Image and Electron Microscopy Processing in ImageJ Software Involving Fossils with Compositional Variation

In fossilization through authigenic mineralization, the chemical compounds present in the fossilized material provide insights into the taphonomic processes to which the organism was subjected. Various methods of compositional analysis are used for this purpose. Thickness calculations of structures with different compositions are sometimes necessary for taxonomic or taphonomic objectives. In the case of the samples used in this study, FIB-STEM, SEM, EDS, SR-micro XRF, and conventional photography were employed to characterize the composition of fossil insects from the Crato Formation (Cretaceous). The analysis of images containing areas with different compositions enables the study of elemental distribution within a fossilized body, thus allowing the determination of anatomical structure thicknesses and facilitating the understanding of the mineralization sequence. The initial processing of samples for identifying thicknesses is carried out manually and visually using the ImageJ software. It is possible to select each interval of points and desired sections, calculating their areas, but at the cost of extensive labor, as image complexity (compositional heterogeneity) increases the measurement time, posing a challenge to be addressed.To meet this demand, we present a method for automating the use of the software with external scripts, which shortens the time required to identify structures with different compositions in a photograph or micrograph (using backscattered electrons validated by an EDS map), thus enabling thickness calculations. The implemented automation consists of combining Python and ImageJ Macro Language (IJM) scripts to map the image, selecting changes in color intensity with their associated coordinates, based on a variable validation parameter. From the selected points, it is possible to obtain a visual reconstruction for validation purposes, generate graphs of composition variation as a function of sample thickness, and calculate the distribution of different compositions. In two samples (FIB-STEM-EDS and conventional photographs with SR-micro XRF maps), it was possible to perform mapping, identification, and thickness calculation of compositionally varied structures. The tool proved effective in the visual reproduction of the samples and in the quantification of areas with distinct compositions, validating its application in studies of fossilization.
[FAPESP 2025/02782-3, 2023/14250-0, 2023/04501-6, 2022/06485-5]