7,348 images of casting products categorized into defective and non-defective classes. The data enables automated visual inspection of submersible pump impellers within industrial manufacturing environments.
Use Cases
- Train a binary image classifier using the 'def_front' and 'ok_front' labels
- Benchmark computer vision models for industrial quality inspection tasks using the 300x300 pixel image set
- Develop automated defect detection systems using the provided casting product images
Strengths
- 7,348 images of casting products
- Binary labels for 'def_front' and 'ok_front' classes
- Grayscale image format at 300x300 pixel resolution