device heat dissipation characteristics is verified. Additionally, the forward Schottky junction characteristics method and infrared thermal imaging are used to measure the channel temperature for verification of the proposed thermal reflection method. These measurements indicate that the proposed method can provide accurate results.

Resolving Intrinsic Modulation Spectral Lines from Electro-Optic Modulation Spectra Based on Boosted Iterative Deconvolution

Optical spectrum analysis is the most direct and effective method for characterizing electro-optic modulation spectra. According to the Rayleigh criterion, the best resolution of an optical spectrum analysis is limited by the point spread function of an optical spectrum analyzer (OSA) and the relative intensity of closely spaced spectral lines. In this paper, we propose a boosted iterative deconvolution (BID) method to resolve the intrinsic modulation of spectral lines from the measured optical carrier and modulation spectra. In our scheme, the electro-optic modulation spectrum is considered the convolution of the optical carrier spectrum and the intrinsic δ-function modulation spectrum, and the BID method enables fast and accurate extraction of the δ-function spectral lines from the measured modulation spectrum. The proof-of-concept experiment demonstrates that our method can improve the resolution of OSA by 10–30 p.m. at different relative intensities, with a best resolution of 10 p.m. in the iso-intensity case and wavelength errors of less than 2 p.m., which largely improves the measurement resolution and accuracy of the modulation spectrum.

Iterative blind deconvolution method for dwell-time adjustment.

Negative values occur in the calculation results when a fast Fourier transform is used in the algorithm to calculate the dwell time via the dwell-time adjustment algorithm for optical element grinding. In the present study, a processing method for negative values is used in the Fourier iterative Ayers-Dainty (AD) algorithm. The method in which the range of the unit removal shape was adopted in the AD algorithm resulted in a smaller error and enabled the error to be minimized using the quantized dwell-time distribution output.

Remote sensing of raindrop size distribution using the coherent Doppler lidar.

The coherent Doppler wind lidar (CDL) shows capability in precipitation detection. Retrieval of the raindrop size distribution (DSD) using CDL is still challenging work, as both accurate backscattering cross section at the working wavelength and reflectivity spectrum of raindrop are required. Firstly, the Mie theory and the vectorial complex ray model (VCRM) are applied to calculate backscattering cross section for small spheric raindrops and large oblate raindrops, respectively. Secondly, an iterative deconvolution method is proposed to separate the reflectivity spectrum of raindrop from the lidar power spectrum, which is a superposition of rain and aerosol components. An accompanying aerosol signal model considering the effect of temporal window, from the same height and time, is used to improve the accuracy and robustness of the iteration. In experiment, a co-located micro rain radar (MRR) is used for comparison. Good agreements are obtained despite tremendous differences in wavelength and scattering characteristics. As an example, at 600 m height, the R2 of linear fitting to the mean rain velocity and mean raindrop diameter between CDL and MRR are 0.96 and 0.93, respectively.

Modularized data-driven reconstruction framework for nonideal focal spot effect elimination in computed tomography.

High-performance computed tomography (CT) plays a vital role in clinical decision-making. However, the performance of CT imaging is adversely affected by the nonideal focal spot size of the x-ray source or degraded by an enlarged focal spot size due to aging. In this work, we aim to develop a deep learning-based strategy to mitigate the problem so that high spatial resolution CT images can be obtained even in the case of a nonideal x-ray source. To reconstruct high-quality CT images from blurred sinograms via joint image and sinogram learning, a cross-domain hybrid model is formulated via deep learning into a modularized data-driven reconstruction (MDR) framework. The proposed MDR framework comprises several blocks, and all the blocks share the same network architecture and network parameters. In essence, each block utilizes two sub-models to generate an estimated blur kernel and a high-quality CT image simultaneously. In this way, our framework generates not only a final high-quality CT image but also a series of intermediate images with gradually improved anatomical details, enhancing the visual perception for clinicians through the dynamic process. We used simulated training datasets to train our model in an end-to-end manner and tested our model on both simulated and realistic experimental datasets. On the simulated testing datasets, our approach increases the information fidelity criterion (IFC) by up to 34.2%, the universal quality index (UQI) by up to 20.3%, the signal-to-noise (SNR) by up to 6.7%, and reduces the root mean square error (RMSE) by up to 10.5% as compared with FBP. Compared with the iterative deconvolution method (NSM), MDR increases IFC by up to 24.7%, UQI by up to 16.7%, SNR by up to 6.0%, and reduces RMSE by up to 9.4%. In the modulation transfer function (MTF) experiment, our method improves the MTF50% by 34.5% and MTF10% by 18.7% as compared with FBP, Similarly remarkably, our method improves MTF50% by 14.3% and MTF10% by 0.9% as compared with NSM. Also, our method shows better imaging results in the edge of bony structures and other tiny structures in the experiments using phantom consisting of ham and a bottle of peanuts. A modularized data-driven CT reconstruction framework is established to mitigate the blurring effect caused by a nonideal x-ray source with relatively large focal spot. The proposed method enables us to obtain high-resolution images with less ideal x-ray source.

