Error compensation and hardware reduction of fixed point 2-D Gaussian filter

Gaussian filter is used as an efficient preprocessing method in wide ranges of image processing applications to reduce the effect of unwanted and destructive pixels of image. An efficient implementation of Gaussian filter on hardware is important especially in the real time applications. In this paper fixed point representation of Gaussian filter is considered and optimization techniques in its hardware implementation are generalized. In the proposed method the hardware cost and accuracy are in tradeoff with other and can be modified in a good manner. The simulation results show that in the case of un-sharp masking with Gaussian kernel, suitable result in term of hardware complexity and accuracy is obtained.