/DEye

Keep an Eye on Defects Inspection.

Primary LanguageC++Apache License 2.0Apache-2.0

DEye (Keep an Eye on Defects Inspection)


1. Abstract

Defect Eye (DEye) is a deep learning-based software for manufacturing surface defect inspection. It provides the basic function modules to facilitate the development of different defect inspection applications. The applications cover the full rang of manufacturing environment, including incoming process tool qualification, wafer qualification, glass surface qualification, reticle qualification, research and development. Also, It can be used for medical image inpsection, including Lung PET/CT,breast MRI, CT Colongraphy, Digital Chest X-ray images. This software library contains the basic function modules about data processing, model training and model inference. It is developed to reduce the burden of programmers who worked in this field. Based on this software, developers can design the added functions according to their requirements..

DEye

2. Usage

Compiled tensorflow-r1.4 GPU version using CMake,VisualStudio 2017, CUDA8.0, cudnn6.0.

How to use DEye

  • Please find the user manual in ./docs/DEye_User_Manual.pdf.

3. Applications

3.1 IC Chips Defects Inspection

3.2 Highway Road Crack Damage Inpection

3.3 Fabric Defects Inpection

3.4 Cover Glass Inpection

3.5 Civil Infrastructure Defect Detection

3.6 Power lines Crack Detection

3.7 Medical Image Classification

4. Datasets

  1. Weakly Supervised Learning for Industrial Optical Inspection

    DAGM: https://hci.iwr.uni-heidelberg.de/node/3616

  2. Micro surface defect database

    https://pan.baidu.com/s/1QM0AxlGjUlkHHyxwamIMmA

  3. Oil pollution defect database

    https://pan.baidu.com/s/1_aU_Bfh7lcxpYW1no2MlUQ

  4. Bridge Crack Image Data

    http://pan.baidu.com/s/1bplPrPl

  5. ETHZ Datasets

    ETHZ: http://www.vision.ee.ethz.ch/en/datasets/

  6. RSDDs dataset

    http://icn.bjtu.edu.cn/Visint/resources/RSDDs.aspx

  7. Crack Forest Datasets

    https://github.com/cuilimeng/CrackForest

  8. CV Datasets on the Web

    http://www.cvpapers.com/datasets.html

  9. An RGB-D dataset and evaluation methodology for detection and 6D pose estimation of texture-less objects

    http://cmp.felk.cvut.cz/t-less/

5. Contact

Notice: Any comments and suggetions are welcomed, kindly please introduce yourself(name, country, organization etc.) when contact with me, thanks for your cooperation.

6. TODO List

  • A user-friendly GUI ( welcome to contact with me if you want to be a collaborator)

7. License

Apache License 2.0

8. Citation

Use this bibtex to cite this repository:

@misc{DEye,
  title={A Deep Learning-based Software for Manufacturing Defect Inspection},
  author={Sundy},
  year={2017},
  publisher={Github},
  journal={GitHub repository},
  howpublished={\url{https://github.com/sundyCoder/DEye}},
}