We discuss inverse problems of determining the time-dependent source coefficient for a general class of subelliptic heat equations. We show that a single data at an observation point guarantees the existence of a (smooth) solution pair for the inverse problem. Moreover, additional data at the observation point implies an explicit formula for the time-dependent source coefficient. We also explore an inverse problem with nonlocal additional data, which seems a new approach even in the Laplacian case.
{"title":"Inverse problems of identifying the time-dependent source coefficient for subelliptic heat equations","authors":"M. Ismailov, T. Ozawa, D. Suragan","doi":"10.3934/ipi.2023056","DOIUrl":"https://doi.org/10.3934/ipi.2023056","url":null,"abstract":"We discuss inverse problems of determining the time-dependent source coefficient for a general class of subelliptic heat equations. We show that a single data at an observation point guarantees the existence of a (smooth) solution pair for the inverse problem. Moreover, additional data at the observation point implies an explicit formula for the time-dependent source coefficient. We also explore an inverse problem with nonlocal additional data, which seems a new approach even in the Laplacian case.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139371668","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This work concerns the direct and inverse potential problems for the stochastic diffusion equation driven by a multiplicative time-dependent white noise. The direct problem is to examine the well-posedness of the stochastic diffusion equation for a given potential, while the inverse problem is to determine the potential from the expectation of the solution at a fixed observation point inside the spatial domain. The direct problem is shown to admit a unique and positive mild solution if the initial value is nonnegative. Moreover, an explicit formula is deduced to reconstruct the square of the potential, which leads to the uniqueness of the inverse problem for nonnegative potential functions. Two regularization methods are utilized to overcome the instability of the numerical differentiation in the reconstruction formula. Numerical results show that the methods are effective to reconstruct both smooth and nonsmooth potential functions.
{"title":"An inverse potential problem for the stochastic diffusion equation with a multiplicative white noise","authors":"Xiaoli Feng, Peijun Li, Xu Wang","doi":"10.3934/ipi.2023032","DOIUrl":"https://doi.org/10.3934/ipi.2023032","url":null,"abstract":"This work concerns the direct and inverse potential problems for the stochastic diffusion equation driven by a multiplicative time-dependent white noise. The direct problem is to examine the well-posedness of the stochastic diffusion equation for a given potential, while the inverse problem is to determine the potential from the expectation of the solution at a fixed observation point inside the spatial domain. The direct problem is shown to admit a unique and positive mild solution if the initial value is nonnegative. Moreover, an explicit formula is deduced to reconstruct the square of the potential, which leads to the uniqueness of the inverse problem for nonnegative potential functions. Two regularization methods are utilized to overcome the instability of the numerical differentiation in the reconstruction formula. Numerical results show that the methods are effective to reconstruct both smooth and nonsmooth potential functions.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":1.3,"publicationDate":"2023-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46720630","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we study time-harmonic electromagnetic scattering in two scenarios, where the anomalous scatterer is either a pair of electromagnetic sources or an inhomogeneous medium, both with compact supports. We are mainly concerned with the geometrical inverse scattering problem of recovering the support of the scatterer, independent of its physical contents, by a single far-field measurement. It is assumed that the support of the scatterer (locally) possesses a conical singularity. We establish a local characterisation of the scatterer when invisibility/transparency occurs, showing that its characteristic parameters must vanish locally around the conical point. Using this characterisation, we establish several local and global uniqueness results for the aforementioned inverse scattering problems, showing that visibility must imply unique recovery. In the process, we also establish the local vanishing property of the electromagnetic transmission eigenfunctions around a conical point under the Hölder regularity or a regularity condition in terms of Herglotz approximation.
