FPGA-based image change detection framework

This research presents a new approach in modeling and implementing a framework for linear

image change detection using Field Programmable Gate Arrays (FPGA) based implementations.

Detecting changes between digital images of the same location or region of interest have attracted

huge interest in research due to its wide range of applications such as medical diagnosis,

astronomy and forensics. In recent years, the steady increase in the use of hardware

implementation in image processing, especially the FPGA architecture that has proven to be

superior in performance and cost-efficiency when compared to the conventional computer

systems. The problem addressed by this research is the lack of an FPGA based generic and

standardized change detection framework that executes image change detection at a low

computational cost, using low processing resources like memory and power consumption.

The framework proposed in this research work is focused on the image differencing, which is an

algebra change detection technique. Image edge detection, image enhancement and restoration

methods were employed in the framework implementations using MATLA/Simulink and the Xilinx

system generator. The success of the framework implementation proved that the objectives of

the thesis were achieved and has been verified by the experimental results. While the

implementation was successful in producing expected results, there is a need for more

experiments to explore different filtering algorithms that could improve the results. Statistical

forms of data interpretation such as Machine Learning algorithms could improve complex image

difference results analysis