Font Size: a A A

Research And Design Of Automatic Fabric Defects Inspection System

Posted on:2006-05-04Degree:MasterType:Thesis
Country:ChinaCandidate:Y J GongFull Text:PDF
GTID:2178360182469953Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
In the process of textile production, quality control is very important. Detection of fabric defects is an important part of this. Presently, much of the fabric inspection is performed manually by human inspectors. Many defects are missed, and the inspection is inconsistent, with its outcome depending on the training and the skill level of the personnel. Since 1990's, automation of fabric inspection has been a topic of considerable research in automation of textile industry. In this paper, a project of design of automatic fabric inspection system is bringing forward. It includes the system hardware structure design, software structure design, image capturing module design, fabric defects detection algorithm and defects classification algorithm design and research. Fabric image capturing and fabric defects inspection algorithm are the key points of system design. First, after briefly introducing the principle of the image capturing card, we give the software design of fabric image capturing. The software is divided into three level parts. It includes Windows driver of image capturing card, the API(Application Programming Interface)of the card, fabric image capturing .This software architecture avail to software maintenance. Second, we discuss the research and design of algorithms of automatic inspection of fabric defects. The arm of fabric inspection is finding the information of fabric defects. We divide the task of fabric inspection into three parts: defects detection, defects classification, defects segmentation. Base on the deeply analysis fabric inspection and summarizing the research before, we propose a fabric detection algorithm using wavelet analysis, and an fabric classification algorithm using BP Neural Network. Fabric detection algorithm creates and optimizes to select the wavelet in order to improve it's detection effect and capability of self-adaptation. Fabric classification algorithm considers the fabric feature parameters selection and optimization of training algorithm of BP neural network. The fabric defect automatic inspection system in the paper is still in research and design. Though some achievement have been got, the system has many parts should improve in order to satisfy the need of the fabric industry. In the end of the paper, we give some idea to improve the system.
Keywords/Search Tags:Fabric Inspection, Machine Vision, Image Capturing, Device Driver, Fabric Defect Detection, Wavelet Analysis, Fabric Defect Classification, BP Neural Network
PDF Full Text Request
Related items