{"title":"Visibility, invisibility and unique recovery of inverse electromagnetic problems with conical singularities","authors":"Huaian Diao, Xiaoxu Fei, Hongyu Liu, Ke Yang","doi":"10.3934/ipi.2023043","DOIUrl":"https://doi.org/10.3934/ipi.2023043","url":null,"abstract":"In this paper, we study time-harmonic electromagnetic scattering in two scenarios, where the anomalous scatterer is either a pair of electromagnetic sources or an inhomogeneous medium, both with compact supports. We are mainly concerned with the geometrical inverse scattering problem of recovering the support of the scatterer, independent of its physical contents, by a single far-field measurement. It is assumed that the support of the scatterer (locally) possesses a conical singularity. We establish a local characterisation of the scatterer when invisibility/transparency occurs, showing that its characteristic parameters must vanish locally around the conical point. Using this characterisation, we establish several local and global uniqueness results for the aforementioned inverse scattering problems, showing that visibility must imply unique recovery. In the process, we also establish the local vanishing property of the electromagnetic transmission eigenfunctions around a conical point under the Hölder regularity or a regularity condition in terms of Herglotz approximation.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136303578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Image demosaicing and denoising play a critical role in the raw imaging pipeline. These processes have often been treated as independent, without considering their interactions. Indeed, most classic denoising methods handle noisy RGB images, not raw images. Conversely, most demosaicing methods address the demosaicing of noise free images. The real problem is to jointly denoise and demosaic noisy raw images. But the question of how to proceed is still not clarified. In this paper, we carry out extensive experiments and a mathematical analysis to tackle this problem by low complexity algorithms. Indeed, both problems have only been addressed jointly by end-to-end heavy-weight convolutional neural networks (CNNs), which are currently incompatible with low-power portable imaging devices and remain by nature domain (or device) dependent. Our study leads us to conclude that, with moderate noise, demosaicing should be applied first, followed by denoising. This requires a simple adaptation of classic denoising algorithms to demosaiced noise, which we justify and specify. Although our main conclusion is 'demosaic first, then denoise,' we also discover that for high noise, there is a moderate PSNR gain by a more complex strategy: partial CFA denoising followed by demosaicing and by a second denoising on the RGB image. These surprising results are obtained by a black-box optimization of the pipeline, which could be applied to any other pipeline. We validate our results on simulated and real noisy CFA images obtained from several benchmarks.
{"title":"How to best combine demosaicing and denoising?","authors":"Yu Guo, Qiyu Jin, Jean-Michel Morel, Gabriele Facciolo","doi":"10.3934/ipi.2023044","DOIUrl":"https://doi.org/10.3934/ipi.2023044","url":null,"abstract":"Image demosaicing and denoising play a critical role in the raw imaging pipeline. These processes have often been treated as independent, without considering their interactions. Indeed, most classic denoising methods handle noisy RGB images, not raw images. Conversely, most demosaicing methods address the demosaicing of noise free images. The real problem is to jointly denoise and demosaic noisy raw images. But the question of how to proceed is still not clarified. In this paper, we carry out extensive experiments and a mathematical analysis to tackle this problem by low complexity algorithms. Indeed, both problems have only been addressed jointly by end-to-end heavy-weight convolutional neural networks (CNNs), which are currently incompatible with low-power portable imaging devices and remain by nature domain (or device) dependent. Our study leads us to conclude that, with moderate noise, demosaicing should be applied first, followed by denoising. This requires a simple adaptation of classic denoising algorithms to demosaiced noise, which we justify and specify. Although our main conclusion is 'demosaic first, then denoise,' we also discover that for high noise, there is a moderate PSNR gain by a more complex strategy: partial CFA denoising followed by demosaicing and by a second denoising on the RGB image. These surprising results are obtained by a black-box optimization of the pipeline, which could be applied to any other pipeline. We validate our results on simulated and real noisy CFA images obtained from several benchmarks.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136366916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nonlinearity parameter tomography leads to the problem of identifying a coefficient in a nonlinear wave equation (such as the Westervelt equation) modeling ultrasound propagation. In this paper we transfer this into frequency domain, where the Westervelt equation gets replaced by a coupled system of Helmholtz equations with quadratic nonlinearities. For the case of the to-be-determined nonlinearity coefficient being a characteristic function of an unknown, not necessarily connected domain $ D $, we devise and test a reconstruction algorithm based on weighted point source approximations combined with Newton's method. In a more abstract setting, convergence of a regularised Newton type method for this inverse problem is proven by verifying a range invariance condition of the forward operator and establishing injectivity of its linearisation.
