Published Paper Details:

The aim of image enhancement is to improve the interpretability or perception of information in images for human viewers, or to provide `better' input for other automated image processing techniques such as biometric based techniques that include face recognition, fingerprint matching, early detection of biological disorder like cancer etc. Image enhancement techniques can be divided into two broad categories: spatial domain techniques, which operate directly on pixels, and frequency domain techniques, which operate on the Fourier transform of an image. Frequency domain image enhancement techniques are based on modifying the Fourier transform of an image. Smoothing domain filters, sharpening domain filters, Homomorphic filters are few of them which are used for image enhancement. In the present work such filters have been designed and implemented. A comparison has been made over aforesaid frequency domain image enhancement techniques using various filters and their performances have been evaluated in terms of PSNR (peak signal to noise ratio), CNR (contrast to noise ratio), mean, variance and invariant moments. For better visual interpretation of a colour image a special technique colour image enhancement has also been successfully implemented. The results of present research work have been applied to one of the forensic science applications i.e. fingerprint matching. It has been observed that matching accuracy is improved if fingerprint image is enhanced in frequency domain before subjecting it to the matching process.