Fabric Stain Detection Using YOLO Algorithm

The Fabric Stain Detection System that we made for this project is really good at fixing the problems with the way of checking for stains by hand and using regular image processing. Checking for stains by hand takes a time and you need to know what you are doing. The old image processing ways do not work well when the light is different or the fabric has a pattern. Even a simple computer model that looks at pictures does not work enough to find the exact spot of the stain so it is not very useful for companies that make fabric.   To fix these problems we used a kind of computer program called a deep learning algorithm that is like the YOLO architecture. This program can find stains tell what kind of stain it is and show where the stain is, in one step. We taught the program by showing it lots of pictures of fabric with stains on it. The program is really good at finding all kinds of stains like oil stains, dirt stains, dye stains, grease stains and water stains, on the Fabric Stain Detection System.   The project has achieved accuracy of 95% When we think about what's coming next we can suggest using newer versions of YOLO and other modern object detection algorithms to make the detection better and faster. If we make our approach work well on devices like smart cameras it can check things in real time without needing a lot of power.   We can also use this to look at how bad the defects in the fabric're not just find them. This will help people who make fabric decide if they should accept or reject it. What we might do next is make our system able to find types of defects like holes and tears in the fabric by adding more defect recognition capabilities, to YOLO and other object detection algorithms.

[1]  “Content-Based Image Retrieval And Feature Extraction:A Comprehensive review”, Mathematical Problems in Engineering, Vol. 2019, Article ID. View at: Publisher Site |Google Scholar. A. Latif, A.Rasheed, U.Sajid et al.  [2]  “Fabric Defect Detection Algorithm Using RDPSO-Based Optimal gabor filter”, The Journal of the textile Institute, vol. 110, no. 4, pp. 487-495, 2019. View at: Publisher Site |Google Scholar. Y. Li, H. Luo, M. Yu, G. Jiang.   [3] “ Computer Vision Techniques for Detecting Fabric Defects" in Applications of Computer Vision in Fashion and Textiles Elsevier, Amsterdam,Netherlands,2018. View at: Google Scholar W.Wong And J.jiang   [4]  “Visual-Based Defect Detection And Classification Approaches For Industrial Applications-a survey” Sensors VOL.20 No.5.1459 2020. view at: publisher site google scholar. T. Czimmermann. Ciuti.