{"title":"Nonlinearity parameter imaging in the frequency domain","authors":"Barbara Kaltenbacher, William Rundell","doi":"10.3934/ipi.2023037","DOIUrl":"https://doi.org/10.3934/ipi.2023037","url":null,"abstract":"Nonlinearity parameter tomography leads to the problem of identifying a coefficient in a nonlinear wave equation (such as the Westervelt equation) modeling ultrasound propagation. In this paper we transfer this into frequency domain, where the Westervelt equation gets replaced by a coupled system of Helmholtz equations with quadratic nonlinearities. For the case of the to-be-determined nonlinearity coefficient being a characteristic function of an unknown, not necessarily connected domain $ D $, we devise and test a reconstruction algorithm based on weighted point source approximations combined with Newton's method. In a more abstract setting, convergence of a regularised Newton type method for this inverse problem is proven by verifying a range invariance condition of the forward operator and establishing injectivity of its linearisation.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135056485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we present our approach for the Helsinki Deblur Challenge (HDC2021). The task of this challenge is to deblur images of characters without knowing the point spread function (PSF). The organizers provided a dataset of pairs of sharp and blurred images. Our method consists of three steps: First, we estimate a warping transformation of the images to align the sharp images with the blurred ones. Next, we estimate the PSF using a quasi-Newton method. The estimated PSF allows to generate additional pairs of sharp and blurred images. Finally, we train a deep convolutional neural network to reconstruct the sharp images from the blurred images. Our method is able to successfully reconstruct images from the first 10 stages of the HDC 2021 dataset. Our code is available at https://github.com/hhu-machine-learning/hdc2021-psfnn.
{"title":"Deblurring photographs of characters using deep neural networks","authors":"Thomas Germer, Tobias Uelwer, Stefan Harmeling","doi":"10.3934/ipi.2022057","DOIUrl":"https://doi.org/10.3934/ipi.2022057","url":null,"abstract":"In this paper, we present our approach for the Helsinki Deblur Challenge (HDC2021). The task of this challenge is to deblur images of characters without knowing the point spread function (PSF). The organizers provided a dataset of pairs of sharp and blurred images. Our method consists of three steps: First, we estimate a warping transformation of the images to align the sharp images with the blurred ones. Next, we estimate the PSF using a quasi-Newton method. The estimated PSF allows to generate additional pairs of sharp and blurred images. Finally, we train a deep convolutional neural network to reconstruct the sharp images from the blurred images. Our method is able to successfully reconstruct images from the first 10 stages of the HDC 2021 dataset. Our code is available at https://github.com/hhu-machine-learning/hdc2021-psfnn.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134955794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Silvia Bonettini, Giorgia Franchini, Danilo Pezzi, Marco Prato
In this paper we present a bilevel optimization scheme for the solution of a general image deblurring problem, in which a parametric variational-like approach is encapsulated within a machine learning scheme to provide a high quality reconstructed image with automatically learned parameters. The ingredients of the variational lower level and the machine learning upper one are specifically chosen for the Helsinki Deblur Challenge 2021, in which sequences of letters are asked to be recovered from out-of-focus photographs with increasing levels of blur. Our proposed procedure for the reconstructed image consists in a fixed number of FISTA iterations applied to the minimization of an edge preserving and binarization enforcing regularized least-squares functional. The parameters defining the variational model and the optimization steps, which, unlike most deep learning approaches, all have a precise and interpretable meaning, are learned via either a similarity index or a support vector machine strategy. Numerical experiments on the test images provided by the challenge authors show significant gains with respect to a standard variational approach and performances comparable with those of some of the proposed deep learning based algorithms which require the optimization of millions of parameters.
