Image restoration methods are commonly used to improve the quality of astronomical images. In recent years, developments of deep neural networks and increments of the number of astronomical images have evoked a lot of data--driven image restoration methods. However, most of these methods belong to supervised learning algorithms, which require paired images either from real observations or simulated data as training set. For some applications, it is hard to get enough paired images from real observations and simulated images are quite different from real observed ones. In this paper, we propose a new data--driven image restoration method based on generative adversarial networks with option--driven learning. Our method uses several high resolution images as references and applies different learning strategies when the number of reference images is different. For sky surveys with variable observation conditions, our method can obtain very stable image restoration results, regardless of the number of reference images.
AstroClearNet: Deep image prior for multi-frame astronomical image restoration
Department of Applied Mathematics and Statistics, Johns Hopkins University（应用数学与统计学系，约翰霍普金斯大学）；Johns Hopkins University（约翰霍普金斯大学）；Department of Computer Science, Johns Hopkins University（计算机科学系，约翰霍普金斯大学）；Department of Physics and Astronomy, Johns Hopkins University（物理与天文学系，约翰霍普金斯大学）
Image Restoration with Point Spread Function Regularization and Active Learning
College of Electronic Information and Optical Engineering, Taiyuan（太原电子信息与光学工程学院）；Peng Cheng Lab, Shenzhen（鹏城实验室）；Department of Physics, Durham University（杜伦大学物理系）；National Astronomical Observatories, Beijing（中国科学院国家天文台）；Yunnan Observatories, Kunming（中国科学院云南天文台）
Astronomical images suffer a constant presence of multiple defects that are consequences of the intrinsic properties of the acquisition equipments, and atmospheric conditions. One of the most frequent defects in astronomical imaging is the presence of additive noise which makes a denoising step mandatory before processing data. During the last decade, a particular modeling scheme, based on sparse representations, has drawn the attention of an ever growing community of researchers. Sparse representations offer a promising framework to many image and signal processing tasks, especially denoising…
Learning Efficient and Effective Trajectories for Differential Equation-based Image Restoration
Department of Computer Science, City University of Hong Kong（香港城市大学计算机科学系）；School of Computing and Information Sciences, Saint Francis University（圣芳济大学计算与信息科学学院）；School of Engineering, Huaqiao University（华侨大学工程学院）；School of Information Science and Engineering, Huaqiao University（华侨大学信息科学与工程学院）
High-capacity image steganography, aimed at concealing a secret image in a cover image, is a technique to preserve sensitive data, e.g., faces and fingerprints. Previous methods focus on the security during transmission and subsequently run a risk of privacy leakage after the restoration of secret images at the receiving end. To address this issue, we propose a framework, called Multitask Identity-Aware Image Steganography (MIAIS), to achieve direct recognition on container images without restoring secret images. The key issue of the direct recognition is to preserve identity information of se…
There are many resources useful for processing images, most of them freely available and quite friendly to use. In spite of this abundance of tools, a study of the processing methods is still worthy of efforts. Here, we want to discuss the possibilities arising from the use of fractional differential calculus. This calculus evolved in the research field of pure mathematics until 1920, when applied science started to use it. Only recently, fractional calculus was involved in image processing methods. As we shall see, the fractional calculation is able to enhance the quality of images, with inte…
In this paper we present a general forward fitting method for multichannel image restoration based on regularized chi2. We introduce separable regularizations that account for the dynamic of the model and take advantage of the continuities present in the data, leaving only two hyper-parameters to tune. We illustrate a practical implementation of this method in the context of host galaxy subtraction for the Nearby SuperNova factory. We show that the image restoration obtained fulfills the stringent requirements on bias and photometricity needed by this program. The reconstruction yields sub-per…
Normalizing flow models using invertible neural networks (INN) have been widely investigated for successful generative image super-resolution (SR) by learning the transformation between the normal distribution of latent variable $z$ and the conditional distribution of high-resolution (HR) images gave a low-resolution (LR) input. Recently, image rescaling models like IRN utilize the bidirectional nature of INN to push the performance limit of image upscaling by optimizing the downscaling and upscaling steps jointly. While the random sampling of latent variable $z$ is useful in generating divers…
