Fabric Inspection

Project Title

Fabric Inspection

Project Area of Specialization Artificial IntelligenceProject Summary

We are going to introduce an automated system named “fabric inspection” who detects “defects” of weave fabric. As Quality is one of the important aspects of business. Customers always demand and expect value for money. As producers of apparel there must be a constant production of work to produce good quality.in a textile field. Quality checking is a highly automated industrial process. Due to small inaccuracies during the production process, different types of weave defects can occur, by which the quality of the produced fabric is heavily impaired. The defects can diminish the selling price by up to 50%. Current automated visual defect detection systems need to be adjusted by a trained operator to every new fabric, making them impractical for industrial use. We present a novel automated visual defect detection framework which localizes and tracks yarns in new and unseen fabrics without the need for tedious settings, and which consecutively detects anomalies. The detection of weave defects is based on three consecutive steps, (1) the identification of single weft and warp float-points with fully convolutional networks, (2) the tracking of single yarns based on a set of rules, and finally (3)the recognition of defects using statistical analysis.

Project Objectives

The project objective is simple as we know that textile industry is one of the biggest industries in the world and produces several million tons of fabric every year. However, fabric defect detection is mostly provided by human operators. Whether due to fatigue, inattention or simply brief distraction, human operators are quite prone to missing even important defects in textiles. Undetected weaving defects lead to low quality finished products. In the end, the selling price of these low quality products diminishes, or they remain unsaleable. Automatic defect detection for fabrics may overcome this problem. The presently most promising approaches are all based on image analysis techniques: it is easy to take pictures of the fabric, either on-loom or off-loom using a digital cam-era, and to analyze the picture with a machine vision system.

Project Implementation Method

Tools Required:	Python 3.6, TensorFlow 1.11 ,OpenCV, NumPy ,matplotlib ,SciPy
Area/Specialization:	Artificial Intelligence

Tools Required:

Area/Specialization:

Benefits of the Project

Quality checking is a highly automated industrial process. Due to small inaccuracies during the production process, different types of weave defects can occur, by which the quality of the produced fabric is heavily impaired. The defects can diminish the selling price by up to 50%. Current automated visual defect detection systems need to be adjusted by a trained operator to every new fabric, making them impractical for industrial use. We present a novel automated visual defect detection framework which localizes and tracks yarns in new and unseen fabrics without the need for tedious settings, and which consecutively detects anomalies

Technical Details of Final Deliverable

we need graphics card of nvidia for training data.
for runing code we need raspberry pi4.
for real time detection we need Raspberry pi camera and connection it.
for table range we need motors wooden peice and led lights.

Final Deliverable of the Project HW/SW integrated systemCore Industry OthersOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Decent Work and Economic GrowthRequired Resources

Elapsed time in (days or weeks or month or quarter) since start of the project	Milestone	Deliverable
Month 1	data gathering	completed
Month 2	labelling data and resizing	completed
Month 3	annotation and augmentation	completed
Month 4	training on google colabs	completed
Month 5	testing and mature dataset	completed
Month 6	installing dependencies and setting environment	completed
Month 7	Run project on raspberry pi 4	completed