Application of Daubechies Wavelets in Digital Processing of Biomedical Signals and Images

Abstract

Wavelet analysis of one-dimensional signals has proven effective in deciphering the electrocardiogram (ECG). Promising results have already been obtained from their analysis. In particular, it has been shown that anomalous effects in the ECG are mainly manifested on much larger scales (low frequencies), while normal structures are characterized by relatively small scales (high frequencies). Denoising is one of the urgent problems of digital processing of biomedical signals and tomographic images. Wavelet methods are relatively new and are a method of denoising using wavelet functions. Wavelets allow for the analysis of various types of signals and effective noise removal, so it is of particular interest to study their potential to improve image quality. It is very convenient to use DWT (Discrete Wavelet Transform) in digital image processing, because it provides deep insight into the main spatial and frequency features. Wavelets provide excellent time-frequency localization, meaning they can capture both transient and stationary features in signals and images. This localization capability is especially valuable in medical applications where signals may contain abrupt changes or irregular patterns. In this paper, we will discuss the method of biomedical signal restoration and denoising in tomographic images using different wavelet functions such as Haar wavelet, Symlet, Meyer wavelet, Daubechies wavelet. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.