Third Military Medical University（第三军医大学）；National University of Defense Technology（国防科技大学）
The recovery of images from the observations that are degraded by a linear operator and further corrupted by Poisson noise is an important task in modern imaging applications such as astronomical and biomedical ones. Gradient-based regularizers involve the popular total variation semi-norm have become standard techniques for Poisson image restoration due to its edge-preserving ability. Various efficient algorithms have been developed for solving the corresponding minimization problem with non-smooth regularization terms. In this paper, motivated by the idea of the alternating direction minimiz…
The implementation of fractional differential calculations can give new possibilities for image processing tools, in particular for those that are devoted to astronomical images analysis. As discussed in arxiv:0910.2381, the fractional differentiation is able to enhance the quality of images, with interesting effects in edge detection and image restoration. Here, we propose the AstroFracTool, developed to provide a simple yet powerful enhancement tool-set for astronomical images. This tool works evaluating the fractional gradient of an image map. It can help produce an output image useful for …
In many professional fields, such as medicine, remote sensing and sciences, users often demand image compression methods to be mathematically lossless. But lossless image coding has a rather low compression ratio (around 2:1 for natural images). The only known technique to achieve significant compression while meeting the stringent fidelity requirements is the methodology of $\ell_\infty$-constrained coding that was developed and standardized in nineties. We make a major progress in $\ell_\infty$-constrained image coding after two decades, by developing a novel CNN-based soft $\ell_\infty$-con…
Institute for Computer Graphics and Vision, Graz University of Technology（格拉茨工业大学计算机图形与视觉研究所）
A trainable filter-based higher-order Markov Random Fields (MRFs) model - the so called Fields of Experts (FoE), has proved a highly effective image prior model for many classic image restoration problems. Generally, two options are available to incorporate the learned FoE prior in the inference procedure: (1) sampling-based minimum mean square error (MMSE) estimate, and (2) energy minimization-based maximum a posteriori (MAP) estimate. This letter is devoted to the FoE prior based single image super resolution (SR) problem, and we suggest to make use of the MAP estimate for inference based on…
We present a performance test of the Point Spread Function deconvolution algorithm applied to astronomical Integral Field Unit (IFU) Spectroscopy data for restoration of galaxy kinematics. We deconvolve the IFU data by applying the Lucy-Richardson algorithm to the 2D image slice at each wavelength. We demonstrate that the algorithm can effectively recover the true stellar kinematics of the galaxy, by using mock IFU data with diverse combination of surface brightness profile, S/N, line-of-sight geometry and Line-Of-Sight Velocity Distribution (LOSVD). In addition, we show that the proxy of the …
We present AMADA, an interactive web application to analyse multidimensional datasets. The user uploads a simple ASCII file and AMADA performs a number of exploratory analysis together with contemporary visualizations diagnostics. The package performs a hierarchical clustering in the parameter space, and the user can choose among linear, monotonic or non-linear correlation analysis. AMADA provides a number of clustering visualization diagnostics such as heatmaps, dendrograms, chord diagrams, and graphs. In addition, AMADA has the option to run a standard or robust principal components analysis…
In a recent paper, published at arXiv:0910.2381, we started a discussion on the new possibilities arising from the use of fractional differential calculus in image processing. We have seen that the fractional calculation is able to enhance the quality of images, with interesting possibilities in edge detection and image restoration. Here, we want to discuss more deeply its role as a tool for the processing of astronomical images. In particular, the fractional differentiation can help produce a 'content-matter' based image from a pretty astronomical image that can be used for more research and …
The world astronomical image archives represent huge opportunities to time-domain astronomy sciences and other hot topics such as space defense, and astronomical observatories should improve this wealth and make it more accessible in the big data era. In 2010 we introduced the Mega-Archive database and the Mega-Precovery server for data mining images containing Solar system bodies, with focus on near Earth asteroids (NEAs). This paper presents the improvements and introduces some new related data mining tools developed during the last five years. Currently, the Mega-Archive has indexed 15 mill…
