A fast color tracking system with automatic exposure control

Abstract

In this paper, we present a high frame rate (HFR) color object tracking system that can automatically control its exposure time by executing brightness histogram-based image processing in real time at a high frame rate. Our aim is to track high-speed color objects even under dynamically changing illumination, such as lamps flickering at 100 Hz, corresponding to an AC power supply at 50 / 60 Hz. Our vision system, which is implemented as a brightness histogram calculation circuit module in parallel hardware logic on an FPGA-based high-speed vision platform, can simultaneously calculate a 256-bin brightness histogram for an 8-bit color image with resolution 512 × 512 px at 2000 fps. On the basis of the HFR brightness histogram calculation, our proposed method realizes automatic exposure (AE) control of 512 × 512 images at 2000 fps using our proposed AE algorithm, which can maximize the number of pixels in the effective range of the brightness histogram, thus excluding pixels that are much darker and much brighter, to improve the dynamic range of the captured image without over- and under-exposure. The effectiveness of our proposed HFR system with AE control is evaluated experimentally for a static scene and for moving objects in a nonuniform illumination environment.