Smart vision in system-on-chip applications

Wells, Cade Cenric (2005) Smart vision in system-on-chip applications. EngD thesis, University of Glasgow.

Full text available as:
[thumbnail of 10754010.pdf] PDF
Download (10MB)

Abstract

In the last decade the ability to design and manufacture integrated circuits with higher transistor densities has led to the integration of complete systems on a single silicon die. These are commonly referred to as System-on-Chip (SoC). As SoCs processes can incorporate multiple technologies it is now feasible to produce single chip camera systems with embedded image processing, known as Imager-on-Chips (IoC). The development of IoCs is complicated due to the mixture of digital and analog components and the high cost of prototyping these designs using silicon processes. There are currently no re-usable prototyping platforms that specifically address the needs of IoC development. This thesis details a new prototyping platform specifically for use in the development of low-cost mass-market IoC applications. FPGA technology was utilised to implement a frame-based processing architecture suitable for supporting a range of real-time imaging and machine vision applications. To demonstrate the effectiveness of the prototyping platform, an example object counting and highlighting application was developed and functionally verified in real-time. A high-level IoC cost model was formulated to calculate the cost of manufacturing prototyped applications as a single IoC. This highlighted the requirement for careful analysis of optical issues, embedded imager array size and the silicon process used to ensure the desired IoC unit cost was achieved. A modified version of the FPGA architecture, which would result in improving the DSP performance, is also proposed.

Item Type: Thesis (EngD)
Qualification Level: Doctoral
Additional Information: Adviser: David Renshaw
Keywords: Electrical engineering
Date of Award: 2005
Depositing User: Enlighten Team
Unique ID: glathesis:2005-74098
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 23 Sep 2019 15:33
Last Modified: 23 Sep 2019 15:33
URI: https://theses.gla.ac.uk/id/eprint/74098

Actions (login required)

View Item View Item

Downloads

Downloads per month over past year