fabric fault

Published: 10 July 2025| Version 1 | DOI: 10.17632/7d37tk8m7d.1
Contributors:
Mohd Shafi Pathan,

Description

Data collection involved gathering a diverse dataset of fabric images, including both defect-free and defective samples. Specific fabric faults such as holes, stains, tears, and deformations were included . Images were meticulously annotated to indicate defect locations and types. For normalization, images were rescaled (1./255) and resized to 128x128 pixels using ImageDataGenerator . The system uses digital images grabbed by industrial camera like Machine Vision Camera Model No - VCXG.2-57C Machine Vision Lens Size - 6mm M118FM16 MACHINE VISION RING LIGHT Model No - CMVRL1366925240210 Machine Vision Camera Enclosure Material and camera enclosure mounting bracket. Setup furniture including motable table and lights

Files

Steps to reproduce

Data collection involved gathering diverse fabric images, which included both defect-free and defective samples , and these were meticulously annotated to indicate defect locations and their specific types . The instruments utilized for this data capture included a Machine Vision Camera (Model No: VCXG.2-57C), a Machine Vision Lens (6mm M118FM16), and a MACHINE VISION RING LIGHT (Model No: CMVRL1366925240210). This setup also encompassed a Machine Vision Camera Enclosure Material with a camera enclosure mounting bracket, along with setup furniture including a movable table and lights [Query]. For data processing and normalization, images were rescaled (1./255) and resized to 128x128 pixels using ImageDataGenerator. The system relies on processing these digital images using TensorFlow and Keras

Institutions

MIT Art Design and Technology University

Categories

Textile Engineering, Textile Industry

Licence