Wavelet transform approach for image processing – A research motivation for engineering graduates

Abstract

The study of wavelet transform and its application to image processing are being incorporated in undergraduate and postgraduate syllabuses. This article discuss the basic concepts of image processing and its importance in real time application. The content of this work elaborates the importance of the image denoising and what is the necessity of image denoising in transmitted images in image enhancement topic among the final year project students as well as an emerging research topic among the research scholars worldwide. Hence this article is prepared in order to trigger those students’ knowledge in basic steps involved in image processing techniques in detail. This study highly focuses on how wavelet transform is involved in denoising and impact of denoising in an image. In general, the basic transforms like Fourier transform and discrete cosine transforms are used to transform the image from one domain to another domain. Further it does not require a priori knowledge of noisy image over a transmission time. Hence the proposed method can provide a satisfactory improvement in image enhancement, assured reduction in root mean square error and improvement in the peak signal-to-noise ratio. The peak signal-to-noise ratio value of noisy image to denoised images is improved more in different wavelet transform methods. Thus, this paper can help the students who are working with visual perception of images and machine learning algorithm under image processing environment.