Industrial Weld Surface Defect Detection and Localization Dataset

#Object detection #Defect classification #Bounding box localization #Industrial visual inspection #Image annotation parsing #Welding line quality inspection #Weld rework localization #In-line industrial camera inspection #Welding process quality traceability #Manufacturing defect screening

6420
0.3GB
piece
CATL
MOBIUSI INC

Updated:2026-11-05

AI Analysis & Value Prop

This dataset contains images of industrial weld surfaces paired with defect categories and location annotations, supporting the recognition and localization of defects in weld imagery. Records also include image identifiers, image and annotation file paths, image width and height, and file format to support data management and integration into training workflows. It is suited to automated weld quality inspection, production-line visual inspection, and the development and evaluation of industrial defect recognition models; sample counts, defect category distribution, and image specifications should be verified against the available samples.

Dataset Insights

Sample Examples

e611da66**.jpg||0 B

Technical Specifications

Field	Type	Description
file_name	string	File name
quality	string	Resolution
image_id	string	A unique identifier for the image.
class_name	string	The name of the defect category assigned to the image or annotation.
image_path	string	The file path to the image.
label_path	string	The file path to the annotation file.
image_width	number	The width of the image in pixels.
image_format	string	The image file format.
image_height	number	The height of the image in pixels.

Compliance Statement

Authorization Type	Proprietary - Commercial AI Training License (No Redistribution)
Commercial Use	Requires exclusive subscription or authorization contract (monthly or per-invocation charging)
Privacy and Anonymization	No PII, no real company names, simulated scenarios follow industry standards
Compliance System	Compliant with China's Data Security Law / EU GDPR / supports enterprise data access logs

Frequently Asked Questions

What does the dataset contain?: It contains industrial weld surface images with defect categories and location annotations. Fields also include image identifiers, image and annotation file paths, image width and height, and file format.

Are sample counts and image specifications provided?: The available information does not specify the total number of samples, resolution ranges, or counts by defect category. Review the available samples and annotation files before use to confirm that the dataset size and specifications meet your project requirements.

Which models and tasks is it suitable for?: The data can support weld surface defect detection, defect classification, and image-region localization, as well as the development and evaluation of industrial visual inspection workflows. Suitability depends on the annotation format, category distribution, and defect types required by your project.

How can I obtain the data, and what are the licensing terms?: Use the dataset product page or the provider's published channel to request access. The available information does not specify the delivery method or license scope, so confirm the included files, commercial-use rights, redistribution restrictions, and applicable terms before use.

Can't find the data you need?

Post a request and let data providers reach out to you.

Get this Dataset

Verified for Enterprise Use

Cite this Work

@dataset{Mobiusi2026,
  title={Industrial Weld Surface Defect Detection and Localization Dataset},
  author={MOBIUSI INC},
  year={2026},
  url={https://www.mobiusi.com/datasets/43065047dc105c3292c754ed8b6def20},
  urldate={2026-11-03},
  keywords={industrial weld defect dataset,weld surface defect detection,weld quality inspection data,industrial object detection data,weld defect annotations,welding visual inspection,surface defect recognition,industrial image dataset},
  version={1.0}
}

Using this in research? Please cite us.

placeholder

Scene

Domain

Modality

Task