Casting defect detection: Datasets & Models
Automated visual inspection of metal castings to identify surface and structural defects like porosity, cracks, and shrinkage. Uses object detection to flag defective parts on the production line.
Casting defect detection
Casting defect detection runs object detection on images of cast metal parts to locate and classify flaws such as porosity, blowholes, cracks, cold shuts, and shrinkage cavities. It takes input images captured from fixed line cameras or handheld devices and returns bounding boxes with defect labels so operators can flag and reject defective castings.
Try it out
Powered by general detection model
Run on custom image
Drop an image here or click to upload
Why automate casting defect detection
Manual visual inspection of castings is slow, subjective, and fatiguing, inspectors miss subtle porosity or hairline cracks, and consistency drops across shifts. Automating casting defect detection brings repeatable, objective inspection to every part instead of a sampled subset, catching defective castings before they move downstream into machining or assembly where scrap costs multiply. For high-volume foundries and metalworking operations, this means fewer escaped defects, tighter documentation for customer audits, and inspection that keeps pace with line speed.
Key benefits
A trained detection model inspects every casting consistently, reduces the labor burden of manual QC, and produces a defect log tied to each part for traceability. It integrates with existing machine-vision cameras and inline conveyors, so you can deploy without re-tooling the line. As volumes grow or new part geometries are introduced, the same pipeline scales by adding labeled examples and retraining.
How to use it
- Upload an image: Drop any photo into the widget above to try casting defect detection for free, right on this page.
- Preview instant results: The model detects porosity, crack, shrinkage cavity, blowhole, metal casting 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 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 casting defect detection
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 casting defect detection, fine-tune it on your own labeled images, then deploy to a Serverless Cloud API endpoint.
Not sure where to start?
Describe your Casting Defect Detection project to Roboflow Agent.