Real-time embedded eye detection system

The detection of a person’s eyes is basic task in applications as important iris recognition biometric identification or fatigue driving assistance systems. Current commercial and research systems use software frameworks that require dedicated computer, whose power consumption, size price are significantly large. This paper presents hardware-based embedded solution for eye real-time. From an algorithmic point-of-view, the popular Viola–Jones approach has been redesigned to enable highly parallel, single-pass image-processing implementation. Synthesized implemented All-Programmable System-on-Chip (AP SoC), this proposal allows us process more than 88 frames per second (fps), taking classifier less 2 ms image. Experimental validation successfully addressed system works with walking subjects. In case, prototype module includes CMOS digital imaging sensor providing 16 Mpixels images, it outputs stream detected 640 × 480 images. Experiments determining accuracy proposed terms performed CASIA-Iris-distance V4 database. Significantly, they show 100%. • A redesign proposed. core can 750 fps. An rate 100% obtained using integrated into framework on motion.