Moho depth variations and Vp/Vs ratio in the Zagros (Iran) from teleseismic converted waves

Variation of Moho depth and Vp/Vs ratio beneath the Zagros using 3 years of teleseismic recording at twenty permanent broadband seismic stations of the Iranian Seismological Center (IRSC) where are scattered all over Zagros is presented in this paper. The results of this study are the first receiver function study for the Moho depth beneath these stations on this new dataset. We determine receiver functions by the time domain iterative deconvolution method and Moho depth and Vp/Vs ratio using a modified Zhu and Kanamori method. Our results indicate that the Moho depth beneath the Zagros Mountain Range varies from ~ 30–35 km in western Zagros near the Arabian platform, up to ~ 45 km beneath Central Zagros. In the main Zagros zone, our results indicate a uniform pattern for the Moho depth. The Moho depth reaches ~ 50–54 km beneath stations near Sanandaj-Sirjan Zone (SSZ), indicating a considerable increase of Moho depth from the west toward the east of Zagros. The depth of Moho in the southeast of Zagros is less than other parts of Zagros. We found deepening of the Moho depths beneath stations near the SSZ which can be due to the overthrusting of the crust of Zagros by the crust of central Iran along the MZT. The average Vp/Vs ratio in Zagros is 1.8.

Sediment and Crustal Structure Beneath East Kalimantan and East Java, Indonesia From Receiver Function

We investigate the sediment and Earth’s crustal structure beneath the East Kalimantan and East Java by modelling the stacked receiver function from six seismic broadband stations belong to the Meteorology, Climatology and Geophysics Agency of Indonesia (BMKG). The calculation of receiver functions from 60 teleseismic earthquake events using iterative time-domain deconvolution method. Inversion result shows a significant variation in the sediment layer thickness from 1 km to 4 km beneath East Java. It is correlated with the South East Java basinal area (station KRK) and the North East Java Basin (station GRJI and BWJI). The sediment layer of about 5 to 10 km beneath East Kalimantan is associated with the Kutai Basin (station SGKI, SMKI, and BKB). Average Sediment layer shear velocity (Vs) beneath all station is observed to be less than 2.31 km/s. Crustal thickness beneath East Java varies between 20 to 40 km and 34 to 50 km beneath East Kalimantan, with average crustal shear wave velocity, Vs 4.60 km/s.

Applications of non-negative iterative deconvolution method in the analysis of alpha-particle spectra

Different from the regular method with peak-shape function fitting, in this work, the idea of alpha-spectra analysis using non-negative iterative deconvolution method is proposed. Two non-negative iterative algorithms, the Boosted Gold and the Boosted Richardson–Lucy, were applied to unfold and analyze alpha-particle spectra. The detector response matrix was constructed with the AASI, which is a Monte Carlo simulation software specifically for alpha-particle spectra. A $$^{244}$$Cm alpha-particle source was measured and tested for peak-position deconvolution. A $$^{238+239+240}$$Pu alpha-particle spectrum was used to prove the validity of quantitative analysis by deconvolution algorithms. Both the Boosted Gold and the Boosted Richardson–Lucy can get accurate results for peak positions and content ratios. In terms of running speed, the Boosted Gold was faster than the Boosted Richardson–Lucy.

Directional Surface Wave Spectra from Point Measurements of Height and Slope

AbstractWe describe here a method for recovering directional ocean surface wave spectra obtained from height and slope measurements made over a small area, the iterative deconvolution method (IDM). We show that IDM is a more reliable method for estimating directional wave spectra than more common spectral estimation techniques by comparing it with the widely used maximum entropy method (MEM). IDM is based on the observation that pitch–roll buoys produce directional spectra that are the true spectra convolved with an angular windowing function and are therefore much broader than the true spectra. We test IDM against simulated data and find that it does a better job of retrieving the known input spectra than does MEM, which often produces false double peaks or incorrect angular widths. We compare IDM recoveries to spectra obtained using a nonstandard processing technique, the wavelet directional method (WDM) on data from a compact array of wave staffs on Lake Ontario. We find that IDM produces directional wave spectra very nearly identical to those obtained using WDM, verifying both techniques. Finally, we processed standard NDBC buoy directional spectra and showed that IDM recovers ocean wave spectra that narrow in the Strait of Juan de Fuca and that follow a changing wind in the expected manner. Neither of these phenomena are reliably obtained using MEM due to its tendency to produce false bimodal peaks and peaks that are too narrow.

