Enhancement of Daytime Crescent Image Using Wiener Filter Based De-Blurring Technique

Abstract

The aim of this research is to enhance the astronomical image, specifically the crescent image. Millions of Muslims care of monitoring the new crescent to determine the beginning of the fasting month (Ramadan), Eid Al-Fitr(Islamic Eid), and the beginning of the Hijri calendar. These timings are based on the appearance of the crescent in the sky. Because it is very difficult to detect the new crescent during day time, it is important to develop a technique to help detecting it. In the proposed method, the image restoration technique has been used to extract the crescent image from the telescope. The captured image is very noisy due to blurring and weather factors. This noise reduces the quality of the image. In this study, some of crescent images have been captured during the day time, where the crescent age was less than 50 hours before or after the new moon by using a portable digital camera installed on the telescope. Both, the deconvolution and De-Blurring technique with Hanning filter have been used to enhance the crescent's image. This indicates that the Wiener filter is a versatile and powerful technique for de-blurring and noise removal which can be successfully used in taking the crescent's images. In fact, our algorithm's performance was compared with that of the Wiener filter, Blind deconvolution, and Lucy Richardson algorithm. The resulted enhanced images show the crescent's image as can be seen visibly and can be recognized by the human eye. We discovered that and that Wiener filter deconvolution method is better than Lucy Richardson algorithm.