Recently, a massive astronomical dataset is being collected to find answers for a variety of unanswered questions about our universe by virtue of modern sky survey instruments. Unfortunately, it is impossible to work on these massive datasets manually to get effective results so, astronomers are seeking approaches to automate the human error borne processes of manual scanning in order to discover astronomical knowledge and information from these large raw datasets i.e. to classify stars, quasars, galaxies and Supernovae (SNe). The problem here, this is done by hand and it is a very time consuming job as well as it is subject to human bias which differs from person to person. In addition, the manual scanning is infeasible for a huge amount of images. From this point of view, I've selected this concrete astronomical classification problem to investigate applying convolutional Neural Networks (CNNs) algorithm to automate this process and then I compared my results to a reference publication as a benchmark model by using the same well-known public dataset of the Sloan Digital Sky Survey (SDSS).