An iterative sparse deconvolution method for simultaneous multicolor 19 F-MRI of multiple contrast agents.

Purpose 19F‐MRI is gaining widespread interest for cell tracking and quantification of immune and inflammatory cells in vivo. Different fluorinated compounds can be discriminated based on their characteristic MR spectra, allowing in vivo imaging of multiple 19F compounds simultaneously, so‐called multicolor 19F‐MRI. We introduce a method for multicolor 19F‐MRI using an iterative sparse deconvolution method to separate different 19F compounds and remove chemical shift artifacts arising from multiple resonances.MethodsThe method employs cycling of the readout gradient direction to alternate the spatial orientation of the off‐resonance chemical shift artifacts, which are subsequently removed by iterative sparse deconvolution. Noise robustness and separation was investigated by numerical simulations. Mixtures of fluorinated oils (PFCE and PFOB) were measured on a 7T MR scanner to identify the relation between 19F signal intensity and compound concentration. The method was validated in a mouse model after intramuscular injection of fluorine probes, as well as after intravascular injection.ResultsNumerical simulations show efficient separation of 19F compounds, even at low signal‐to‐noise ratio. Reliable chemical shift artifact removal and separation of PFCE and PFOB signals was achieved in phantoms and in vivo. Signal intensities correlated excellently to the relative 19F compound concentrations (r−2 = 0.966/0.990 for PFOB/PFCE).ConclusionsThe method requires minimal sequence adaptation and is therefore easily implemented on different MRI systems. Simulations, phantom experiments, and in‐vivo measurements in mice showed effective separation and removal of chemical shift artifacts below noise level. We foresee applicability for simultaneous in‐vivo imaging of 19F‐containing fluorine probes or for detection of 19F‐labeled cell populations.

Improvement in quantitative phase mapping by a hard x-ray microscope equipped with a Lau interferometer

X-ray phase contrast imaging offers a complementary modality next to conventional absorption. To reveal internal structures of soft materials in high resolution, an X-ray phase imaging microscope has been developed by installing a Lau interferometer into a commercially available ZEISS Xradia 800 Ultra-nano computed tomography X-ray microscope. On this system, a twin phase image is once generated through measurements of interference fringes. In order to produce a quantitative X-ray phase map, the twin phase image should be treated by deconvolution. However, conventional deconvolution results showed a reduction in image quality due to noise and artifacts. Therefore, we propose and evaluate the performance of an iterative deconvolution method. The results show that the proposed deconvolution method does not affect spatial resolution, increases the signal to noise ratio compared to that of the twin phase image, and is capable of imaging complex structures.

Moho depth variations and Vp/Vs ratios in the seismotectonic zones of Central Iran, Eastern Iran, and Makran: using a modified Zhu and Kanamori method

Variations of the Moho depth and Vp/Vs ratio beneath the Central Iran, Eastern Iran, and Makran regions using a modified Zhu and Kanamori method are presented in this paper. Receiver functions have been calculated from 3 years of teleseismic recordings at seventeen broadband stations using a time-domain iterative deconvolution method. Our results indicate that Moho depth changes from 56 km beneath the western part of Central Iran to 40 km in Central Iran and 38 km in Eastern Iran. The relatively thick crust beneath western margin of Central Iran is due to the close proximity of this region to the Sanandaj-Sirjan Zone and the Urumieh-Dokhtar magmatic arc that are associated with underthrusting of the Arabian plate beneath Central Zagros along the main Zagros reverse fault. Also, crustal thickening of this region would be due to volcanic intrusions, magmatic. The reduction of crustal thickness beneath the eastern stations in this region could be due to the absence of a collisional zone in the tectonic evolution of the region. The results beneath the CHBR station located in Makran indicate a value of 31 km for crustal thickness and 1.70 for the Vp/Vs ratio. It is associated to subduction of the oceanic crust of the Arabian plates, beneath the southern part of the Makran crust with a very low slope. The crust beneath Central and Eastern Iran has a moderate thickness which decreases from west to east.

