Published 2025-07-08

Study on the automated characterization of particle size and shape of stacked gravelly soils via deep learning

Data source language:

  • English

Data reference:

  • Study on the automated characterization of particle size and shape of stacked gravelly soils via deep learning

Contributors of raw data:

  • 龚健/广西大学

Data description:

  • This dataset is designed to support the automatic recognition and segmentation of nonoverlapping and overlapping particles in granular images. All images were captured from a top-down view using a Canon 5D Mark IV camera equipped with an EF24-105mm lens, with a resolution of 6720×4480 pixels. The dataset contains a total of 420 images. Each image has been annotated at the instance level using the LabelMe tool, with particles categorized into two types of labels: “nonoverlapping particles” and “overlapping particles” based on whether they are occluded by other particles. In total, the dataset includes 1814 nonoverlapping particles and 17,321 overlapping particles, covering three typical particle shape types: angular ballast, sub-rounded particles, and rounded river pebbles. It is worth noting that the original version of the dataset, as described in the related publication, was divided into two separate subsets for training models on nonoverlapping and overlapping particles, respectively. However, to unify the data structure and reduce storage requirements for open access, the dataset was reorganized into a single collection of 420 images. Each image in this updated version contains both types of particles, allowing users to flexibly select target categories based on their specific tasks. Additionally, due to the corruption of a small number of original images during storage, those images were retaken and re-annotated. As a result, the total number of images and annotations in this released version may differ slightly from those reported in the paper, but the dataset remains representative and suitable for practical use.

Keywords and subjects:

  • Gravelly soils, Stacked particle

Availability:

  • publicly accessible
0