This paper introduces an algorithm for automating the analysis of petrographic thin-section images of sandstones and siltstones. The images of thin sections are obtained in polarized light at magnifications providing good image quality. In addition, the images for each section are obtained at different angles of rotation of the microscope stage. Augmentation is applied to the obtained photographs: the number of images increases due to rotations, shifts, and rescaling of the image. For training the neural network of the Mask R-CNN architecture, transfer learning is used, with initial weights obtained from a huge variety of nongeological images. The results of image segmentation using Mask R-CNN are compared to the Watershed algorithm results and the U-Net network for two metrics. According to the standard Intersection over Union metric, U-Net for high-quality images and Watershed for blurry images show the best results with a slight superiority. However, according to the Grain Size Metric, which evaluates the accuracy of grain-size measurement, the best accuracy (over 95%) is shown by Mask R-CNN. The grain-size analysis is done, and the porosity of the studied petrographic sections is determined. The use of the proposed approaches in the study of thin sections will significantly reduce the time for obtaining the results of grain-size-distribution analysis and porosity determination.
This article is the result of multidisciplinary collaboration between geologists and programmers. This has allowed for the merging of profound knowledge in the field of geology with cutting-edge data processing technologies. By employing the presented methodology, geologists can devote more time to interpreting results rather than obtaining them, which in turn enhances the efficiency of research work. The benefits of using this methodology are not limited to just speeding up the process: it also allows for increased accuracy and reliability of the analysis, minimizing human error.