Image Denoising Techniques Using Wavelets
The focus of this work is to develop performance-enhancing algorithm for denoising the signal by using wavelet transformation. The earlier methods used for denoising were based on FFT, where signal is transformed in to frequency domain and soft and hard threshold has been carried out for denoising. After comparing the performances, it has been seen if temporal characteristics of signal can be preserved, it would give better result .Thus, wavelet based denoising came into picture where transformation results in perseverance of frequency and temporal characteristics of the signal. In wavelet based denoising, while applying threshold techniques few signals are also lost. If the lost signal can be retrieved using signal statistical properties, it would give better result in terms of SNR. We tried to recover the lost signal in details part Importance of denoising comes when we talk about images, which play an important role in daily life application. Different techniques have been used for denoising of image, but these lose some of the image characteristics. We modified the existing stochastic algorithm to make it more adaptive. The results for Lena image are presented to establish the advantages that our modified stochastic algorithm provides over other techniques.
To read the full article Download Full Article