aggregate dataset

Description of the Aggregate Dataset Our aggregate dataset was constructed through the following steps: First, we extracted planar parameters from tiled photographs of each coarse aggregate (corresponding to the files photo_group_1.rar and photo_group_2.rar) and measured their volumes. Then, we scanned the 3D models of these aggregates. Finally, we compiled the aggregate ID numbers, planar parameters, and volumes into the dataset (corresponding to the file dataset.xlsx). The aggregates are divided into two groups, classified by sieve size (into 7 categories). Data were collected in multiple batches, so the ID numbers follow a specific encoding rule: The first digit indicates the group number. The second digit represents the aggregate category. The third and fourth digits denote the collection batch. The fifth (and sometimes sixth) digit(s) indicate the specific aggregate number. For example: An aggregate with ID 141610 belongs to Group 1, Category 4, Batch 16, and is the 10th aggregate in that batch. An aggregate with ID 23033 belongs to Group 2, Category 3, Batch 03, and is the 3rd aggregate in that batch. These IDs serve as unique identifiers, ensuring a one-to-one correspondence between physical aggregates, photographs, and 3D models, facilitating traceability and providing a foundation for future research. Additionally, due to the challenges of 3D scanning, some STL models did not accurately represent the true shapes of the aggregates. These were excluded and marked with yellow highlighting in dataset.xlsx. It should be noted that, due to limited shooting space, each photograph only captures a subset of the aggregates (the specific subset can be identified from the photo’s filename). The aggregates in the photos are arranged from the bottom-left corner, progressing rightward row by row, and ending at the top-right corner.