Approach to reduce the computational image processing requirements for a computer vision system using sensor preprocessing and the Hotelling transform

摘要

We describe a new development approach to computer vision for a compact,low-power, real-time system such as mobile robots. We take advantage of preprocessing in a biomimetic vision sensor and employ a computational strategy using subspace methods and the Hotelling transform in an effort to reduce the computational imaging load. The combination, while providing an overall reduction in the computational imaging requirements, is not optimized to each other and requires additional investigation.