Analog VLSI and neural systems
Analog VLSI and neural systems
Analog VLSI circuits for stimulus localization and centroid computation
International Journal of Computer Vision - Special issue: VLSI for computer vision
Communicating neuronal ensembles between neuromorphic chips
Neuromorphic systems engineering
A general purpose image processing chip orientation detection
NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
VLSI implementation of motion centroid localization for autonomous navigation
Proceedings of the 1998 conference on Advances in neural information processing systems II
Computation of smooth optical flow in a feedback connected analog network
Proceedings of the 1998 conference on Advances in neural information processing systems II
Robot Vision
Vision Chips: Implementing Vision Algorithms with Analog VLSI Circuits
Vision Chips: Implementing Vision Algorithms with Analog VLSI Circuits
Analog VLSI Circuits for Covert Attentional Shifts
MICRONEURO '96 Proceedings of the 5th International Conference on Microelectronics for Neural Networks and Fuzzy Systems
Motion-Driven Segmentation by Competitive Neural Processing
Neural Processing Letters
| Hi-index | 0.00 |
Traditional approaches for solving real-world problemsusing computer vision have depended heavily on CCD cameras andworkstations. As the computation power of workstations doubles every1.5 years, they are now better able to handle the large amount ofdata presented by the cameras; yet real-time solutions for physicalinteraction with the real-world continues to be very hard, andrelegated to large and expensive systems. Our approach attempts tosolve this problem by using computational sensors andsmall/inexpensive embedded processors. The computational sensors arecustom designed to reduce the amount of data collected, to extractonly relevant information and to present this information to thesimple processor, microcontrollers (μCs) or DSPs, in a formatwhich reduces post-processing latency. Consequently, thepost-processors are required to perform only high level computationon features rather than data. These systems are applied to problemssuch as target acquisition and tracking for image stabilization andautonomous data driven autonavigation for mobile robots. We presentan example of a system that uses a pair of computational sensors anda μC to solve a toy autonavigation problem.The computational sensors, however, have wide applications in manyproblems that require image preprocessing such as edge detection,motion detection, centroid localization and other spatiotemporalprocessing. This paper also presents a general-purpose computationalsensor capable of extracting many visual information components atthe focal plane.