EVALUATION OF AN IMAGE QUALITY ENHANCEMENT TECHNIQUE

摘要

A technique for enhancing the quality of images was investigated. The technique consists of obtaining a video signet from a transparency and adding to this signal its negative second derivative. The technique was investigated by comparing human performance on images produced from the video signal and its derivative with images produced from the video signal alone. Four classes of images, defined by their content, were investigated: tactical, airfields, aircraft and artificial "C" forms. A variety of target detection, classification and identification tasks were studied and performance indices such as completeness, accuracy, and speed were calculated from the basic performance data. The results indicated that the enhancement technique produces both a statistically and a practically significant improvement in performance. The relative magnitude of the improvement increased as the difficulty of the performance task increased. The enhancement technique improves performance principally by increasing the number of correct responses, and to a lesser extent by reducing the number of incorrect responses.