Three-dimensional surface information of material surfaces using focused ion or photon induction methods is a significant challenge in materials science which have application in various fields. Ionizing radiations, such as protons, alpha particles, and beta electrons, with their interactions in matter, lead to the production of characteristic X-rays, which are particularly important for analyzing the elemental composition of materials. Surface roughness greatly impacts the accuracy of data analyses, so its reconstruction is very important. In this study, using the multi-segment detectors in mirco-PIXE analysis the surface roughness of an ancient Roman coin from the fourth century AD has been reconstructed. Particle-induced X-ray emission (PIXE) and X-ray fluorescence analysis are employed to determine the elemental composition of the coin. Then, using a four-segment silicon drift detectors (SDD), a set of dual images obtained by the orthogonal parts of the detector was considered to extract the gradient components of the roughness. This was followed by numerical integration of the gradient components in two dimensions to reconstruct the surface topography. The result surface roughness with increased precision by reducing shadow effects and enhancing resolution. The results of this study in reconstructing the roughness of the surface based on its constituent elements demonstrate that these findings could play a vital role in archaeological studies and the analysis of historical objects, allowing researchers to examine the physical and chemical characteristics of ancient works with greater accuracy.