Proper installation is the key to the effectiveness of the system. The video cameras customarily used for surveillance application have low image quality, such that there is often not sufficient detail in the images to identify people and objects. This report assessed five approaches that might correct this problem. Only two of the approaches were judged to have a high likelihood of success. These were the use of a few digital still cameras and the proper installation of all cameras from a photographic perspective. The most effective system involved replacing a few cameras that have a very wide angle of view with a larger number of cameras, some with wide angles of view and others with specific target locations. Since this increases the total amount of information that would be collected if all of the cameras operate and record all of the time, this report recommends that cameras not record all of their information all of the time. This requires an "intelligent" system similar to the ones tested during this project. Three ways that the recording can be managed with these systems are described in this report. Appended papers on measuring the use of safety technology in American Schools, camera testing for school safety, and audio surveillance discriminator