Implementing Planar Motion Tracking Algorithms on CMOS+FPGA Vision System

This paper describes a 1,000 Hz visual feedback using the CMOS+FPGA vision. It is required to obtain positional and angular signals around 1,000 Hz to control a mechanical system. A vision sensor must obtain visual features of a target object, synchronizing its sampling rate to the sampling rate of the control. Thus, we need 1) image capturing over 1,000 Hz with high resolution, 2) visual feature computation at the capturing rate, and 3) visual feature transmission to a control system with little delay. We propose the CMOS+FPGA vision system to realize 1,000 Hz visual feedback. This system consists of a CMOS image detector to capture images at 1,000 Hz and an FPGA to compute image features at this sampling rate. We implement two vision algorithms to track planar motion on the CMOS+FPGA vision system. First we introduce the CMOS+FPGA vision system. Second, we show the implementation of the computation of the image gravity center. Then, we show the implementation of matched filter