Average crustal thickness and Poisson’s ratio beneath the Kali River Valley, Kumaon Himalaya

Crustal thickness and Poisson's ratios were estimated at 13 broadband seismological stations established in a profile along the Kali river valley, Kumaon (Central) Himalaya. The seismological profile extends from the Indo-Gangetic plain (IGP) in the south to the Higher Himalaya in the north, passing through the Sub and Lesser Himalaya. The Receiver Function (RF) method has been adopted to investigate the crustal structure beneath the profile. Time domain iterative deconvolution method has been adopted for RF computation. The H-k stacking method has been adopted to analyze the RFs for obtaining average crustal thickness and Poisson’s ratio (σ) of the crust beneath each station. The study reveals that the crustal thickness beneath the IGP is ∼38 km which gradually increases up to ∼41 km at the northernmost station located in the Higher Himalaya. The values of σ vary from 0.23 to 0.28. The value of σ is low in the Sub-Himalaya and outer Lesser Himalaya suggesting more of a felsic composition of crust. A Significantly high value of σ (∼0.28) is observed in the Dharchula region of the Chiplakot Crystalline belt which is coincident with a large number of micro-to-moderate magnitude earthquakes forming a cluster at shallow to mid-crustal depths. The high Poisson’s ratio is possibly due to the presence of fluids/partial melts at mid-crustal depth. The presence of fluids influences the rheological property and controls the mechanical and shear strength of crustal rocks producing a cluster of seismicity observed beneath the Dharchula region.

A spectrometer for ultrashort gamma-ray pulses with photon energies greater than 10 MeV.

We present a design for a pixelated scintillator based gamma-ray spectrometer for non-linear inverse Compton scattering experiments. By colliding a laser wakefield accelerated electron beam with a tightly focused, intense laser pulse, gamma-ray photons up to 100 MeV energies and with few femtosecond duration may be produced. To measure the energy spectrum and angular distribution, a 33 × 47 array of cesium-iodide crystals was oriented such that the 47 crystal length axis was parallel to the gamma-ray beam and the 33 crystal length axis was oriented in the vertical direction. Using an iterative deconvolution method similar to the YOGI code, modeling of the scintillator response using GEANT4 and fitting to a quantum Monte Carlo calculated photon spectrum, we are able to extract the gamma ray spectra generated by the inverse Compton interaction.

Spatial and Temperature Resolutions of Magnetic Nanoparticle Temperature Imaging with a Scanning Magnetic Particle Spectrometer.

This paper quantitatively investigates the spatial and temperature resolutions of magnetic nanoparticle (MNP) temperature imaging with a multiline phantom filled with MNPs. The multiline phantom in total consists of seven lines with different distances between two adjacent lines. A scanning magnetic particle spectrometer is used to measure the spatial distributions of the MNP harmonics for MNP concentration and temperature imaging, whereas an iterative deconvolution method is used to improve the spatial resolution. A modulation transfer function calculated from the MNP concentration image is used to quantitatively present the spatial resolution, whereas the standard deviation of the measured temperatures is used to quantitatively present the temperature resolution. The spatial resolution is about 4 mm while the temperature resolution is about 1.0 K without deconvolution. With increasing the number of the iterative loops in the deconvolution, the spatial resolution is improved to 2 mm while the temperature resolution is worsened to about 9.6 K due to deconvolution-based oscillation.

A mathematical deconvolution formulation for superficial dose distribution measurement by Cerenkov light dosimetry.

Cerenkov photons are created by high-energy radiation beams used for radiation therapy. In this study, we developed a Cerenkov light dosimetry technique to obtain a two-dimensional dose distribution in a superficial region of medium from the images of Cerenkov photons by using a deconvolution method. An integral equationwas derived to represent the Cerenkov photon image acquired by a camera for a given incident high-energy photon beam by using convolution kernels. Subsequently, an equationrelating the planar dose at a depth to a Cerenkov photon image using the well-known relationship between the incident beam fluence and the dose distribution in a medium was obtained. The final equationcontained a convolution kernel called the Cerenkov dose scatter function (CDSF). The CDSF function was obtained by deconvolving the Cerenkov scatter function (CSF) with the dose scatter function (DSF). The GAMOS (Geant4-based Architecture for Medicine-Oriented Simulations) Monte Carlo particle simulation software was used to obtain the CSF and DSF. The dose distribution was calculated from the Cerenkov photon intensity data using an iterative deconvolution method with the CDSF. The theoretical formulation was experimentally evaluated by using an optical phantom irradiated by high-energy photon beams. The intensity of the deconvolved Cerenkov photon image showed linear dependence on the dose rate and the photon beam energy. The relative intensity showed a field size dependence similar to the beam output factor. Deconvolved Cerenkov images showed improvement in dose profiles compared with the raw image data. In particular, the deconvolution significantly improved the agreement in the high dose gradient region, such as in the penumbra. Deconvolution with a single iteration was found to provide the most accurate solution of the dose. Two-dimensional dose distributions of the deconvolved Cerenkov images agreed well with the reference distributions for both square fields and a multileaf collimator (MLC) defined, irregularly shaped field. The proposed technique improved the accuracy of the Cerenkov photon dosimetry in the penumbra region. The results of this study showed initial validation of the deconvolution method for beam profile measurements in a homogeneous media. The new formulation accounted for the physical processes of Cerenkov photon transport in the medium more accurately than previously published methods.