ITDD
收藏资源简介:
工业纺织品缺陷检测(ITDD)数据集包含1885张工业纺织品图像,分为4类:棉织物、染色织物、麻织物和格子织物。这些类别是从[WEIQIAO Textile](http://www.wqfz.com/en/)的工业生产现场收集的。ITDD是[WFDD](https://github.com/cqylunlun/GLASS?tab=readme-ov-file#1wfdd-download-link)的升级版本,重新组织了三个原始类别并添加了一个新类别。
The Industrial Textile Defect Detection (ITDD) dataset includes 1,885 industrial textile images grouped into four classes: cotton fabric, dyed fabric, linen fabric, and plaid fabric. These classes were collected from the industrial production site of [WEIQIAO Textile](http://www.wqfz.com/en/). ITDD is an upgraded version of [WFDD](https://github.com/cqylunlun/GLASS?tab=readme-ov-file#1wfdd-download-link), which reorganizes three original categories and adds one new category.
CRAS数据集概述
数据集基本信息
- 名称:CRAS (Center-aware Residual Anomaly Synthesis)
- 类型:工业异常检测数据集
- 包含内容:
- 源代码(PyTorch实现)
- 自建数据集ITDD
数据集详情
ITDD数据集
- 全称:Industrial Textile Defect Detection
- 数据量:1885张工业纺织品图像
- 类别:
- 棉织物 (cotton fabric)
- 染色织物 (dyed fabric)
- 麻织物 (hemp fabric)
- 格子织物 (plaid fabric)
- 来源:中国WEIQIAO纺织工业生产现场
- 特点:WFDD数据集的升级版本,重组了三个原始类别并新增一个类别
- 下载链接:https://drive.google.com/file/d/1Iy-f_jxJFhSxDc4J0f85wwQKuwj1NzvX/view?usp=sharing/
相关公开数据集
- MVTec AD:https://www.mvtec.com/company/research/datasets/mvtec-ad/
- VisA:https://github.com/amazon-science/spot-diff/
- MPDD:https://github.com/stepanje/MPDD/
实验环境
-
Python版本:3.9.21
-
GPU要求:NVIDIA GeForce RTX 4090 (24GB)
-
依赖安装:
conda create -n cras_env python=3.9.21 conda activate cras_env pip install -r requirements.txt
引用信息
bibtex @article{chen2025center, title={Center-aware Residual Anomaly Synthesis for Multi-class Industrial Anomaly Detection}, author={Chen, Qiyu and Luo, Huiyuan and Yao, Haiming and Luo, Wei and Qu, Zhen and Lv, Chengkan and Zhang, Zhengtao}, journal={arXiv preprint arXiv:2505.17551}, year={2025} }
许可协议
- 许可证类型:MIT许可证
- 许可证链接:https://github.com/cqylunlun/CRAS?tab=MIT-1-ov-file/




