Although limited alternatives have been proposed to describe soils by using proximal sensors under natural conditions, efforts must be made to apply the newest technologies to assist in soil characterization and mapping. Our main objective was: to create a protocol incorporating technology that assists pedologists: (i) to compare soil horizonation by using visible, near, and shortwave infrared (Vis-NIR-SWIR) spectroscopy, mid-infrared spectroscopy (MIR), portable X-ray fluorescence (pXRF) indices, and conventional field observations; (ii) to identify mineralogical and physico-chemical variations between soil horizons by using proximal sensors, with inferences about pedogenic processes; and (iii) to expand punctual information to spatial dimensions with RGB images and terrain attributes. Seven soil profiles located through a toposequence evaluation were analyzed on a farm in the São Paulo State, Brazil. First, when the area consisted of exposed soils, images were obtained by Landsat and an unmanned aerial vehicle (UAV), and a digital surface model (DSM) was then built by stereoscopy, which was subsequently used to calculate thirteen terrain attributes. Each soil profile was evaluated with devices directly in the field (except for the MIR analyses, for which soil samples were brought to the laboratory and dried). Several indices were calculated from the pXRF data during the profile analysis, while principal component analyses (PCA) were performed to reduce the dimensions of the spectral data. Four datasets for each soil profile (VIS-NIR-SWIR, pXRF, MIR, and a combination of all the data) were submitted for k-means clustering analysis and the similarities among the layers were evaluated. Following the profile evaluation, spatialization was performed based on single bands from the RGB image and PCA using terrain attributes. The data were sampled to a punctual view of 370 points, classified, and then interpolated with the spline method. The Vis-NIR-SWIR spectral range, together with soil color, provided better correspondence with conventional field observations. We observed that the combined use of Vis-NIR-SWIR, MIR, and pXRF helped to identify and describe the soil profiles appropriately. Proximal sensors also allowed us to make more robust inferences regarding pedogenesis. We delineated soil information profiles in soil units by using the spatial dimension's terrain attributes and, RGB images from a UAV.