Rock fragmentation analysis: Datasets & Models
Automated measurement of rock fragment size distribution from images of muck piles, conveyor belts, and haul truck loads using instance segmentation. Each rock is segmented individually so fragment dimensions and gradation can be computed.
Rock fragmentation analysis
This segmentation workflow processes images of blasted rock from muck piles, conveyor belts, and haul truck loads to measure fragment size distribution across mining and quarrying operations. It uses instance segmentation to isolate each rock fragment, boulder, and fine particle individually, letting you compute gradation curves, P80 values, and oversize counts from any rock fragmentation dataset.
Try it out
Powered by general detection model
Run on custom image
Drop an image here or click to upload
Why automate rock fragmentation analysis
Manual rock fragmentation analysis is slow and inconsistent: geologists sieve samples or eyeball muck piles, and results vary by operator and lighting. Automating with a segmentation model lets you process fragmentation images continuously and objectively, catching oversize boulders and excessive fines that drive up crusher wear, secondary breakage costs, and downstream mill throughput problems. A trained model turns routine site photos into quantitative fragment size distributions without pulling crews off other work.
Key benefits
Instance segmentation delivers per-rock measurements rather than coarse averages, so you get accurate size distributions and reliable oversize detection. The workflow runs on existing pit cameras, conveyor-mounted cameras, or phone photos, and scales from a single quarry face to fleet-wide monitoring. Consistent, repeatable fragmentation data lets teams tie blast design changes directly to downstream loading and crushing performance.
How to use it
- Upload an image: Drop any photo into the widget above to try rock fragmentation analysis for free, right on this page.
- Preview instant results: The model detects rock fragment, boulder, fine particle, muck pile, scale reference within seconds and returns an annotated image with structured predictions, so you can immediately see how well it works. Adjust the detection classes and re-run until the results fit your scenario.
- Like what you see?: With a free Roboflow account, you can deploy this workflow with the Serverless Cloud API, customize it in your workspace, or train RF-DETR (instance segmentation) on your own images for production accuracy.
- Get started for free: No code or training data required to begin: upload, review the results, and make it part of your stack in a single click.
Recommended model for rock fragmentation analysis
RF-DETR is state of the art for real-time object detection, instance segmentation, and keypoint detection, with higher accuracy and lower latency than comparable models, and it's Apache 2.0 licensed. When you're ready to build a production model for rock fragmentation analysis, fine-tune it on your own labeled images, then deploy to a Serverless Cloud API endpoint.
Not sure where to start?
Describe your Rock Fragmentation Analysis project to Roboflow Agent.