{"title":"Explainable bilevel optimization: An application to the Helsinki deblur challenge","authors":"Silvia Bonettini, Giorgia Franchini, Danilo Pezzi, Marco Prato","doi":"10.3934/ipi.2022055","DOIUrl":"https://doi.org/10.3934/ipi.2022055","url":null,"abstract":"In this paper we present a bilevel optimization scheme for the solution of a general image deblurring problem, in which a parametric variational-like approach is encapsulated within a machine learning scheme to provide a high quality reconstructed image with automatically learned parameters. The ingredients of the variational lower level and the machine learning upper one are specifically chosen for the Helsinki Deblur Challenge 2021, in which sequences of letters are asked to be recovered from out-of-focus photographs with increasing levels of blur. Our proposed procedure for the reconstructed image consists in a fixed number of FISTA iterations applied to the minimization of an edge preserving and binarization enforcing regularized least-squares functional. The parameters defining the variational model and the optimization steps, which, unlike most deep learning approaches, all have a precise and interpretable meaning, are learned via either a similarity index or a support vector machine strategy. Numerical experiments on the test images provided by the challenge authors show significant gains with respect to a standard variational approach and performances comparable with those of some of the proposed deep learning based algorithms which require the optimization of millions of parameters.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135076321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this note, we generalize the nonlinearity-recovery result in [7] for classical cubic nonlinear Schr"odinger equations to higher-order Schr"odinger equations with a more general nonlinearity. More precisely, we consider a spatially-localized nonlinear higher-order Schr"odinger equation and recover the spatially-localized coefficient by the solutions with data given by small-amplitude wave packets.
{"title":"On recovering the nonlinearity for generalized higher-order Schrödinger equations","authors":"Zachary Lee, Xueying Yu","doi":"10.3934/ipi.2023039","DOIUrl":"https://doi.org/10.3934/ipi.2023039","url":null,"abstract":"In this note, we generalize the nonlinearity-recovery result in [7] for classical cubic nonlinear Schr\"odinger equations to higher-order Schr\"odinger equations with a more general nonlinearity. More precisely, we consider a spatially-localized nonlinear higher-order Schr\"odinger equation and recover the spatially-localized coefficient by the solutions with data given by small-amplitude wave packets.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135400786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper investigates the interior transmission problem for homogeneous media via eigenvalue trajectories parameterized by the magnitude of the refractive index. In the case that the scatterer is the unit disk, we prove that there is a one-to-one correspondence between complex-valued interior transmission eigenvalue trajectories and Dirichlet eigenvalues of the Laplacian which turn out to be exactly the trajectorial limit points as the refractive index tends to infinity. For general simply-connected scatterers in two or three dimensions, a corresponding relation is still open, but further theoretical results and numerical studies indicate a similar connection.
{"title":"On trajectories of complex-valued interior transmission eigenvalues","authors":"Lukas Pieronek, Andreas Kleefeld","doi":"10.3934/ipi.2023041","DOIUrl":"https://doi.org/10.3934/ipi.2023041","url":null,"abstract":"This paper investigates the interior transmission problem for homogeneous media via eigenvalue trajectories parameterized by the magnitude of the refractive index. In the case that the scatterer is the unit disk, we prove that there is a one-to-one correspondence between complex-valued interior transmission eigenvalue trajectories and Dirichlet eigenvalues of the Laplacian which turn out to be exactly the trajectorial limit points as the refractive index tends to infinity. For general simply-connected scatterers in two or three dimensions, a corresponding relation is still open, but further theoretical results and numerical studies indicate a similar connection.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135595500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this article, we introduce a novel direct sampling indicator function for solving the inverse source problem (ISP) associated with the Helmholtz equation. The proposed method is rigorously evaluated through numerical analysis of both singular and general sources, providing a comprehensive assessment of its performance. Several numerical examples are given to validate the theoretical estimations.
{"title":"Error estimation to the direct sampling method for the inverse acoustic source problem with multi-frequency data","authors":"Xia Ji, Yuling Jiao, Xiliang Lu, Fengru Wang","doi":"10.3934/ipi.2023042","DOIUrl":"https://doi.org/10.3934/ipi.2023042","url":null,"abstract":"In this article, we introduce a novel direct sampling indicator function for solving the inverse source problem (ISP) associated with the Helmholtz equation. The proposed method is rigorously evaluated through numerical analysis of both singular and general sources, providing a comprehensive assessment of its performance. Several numerical examples are given to validate the theoretical estimations.","PeriodicalId":50274,"journal":{"name":"Inverse Problems and Imaging","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135700484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：