A digital image processing tool for characterizing dendritic trunks

The formation of tree-like dendritic structures is ubiquitous in many natural and industrial processes. These multi-scale structures manifest complex dynamics, and it is often challenging even to characterize rudimentary features such as dendritic trunks that can yield valuable insights into the underlying physical phenomena. Attempts to understand the evolution of trunks typically involve the use of fast Fourier transforms (FFTs). While FFTs can help resolve spatial structures, they are incapable of distinguishing between dendritic trunks and the omnipresent side branches. In the current work, we present a simplified image-processing procedure that focuses on isolating and estimating the attributes of dendritic trunks. A novel combination of techniques, including grayscale thresholding, Savitzky–Golay filtration, and curvature estimation, has been utilized to accurately evaluate the average height and the dominant wavelength of dendritic trunks. The approach also helps ascertain the transient evolution of dominant modes of the dendritic manifestation. The developed technique has been extensively tested on dendrites that evolve during the electrodeposition process, although the approach is general and can be extended to various other applications.