|
Harrison, Reid A Robust Analog VLSI Reichardt Motion Detector Silicon imagers with integrated motion-detection circuitry have been developed and tested for the past 15 years. While much progress has been made, the performance of most of these sensors has been evaluated only with noiseless stimuli. For integrated motion sensors to become practical for real-world applications, they must exhibit robustness in low signal-to-noise ratio (SNR) environments. Many previous circuits estimate motion by identifying and tracking spatial or temporal features. These approaches are prone to failure at low SNR conditions, where feature detection becomes unreliable. An alternate approach to motion detection is an intensity-based spatiotemporal correlation algorithm, such as the one proposed by Hassenstein and Reichardt in 1956 to explain aspects of insect vision. This fully analog approach, widely used in biological sensory systems, does not identify explicit image features, and is robust in noisy environments. We implemented a Reichardt motion sensor with integrated photodetectors in a standard CMOS process. Our circuit operates at sub-microwatt power levels, and is suitable for constructing 2-D VLSI arrays. Traditional correlation-based sensors suffer from strong contrast and spatial frequency dependence. We investigate circuit architectures that lessen these dependencies. We also discuss and measure the effects of cross-chip device mismatch on these parallel, analog circuits. Finally, we evaluate the robustness of the sensor in the presence of spatial and temporal noise.
|