具有泊松噪声分布的图象复原方法

本文以ＣＣＤ图象为技术背景，指出ＣＣＤ图象噪声主要将以与信号有关的泊松噪声形式存在。根据图象形成的模型及图象统计特性，提出了最大似然法，最大后验概率法及最大复原方法，针对ＭＡＰ法提出了具体算法和参数估计方法，并提出应用图象分割法来提高计算机复原速度，节省存储空间。     

具有泊松噪声分布的图象复原方法

本文以ＣＣＤ图象为技术背景，指出ＣＣＤ图象噪声主要将以与信号有关的泊松噪声形式存在。根据图象形成的模型及图象统计特性，提出了最大似然（ＭＩ）法，最大后验概率（ＭＡＰ）法及最大熵（ＭＥ）复原方法。针对ＭＡＰ法提出了具体算法和参数估计方法。并提出应用图象分割法来提高计算机复原速度，节省存储空间。     

图象的模型及其统计特性

本文对图象的统计特性及其适用模型作了分析。讨论了以前常用的平稳图象模型的缺陷,对实际图象的统计特性作了分析,认为它是非平稳的,不满足各态历经性,同时在空间上是高度相关的;讨论了相应的随机参数统计模型和描述性统计模型。指出作用于整幅图象的“全局”图象复原算法比每次只独立计算单个象元的“点”图象复原算法更为优越。     

海尔-波普彗星和α Gem星图象检测结果

利用自行研制的天文视频图象采集系统,对１９９７年５月１日在武定狮山用ＰａｎａｓｏｎｉｓＮＶ－Ｓ８８０ＥＮ型摄像机拍摄的海尔－波普彗星和双子座αＧｅｍ星进行了图象积分处理,并给出检测结果。结果表明：１）经多帧积分处理的海尔－波普彗星图象比单帧图象细节明显的清晰。２）单帧αＧｅｍ星周围淹没在噪声信号中看不到的暗星经多帧叠加后可以“看”到了。     

天文CCD片内放大器的噪声特性及其抑制方法

分析了低光度天文应用时CCD器件的主要噪声源的噪声特性和三种典型CCD读出信号处理方式的噪声抑制效果;提出了天文CCD读出信号的相关多次采样(CMS)处理新方法,并分析了四次采样抑制CCD读出噪声的实验结果。     

图象质量的测度

本文论述了图象质量的各种定量测度,包括连续的和离散的形式。着重讨论了基于视觉特性的图象质量测度,它们与主观评价具有很好的相关性;提出了用图象信息量作图象质量测度的方法。这些测度在图象压缩、图文传真中具有重要的应用。图象质量的测度是对图象质量的评价,它对复原算法优劣的评价是一个十分重要的依据。     

对弱信号CCD图像噪声的最佳化处理

本文在分析CCD读出图像噪声的统计特性和目前CCD图像预处理的基础上,提出了对CCD读出数据的最佳化处理方法。该方法对CCD读出噪声和其它随机噪声都有较好的抑制效果,使用该方法可以从迭加有随机噪声的数据中得到信号的最佳估值,提高弱信号CCD读出图像的信噪比。     

图象质量的测度

本文论述了图象质量的各种定量测度、包括连续和离散的形式。着重讨论了基于视觉特性的图象质量测度，它们与主观评价具有很好的相关性；提出了用图象信息量作图象质量测试的方法。这些测试在图象压缩、图文传真中具有重要的应用。图象质量的测试是对图象质量的评价，它对复原算法优劣的评价是一个十分重要的依据。     

边带分离型超导SIS接收机本振耦合噪声研究

超导SIS (Superconductor-Insulator-Superconductor)接收机因极低的接收机噪声温度成为毫米波和亚毫米波段射电天文观测的首选.本振系统耦合噪声也是接收机噪声的一部分,在多年的天文观测中,发现本振耦合噪声无法完全忽略,对天文观测的灵敏度有一定影响.采用两个不同种类的信号发生器作为本振系统初级信号源,测试了超导SIS接收机的噪声温度,发现信号发生器输出的基底噪声能够耦合到接收机内部,从而增加接收机噪声强度.分析研究了本振系统热噪声和信号发生器基底噪声对接收机噪声的影响.通过在信号发生器输入端加入窄带滤波器滤除其基底噪声,消除了信号发生器基底噪声引入的接收机噪声,降低了接收机的整体噪声,提高了望远镜的灵敏度.     

用状态空间模型递推模拟振荡器噪声

对振荡器的五种噪声之间的关系作了介绍,并对相位时间信号的状态空间模型的基本理论作了描述.在此基础上导出了五种噪声的递推模拟方法.最后模拟了这五种噪声并分析了它们的功率谱.结果表明,功率谱符合这几种噪声的定义.     

