VLSI array architectures for navigation and human recognition in vision-guided embedded systems

• Main Author: Mohan Aruppukottai Rajarathinam
• Other Authors: Sudha Natarajan
• Format: Theses and Dissertations
• Language: English
• Published: 2011
• Subjects: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
• Online Access: https://hdl.handle.net/10356/43563
• Institution: Nanyang Technological University

This thesis focuses on hardware solutions for vision-based navigation and human recognition. The application scenarios include mobile robotics, human-robot interaction, assistive technology for visually-challenged and video surveillance. Hardware accelerators are necessary for the navigation of robots and guiding systems for visually-challenged in a dynamic environment. They are also necessary for real-time identification of a human by a service robot or a visually-challenged for interaction. Single chip solutions with focus on reconfiguration, low area, and high speed are desirable for such applications. This thesis reports hardware-directed algorithms and array-type Very Large Scale Integrated (VLSI) architectures for navigation and face-based human recognition.