Back to datasets
Dataset assetOpen Source CommunityMaterial Defect DetectionIndustrial Image Processing
ZDSteel
The Belt‑shaped Alloy Functional Material Surface Defect Dataset was collected in an industrial processing workshop using two line‑scan cameras and lenses to capture upper and lower surface defect images of alloy functional materials. An on‑line coaxial red light source was used. The dataset comprises 2,942 defect images of eight types (pits, peeling, bulging, welds, scratches, notches, indentations, and perforations) at a resolution of 4096 × 2048 pixels.
Source
github
Created
Oct 22, 2024
Updated
Oct 22, 2024
Signals
248 views
Availability
Linked source ready
Overview
Dataset description and usage context
ZDSteel Dataset Overview
Dataset Source
- The dataset was obtained from the backend winding section of the production line in an alloy functional material industrial processing workshop.
Data Acquisition Equipment
- Two line‑scan cameras and lenses were used to capture upper and lower surface defect images of belt‑shaped alloy functional materials in real time.
- A coaxial red light source was employed.
Dataset Scale
- A total of 2,942 defect images were collected.
Image Resolution
- Each image has a resolution of 4096 × 2048 pixels.
Defect Types
- The dataset includes eight defect categories:
- Pit
- Peeling
- Bulge
- Weld
- Scratch
- Notch
- Indentation
- Perforation
Need downstream help?
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.