Fast Automatic Target Detection System Based on a Visible Image Sensor and Ripple Algorithm

摘要

In this paper, we present a novel embedded-target detection system built upon a visible image sensor. This system can detect both static and dynamic targets under difficult situations. It is small in size, light in weight, has low power consumption, and can be embedded into the system of robotics and intelligent systems. In this system, the target images are captured by a visible image sensor and then processed by a digital signal processor based on the ripple algorithm proposed in this paper. The ripple algorithm uses several templates to represent the target in different view angles and sizes. Then, it extracts the features of objects by a group of concentric circles, and compares the features to those of the templates. These features are invariant to image rotations or target translations and scaling to overcome the robotics moving characteristics of swing and sway. In addition, the algorithm has low computational complexity, and can fulfill the requirement of real-time applications. From this experiment, it is shown that the developed automatic target detection system is effective to detect static or dynamic targets in the real-time application of robotics, and the ripple algorithm is correct and effective.