updated 20210526

some of githubers said they cannot reach some old datasets, I am sharing them with baidu cloud. here https://pan.baidu.com/s/1kM5Lh9-s2yImuQE2ks0dqg with pass word : lvth

Researchers of surface defect often suffer from the lack of the corresponding dataset.
Thanks to the data provider's generous support, we can collect these datasets.

You can download some of these datasets which are modified by us, they are in the zip files in this project. Others datasets you can visit the links we provide.

The image datasets are only for academic research, no commercial purposes are allowed. If you use any datasets, please cite the paper of the corresponding provider

There are several different surfaces, each surface contains one or several defects. For image segmentation, object detection, saliency detection, classification, etc.

The image datasets are:

Magnetic tile dataset

Magnetic tile dataset by ourselves. Can be downloaded from https://github.com/abin24/Magnetic-tile-defect-datasets. which was used in our paper "Surface defect saliency of magnetic tile", the paper can be reach by here or here

Some datasets we have used before

these three datasets are used in our paper A Compact Convolutional Neural Network for Surface Defect Inspection

Texture dataset https://github.com/abin24/Textures-Dataset

RSDDs dataset http://icn.bjtu.edu.cn/visint/resources/RSDDs.aspx

wood defect dataset http://www.ee.oulu.fi/~olli/Projects/Lumber.Grading.html

Cracks on the surface of the construction.

Cracks on the bridge(left) and crack on the road surface.

Bridge cracks. There are 2688 images of bridge crack without pixel-level ground truth. From the authors "Liangfu Li Weifei Ma Li Li Xiaoxiao Gao". Files can be reached by visiting https://github.com/maweifei/Bridge_Crack_Image_Data.
Crack on road surface. From Shi, Yong, and Cui, Limeng and Qi, Zhiquan and Meng, Fan and Chen, Zhensong. Original dataset can be reached at https://github.com/cuilimeng/CrackForest-dataset. We extract the image files of the pixel level ground truth.

Steel surface defects

These Datasets from Kechen Song, Northeastern University (NEU). You can visit their Homepage http://faculty.neu.edu.cn/me/songkc/Vision-based_SIS_Steel.html to download 1800 images.

There is something wrong with the original website of this dataset now(2019-03-20). There is a pdf file named "NEU_surface_defect.pdf" in this repo, and google drive link of this dataset is given in the pdf file.

-examples of this dataset are shown in the figure below:

these 6 class NEU surface defect dataset has only the image level label, without pixel level groundtruth.

Their "Micro surface defect database" and "Oil pollution defect database" are labeled with pixel level ground truth by us. You can clone this repo and download this file.

Weakly Supervised Learning for Industrial Optical Inspection

You can download these 10 datasets by visit https://hci.iwr.uni-heidelberg.de/node/3616 Samples like the figure below:

If you have any difficulty to access these datasets, contact me by huangyibin2014@ia.ac.cn or [727948828@qq.com]