This dissertation focuses on increasing the ability to detect space objects and increase Space Domain Awareness (SDA) with space surveillance sensors through image processing and optical theory. SDA observations are collected through ground-based radar and optical systems as well as space based assets. This research focuses on a ground-based optical telescope system, the Space Surveillance Telescope (SST). By increasing the number of detectable Resident Space Objects (RSOs) through image processing, SDA capabilities can be expanded. This is accomplished through addressing two main degrading factors present in typical SDA sensors; spatial undersampling in the collected data and noise models and assumptions used in current algorithms. The assigned cost and a priori probabilities of a Bayes Multiple Hypothesis Test (MHT) are investigated in this dissertation to address the spatial undersampling. New algorithms are developed and tested, and demonstrated improved detection capabilities at operationally realistic false alarm rates. Additionally, a new noise model is developed which more accurately represents the received noise present in data collected with surveillance telescopes under certain atmospheric conditions. These algorithm have demonstrated probability of detection improvement of up to 80 percent in collected SST data over the currently employed detection techniques.