真实大气望远镜条件下斑点掩模法像复原模拟实验

我们于1989—1990年对一种有效的天文高分辨率像复原方法—斑点掩模法进行了真实大气、望远镜条件下的模拟实验,得到了较好的结果。实验结果表明,取样期间大气统计特性不变和小方差条件的重要性。为满足这一条件,合适的探测器是必须的。     

Astronomical Image Co-adding Method based on Wide-field Image Deconvolution

Due to the atmospheric turbulence, the static aberration, tracking and pointing errors of telescopes, the point spread functions (PSFs) in different fields of view are different. Meanwhile, there are different PSFs in the images obtained by different telescopes. The quality of co-adding image is limited by the image with the poorest quality, and finally the resolution and sensitivity of the quad-channel telescope will also be affected. Dividing the image into some regions with the same type of PSF, and deconvolving these regions can improve the quality of the co-adding image. According to this theory, an image restoration algorithm based on the PSF clustering is proposed. Firstly, this paper makes the PSF clustering analysis by using Self-Organizing Maps, and makes the image segmentation based on the result of the PSF clustering analysis, then using the clustered PSFs to make deconvolutions on the sub-images. Then, the restored sub-images after deconvolution are joined together. Finally, by through the image registration and co-adding, the image with a high signal to noise ratio can be obtained. The result shows that the signal to noise ratio of the astronomical images are improved with our method, and the detection capability on faint stars is also improved.     

The spatial distribution of polar hydrogen deposits on the Moon

A new analysis of the Lunar Prospector epithermal neutron data is presented, providing an improved map of the distribution of hydrogen near to the lunar poles. This is achieved using a specially developed pixon image reconstruction algorithm to deconvolve the instrumental response of the Lunar Prospector's neutron spectrometer from the observed data, while simultaneously suppressing the statistical noise. The results show that these data alone require the hydrogen to be concentrated into the cold traps at up to 1 wt% water-equivalent hydrogen. This combination of localisation and high concentration suggests that the hydrogen is present either in the form of a volatile compound or as solar wind protons implanted into small regolith grains.     

Segmentation of Loops from Coronal EUV Images

We present a procedure to extract bright loop features from solar EUV images. In terms of image intensities, these features are elongated ridge-like intensity maxima. To discriminate the maxima, we need information about the spatial derivatives of the image intensity. Commonly, the derivative estimates are strongly affected by image noise. We therefore use a regularized estimation of the derivative, which is then used to interpolate a discrete vector field of ridge points; these “ridgels” are positioned on the ridge center and have the intrinsic orientation of the local ridge direction. A scheme is proposed to connect ridgels to smooth, spline-represented curves that fit the observed loops. Finally, a half-automated user interface allows one to merge or split curves or eliminate or select loop fits obtained from this procedure. In this paper we apply our tool to one of the first EUV images observed by the SECCHI instrument onboard the recently launched STEREO spacecraft. We compare the extracted loops with projected field lines computed from near-simultaneous magnetograms measured by the SOHO/MDI Doppler imager. The field lines were calculated by using a linear force-free field model. This comparison allows one to verify faint and spurious loop connections produced by our segmentation tool and it also helps to prove the quality of the magnetic-field model where well-identified loop structures comply with field-line projections. We also discuss further potential applications of our tool such as loop oscillations and stereoscopy.     

A realistic model for point-sources imaged on array detectors: The model and initial results

We have constructed a computer model for simulation of point-sources imaged on two-dimensional detectors. An attempt has been made to ensure that the model produces data that mimic real data taken with 2-D detectors. To be realistic, such simulations must include randomly generated noise of the appropriate type from all sources (e.g. source, background, and detector). The model is generic and accepts input values for parameters such as pixel size, read noise, source magnitude, and sky brightness. Point-source profiles are then generated with noise and detector characteristics added via our model. The synthetic data are output as simple integrations (onedimensional), as radial slices (two-dimensional), and as intensity-contour plots (three-dimensional). Each noise source can be turned on or off so that they can be studied separately as well as in combination to yield a realistic view of an image. This paper presents the basic properties of the model and some examples of how it can be used to simulate the effects of changing image position, image scale, signal strength, noise characteristics, and data reduction procedures.Use of the model has allowed us to confirm and quantify three points: 1) The use of traditionalsize apertures for photometry of faint point-sources adds substantial noise to the measurement which can significantly degrade the quality of the observation; 2) The number of pixels used to estimate the background is important and must be considered when estimating errors; and 3) The CCD equation normally used by the astronomical community consistently overestimates the signal-to-noise obtainable by a measurement while a revised equation, discussed here, provides a better estimator.     

用CCD照相多幅叠加处理检测弱星的试验

本文利用CCD动态范围大且便于数值处理的优点,观测月晕中(强背景上)的弱源。在背景尚未使CCD片子饱和时,即停止露光,存入计算机,多次露光,然后在计算机内多幅叠加,减去背景,显出弱源。通过观测试验和定量估算,结果表明,曝光重叠法,用10幅图叠加时能提高信噪比2.65倍。实测与理想情况符合较好。     

上海天文台1．56米望远镜斑点干涉成像实验进展

斑点干涉成像技术是克服大气湍流影响,提高地面大口径望远镜分辨本领的有效途径之一。该技术利用斑点相机拍摄一系列的短曝光像,使得大气湍流冻结,再经过图像处理获得高分辨率重建像。该技术设备简单,易于实现,很快在观测天文学中得到了广泛的应用,尤其是对双星的研究。首先回顾了天文高分辨率重建技术的发展,并介绍了相关研究成果。描述了几种典型的斑点干涉成像处理方法及其优缺点。对图像噪声类型及滤波方法进行了分析。在上海天文台1．56m望远镜上开展了双星斑点干涉观测实验,目标星等4～7mag,双星目标星等差小于2。分别采用斑点干涉术和迭代位移叠加法成功实现了双星目标的高分辨率成像,初步证明了在1．56m望远镜上进行斑点干涉成像实验,能够达到接近望远镜衍射极限的分辨率水平。