Comparison Of Removing Cosmic-Ray Algorithms From A CCD Image

CCD astronomical images are affected by a variety of noises in their acquition, and the noise of cosmic rays has sometimes direct effect on the useful information of image. It has a law of random distribution, and significantly higher than that of its surrounding pixel gray value. Some of cosmic rays are likely located in a star image, they will change the original information in the star. Cosmic rays result in the offset and the luminosity measurement bias in the center of stars, which will affect certainly in high-precision measurement. Therefore, it is important to eliminate the cosmic-ray for astronomical image analysis and high-precision measurement.This paper focuses on how to effectively identify and eliminate cosmic-ray, which is very important for the information extraction from an astronomical image. And in order to eliminate cosmic-ray in a single image, there are a variety of methods, but no one try to compare the effects of these methods. At present, for the three kinds of algorithms to remove cosmic rays-A Laplacian Edge Detection Algorithm, A Fast Algorithm Based On Histogram, and A Universal Noise Removal Algorithm, we use them to process CCD images taken at Yunnan Observatory with the 1-meter telescope with either actual or simulated cosmic-ray hits, and add noise in the star and the galaxy. Through simulation experiments and the process of true cosmic-ray of the practical images, we compare with these algorithms in effects of removing cosmic rays.Experimental results show that the Laplacian edge detection algorithm can accurately detect cosmic rays in the stars and galaxies best. Because this algorithm is more complex, it is slower than the other two algorithm when they run. At last in this paper we also discuss simply how to accurately substitute those pixels affected by cosmic rays.