VLSI Implementation of Computation Sensors for Visual Information Processing

该建议描述了实时动态任务的VLSI视觉系统的设计和实现。 两部分的职业生涯计划开发智能应用特定和通用视觉计算传感器，可与其他智能系统一起使用，以实现各种机器人视觉和图像处理问题的解决方案。 本文的研究工作分为三个部分：（1）图像采集、边缘检测和运动检测的紧凑型VLSI算法的开发与实现;（2）将这些算法应用于特定问题，例如图像测速、目标跟踪和基本的线跟踪自动导航，以及（3）开发执行图像采集的通用VLSI计算传感器，边缘检测和2D光流测量在一个紧凑的系统。 本提案的教育部分的目标包括将VHDL工具纳入介绍性本科数字设计课程，以及在成像技术和混合数字/模拟信号处理与制造和测试要求的VLSI课程的开发。

This proposal describes the design and implementation of VLSI vision systems for real-time dynamic tasks. The two part career plan develops smart application specific and general purpose visual computational sensors which may be used with other intelligent systems to realize solutions to various robot vision and image processing problems. The three part research tasks are: (1) the development and implementation of compact VLSI algorithms for image acquisition, edge detection and motion detection; (2) applications of these algorithms to specific problems such as image velocitometry, target tracking and rudimentary line-following auto- navigation and (3) development of a general purpose VLSI computational sensor which performs image acquisition, edge detection and 2D optical flow measurement in a single compact system. The goals of the educational component of this proposal include inclusion of VHDL tools into introductory undergraduate digital design courses, and the development of VLSI courses in imaging technology and mixed digital/analog signal processing with fabrication and testing requirements.