Image Filtering on .NET-based Desktop Grids

Image filtering is the use of computer graphics algorithms to enhance the quality of digital images or to extract information about their content. However rendering very large size digital images on a single machine is a performance bottleneck. To address this we propose parallelising this application on a desktop Grid environment. For parallelizing this application we use the Alchemi Desktop Grid environment and the resulting framework is referred to as ImageGrid. ImageGrid allows the parallel execution of linear digital filters algorithms on images. We observed acceptable speed up as a result of parallelising filtering operation through ImageGrid. We run the tests on different data sets by varying the dimension of the images and the complexity of the filters. Results demonstrate potential of grid computing for desktop applications and that the speed up obtained is more consistent for large images and complex filters.