A Proposal to Study the Applicability of Astronomical Image Processing Methods to Low Contrast Digital Images

在过去的几十年里，天文图像处理方法已经发展到一个非常复杂的程度，包括图像重建和恢复（包括无线电干涉测量和像差的哈勃太空望远镜），图像滤波以及目标检测和分类的设施。该项目旨在确定将天文图像处理技术应用于其他领域的低对比度数字图像的可行性。数字化乳房x光照片在许多方面与漫射天体的图像相似。分析乳房x光照片的挑战在于识别被称为微钙化的微弱点状物体；天文学上的类比是暗淡的恒星，或者可能是宇宙射线击中CCD探测器，在一个扩展的星云区域。一些关于乳房x光检查的初步实验令人鼓舞，该项目将进行更系统的可行性研究。* * *

ABSTRACT AST-9520463 Salzberg, S. L. Astronomical image processing methods have been developed in the past several decades to a point of great sophistication, including facilities for image reconstruction and restoration (both for radio interferometry and the aberrated Hubble Space Telescope), image filtering, and object detection and classification. This project is aimed at determining the feasibility of applying astronomical image processing techniques to low contrast digital images from other fields. Digitized mammograms are similar in many ways to images of diffuse astronomical objects. The challenge in analyzing mammograms is to identify clusters of faint, point-like objects called microcalcifications; the astronomical analog would be faint stars, or perhaps cosmic ray hits on a CCD detector, in a region of extended nebulosity. Some initial experiments on mammograms are encouraging and this project will conduct a more systematic feasibility study. ***