2D Discrete Wavelet Transformation (2D-DWT) for Nanoscale Morphological Analysis

Abstract

Digital image processing nowadays is widely used in various applications in micro to macro scale such as Nano-structure for medical, defense, natural resource management, security purposes. This article reviewed the 2-Dimensional Discrete Wavelet Transformation (2D-DWT) for morphological analysis of Scanning Electron Microscope (SEM) image of Ni-P-CNF nanocomposite plated on mild steel substrate (grade AISI1040) for analyzing the multi-level decomposition, denoising and compression test. It was concluded that the 2D-DWT method is more efficient and precise as compared to the conventional methods like Power Spectral Density (PSD) and histogram equivalence. These methods are limited by the Heisenberg uncertainty principle, whereas, the wavelet theorem provides a multi-resolution analysis. The wavelet function can capture localized characteristics and transients in the data since it is often localized in both time and frequency. These features make wavelets ideal for storing transient and steady-state components of a signal or image, allowing them to simultaneously offer excellent time and frequency localization. SEM images usually contain huge information which can lead to computational complexities. 2D-DWT is a very effective tool to de-noise the image. In order to test its efficiency, we have intentionally added some noise in the image and de-noise it. Also, we have compressed the image at different at different levels. This study provides the utility of the 2D-DWT for image processing as well as compared with other approaches for image decomposition, denoising and image compression.