This paper presents a hybrid technique for image enhancement with ability of de-noising it integrates two different processing aspects into one. The proposed algorithm uses the image abstraction technique for detecting the information density in different parts of image then accordingly operates the smoothing filter and after filtering the information's of edges are recombined with the filtered image. The proposed technique also utilizes the Neural Network for filtering noise generated edge patterns. Hence the approach not only enhances the image but also avoids the enhancement of noise. The simulation of algorithm shows that it improves the perception; remove noise while maintaining the structure information intact it is also found that the proposed technique is quite fast.
Image enhancement, Neural Network, empirical analysis
