Deep Neural Networks for Automatic Grain-matrix Segmentation in Plane and Cross-polarized Sandstone Photomicrographs

TL;DR

This paper introduces DSGSN, a novel deep learning framework for automatic grain segmentation in sandstone images, outperforming existing methods in accuracy and robustness for petrographic analysis.

Contribution

First application of deep neural networks to sandstone grain segmentation, providing a robust, end-to-end data-driven solution for petrographic image analysis.

Findings

DSGSN achieves higher segmentation accuracy than existing architectures.

The method is robust against varied sandstone petrography patterns.

Deep learning improves grain-matrix segmentation reliability.

Abstract

Grain segmentation of sandstone that is partitioning the grain from its surrounding matrix/cement in the thin section is the primary step for computer-aided mineral identification and sandstone classification. The microscopic images of sandstone contain many mineral grains and their surrounding matrix/cement. The distinction between adjacent grains and the matrix is often ambiguous, making grain segmentation difficult. Various solutions exist in literature to handle these problems; however, they are not robust against sandstone petrography's varied pattern. In this paper, we formulate grain segmentation as a pixel-wise two-class (i.e., grain and background) semantic segmentation task. We develop a deep learning-based end-to-end trainable framework named Deep Semantic Grain Segmentation network (DSGSN), a data-driven method, and provide a generic solution. As per the authors' knowledge, this is the first work where the deep neural network is explored to solve the grain segmentation problem. Extensive experiments on microscopic images highlight that our method obtains better segmentation accuracy than various segmentation architectures with more parameters.