WorkspaceCrack AnalysisNo active modelSign in
Computer vision workstation
See beyond the surface.
Upload your media. Reveal the cracks. Understand the geometry.
1Select model
2Upload media
3Explore results
Analysis setup
01Segmentation model
Input media Original resolution retained
Drop your image here
or browse files to upload
JPG, PNG, BMP, TIFF · up to 20 MBAdvanced settings
Defined by the model checkpoint.
Upload a model to enable analysis.
Segmentation viewer
LIVE WORKSPACEAWAITING INPUTIMAGE ANALYSIS
A closer look starts here.
Upload an image or video to preview your surface.
Cracks detected
—
Connected components
Total crack area
—
Segmented pixel area
Estimated length
—
Skeleton path length
Average width
—
Distance-transform estimate
Your model. A consistent workflow.
Use PyTorch (.pth) or Keras (.h5) models. Validation checks compatibility before a model can be activated.
Manage your modelsMeasurements that stay honest.
Area, length, and width are image-space estimates, not physical dimensions or model accuracy scores.
Understand your results