Digital image filter design using evolvable hardware

In this paper, a new image processing architecture through complete evolvable hardware has been described that will be able to adapt according to the particular images encountered. This will serve as an alternative to conventional approaches where the structure is irreversibly fixed in the design process. The simulation of such a system through the use of evolutionary algorithm is demonstrated for the conventional image processing operation of image filtering. Bitmap images of size 128×128, contaminated by Gaussian noise with mean 0 and different level of variance were taken and filtering operation is performed. This evolvable digital image filter was compared with conventional image filters in terms of mean difference per pixel. The result shows that the evolvable hardware digital image filter outperforms traditional approaches.