Face Detection in Digital Imagery Using Computer Vision and Image Processing

By adding a failure fault limit to an existing implementation of a face detection system application and a tolerance limit for detection time, a desired throughput for detected objects could be established. The aim of this thesis was to add an increased detection rate for pro le/partial faces while increasing the stability and run-time of the system. The results were obtained through an empirical analysis of test data which was compared between the implementation done for this thesis and the older implementation. The results showed an increase in detected faces (in low sized images) by 10% while also increasing the number of false-positives by 0.725 detections per average image. In large size image cases, an automatic scaling functionality was added, to decrease detection time and decrease false-negatives. The results indicated a decrease in average detection time from (old implementation) 15 seconds to 2 seconds, while still increasing positive detection with 23%, from an average of 42% to 65%. False-positives were also decreased from 5.8 to 0.2 detections per average image used in test