Date of Award


Document Type

Master Thesis

Degree Name

Master of Engineering (Research)


Electronic Engineering

First Advisor

Dr. Joe Connell


As the need to process multi-spectral images in real-time is becoming more prevalent, the pressure on Integrated Circuit (IC) manufacturers to produce cheap and reliable devices capable of handling the demands of such operations is growing. This project investigates the ADSP-BF533 Blackfin® Digital Signal Processor (DSP) and its ability to support a real-time multi-spectral imaging system. The Blackfin® DSP, incorporating the Analog Devices/Intel Micro Signal Architecture (MSA), are a broad family of 16-bit fixed-point product with a dual Multiply Accumulate (MAC) core.

Using the MicroC/OS-II Real-time Operating System (RTOS), which is a completely portable and scaleable pre-emptive real-time kernel, the ADSP-BF533’s ability to host an RTOS is investigated. An OV7141 CMOS Image sensor is used to assess the DSP’s ability to acquire multi-spectral data sets and using the VisualDSP++ software development environment, two standard image-processing algorithms are developed so as to benchmark the computational ability of the ADSP-BF533 DSP.


Submitted to the Higher Education and Training Awards Council in partial fulfilment of the requirements of Degree of Master of Engineering.

Access Level