We introduce and publicly release a new code, ADAPTSMOOTH, which serves to smooth astronomical images in an adaptive fashion, in order to enhance the signal-to-noise ratio (S/N). The adaptive smoothing scheme allows to take full advantage of the spatially resolved photometric information contained in an image in that at any location the minimal smoothing is applied to reach the requested S/N. Support is given to match more images on the same smoothing length, such that proper estimates of local colours can be done, with a big potential impact on multi-wavelength studies of extended sources (ga…
PAPERCLIP: Associating Astronomical Observations and Natural Language with Multi-Modal Models
The NSF AI Institute for Artificial Intelligence and Fundamental Interactions（美国国家科学基金会人工智能与基础交互研究所）；Center for Theoretical Physics, Massachusetts Institute of Technology（麻省理工学院理论物理中心）；Department of Physics, Harvard University（哈佛大学物理系）；Department of Physics, Massachusetts Institute of Technology（麻省理工学院物理系）
We present the 'simage' software suite for the simulation of artificial extragalactic images, based empirically around real observations of the Hubble Ultra Deep Field (UDF). The simulations reproduce galaxies with realistic and complex morphologies via the modeling of UDF galaxies as shapelets. Images can be created in the B, V, i and z bands for both space- and ground-based telescopes and instruments. The simulated images can be produced for any required field size, exposure time, Point Spread Function (PSF), telescope mirror size, pixel resolution, field star density, and a variety of detec…
Astronomical instruments make intensity measurements; any precise astronomical experiment ought to involve modeling those measurements. People make catalogues, but because a catalogue requires hard decisions about calibration and detection, no catalogue can contain all of the information in the raw pixels relevant to most scientific investigations. Here we advocate making catalogue-like data outputs that permit investigators to test hypotheses with almost the power of the original image pixels. The key is to provide users with approximations to likelihood tests against the raw image pixels. We…
Written in Python and utilising ParselTongue to interface with the Astronomical Image Processing System (AIPS), the e-MERLIN data reduction pipeline is intended to automate the procedures required in processing and calibrating radio astronomy data from the e-MERLIN correlator. Driven by a plain text file of input parameters, the pipeline is modular and can be run in stages by the user, depending on requirements. The software includes options to load raw data, average in time and/or frequency, flag known sources of interference, flag more comprehensively with SERPent, carry out some or all of t…
The New Vacuum Solar Telescope (NVST) is a 1-m solar telescope that aims to observe the fine structures in both the photosphere and the chromosphere of the Sun. The observational data acquired simultaneously from one channel for the chromosphere and two channels for the photosphere bring great challenges to the data storage of NVST. The multi-channel instruments of NVST, including scientific cameras and multi-band spectrometers, generate at least 3 terabytes data per day and require high access performance while storing massive short-exposure images. It is worth studying and implementing a sto…
INAF, Osservatorio Astronomico di Padova（意大利国家天体物理研究所帕多瓦天文台）；Dipartimento di Fisica e Astronomia ‘Galileo Galilei’, Università di Padova（帕多瓦大学伽利莱物理与天文学系）
We present a new Python library called vaex, to handle extremely large tabular datasets, such as astronomical catalogues like the Gaia catalogue, N-body simulations or any other regular datasets which can be structured in rows and columns. Fast computations of statistics on regular N-dimensional grids allows analysis and visualization in the order of a billion rows per second. We use streaming algorithms, memory mapped files and a zero memory copy policy to allow exploration of datasets larger than memory, e.g. out-of-core algorithms. Vaex allows arbitrary (mathematical) transformations using …
A Versatile Framework for Analyzing Galaxy Image Data by Implanting Human-in-the-loop on a Large Vision Model
National Astronomical Observatories, Chinese Academy of Sciences（中国科学院国家天文台）；School of Astronomy and Space Science, University of Chinese Academy of Sciences（中国科学院大学天文与空间科学学院）；College of Electronic Information and Optical Engineering, Taiyuan University of Technology（太原理工大学电子信息与光学工程学院）；School of Computer Science, South China Normal University（华南师范大学计算机学院）
