The current status of low energy gamma-ray astronomy is reviewed and the conclusion drawn that the next generation of low energy gamma-ray telescopes will require high sensitivity, good timing and spectral resolution. High angular resolution imaging capability is also considered essential. The imaging of low energy gamma-rays is hampered by the difficulties encountered in developing high resolution position sensitive detectors. To this end, an improved Rotation Modulation Collimator (RMC) imaging technique, the phoswich multi-pitch RMC, is discussed. This imaging method employs currently available detector and pulse-share discrimination techniques and is capable of producing images with an angular resolution which is comparable with that of the Coded Aperture Mask (CAM) imaging techniques. The imaging property of this new technique is studied in detail. A comparison of three of the main imaging techniques used in low energy gamma-ray astronomy, CAM, Fourier Transform Collimator and RMC, is carried out in terms of both their theoretical performances and simulated imaging of the Galactic Centre region. A realistic assessment of the possible applications of these three techniques to different types of astronomical observations is also given. The atmospheric gamma-ray radiation has been measured by a number of balloon and satellite borne instruments. A model of this radiation is developed based on the data as measured by a Compton telescope designed by the Max-Planck group. Emission spectra, as predicted by the model, are compared with various observed data, both inside and outside the atmosphere. The sources of background noise in low energy gamma-ray telescopes are discussed. A comprehensive model of the background sources is developed and a generalised simulation program is presented which uses a combination of Monte Carlo methods and empirical calculations. Photon, proton and neutron induced backgrounds are all considered as well as the effects of spacecraft and orbit. The model is applied to some existing telescopes and good agreements are obtained, and interpreted as a demonstration of the model validity. Predictions of background level are made for several future missions. A proposed satellite gamma-ray telescope, GRASP, is briefly described in the last chapter.