الفهرس 
acknowledgmentOnly 14 pages are availabe for public view
 |  | from | 104 |  |  |
| | | from | 104 | | |
Abstract
The textile industry is continually pressing for higher product quality and improved productivity to meet both customer demands and to reduce the cost associated with quality. Also higher production speeds make the timely detection of fabric defects more important than ever. The work of inspectors is very tedious and time consuming, and the effectiveness of visual inspection decreases quickly with fatigue .So, the development of a flexible, efficient, reliable, and integrated real-time vision system is an essential issue in quality control process for textile manufacturers.
Description of problems in the textile industry is often uncertain, vague, or subjective to be useful .To overcome this uncertainty and achieve automated on-line textile defects recognition control, fuzzy expert systems have been used. Fuzzy logic system has been widely and successfully used in many areas such as system modeling and control, data analysis, and pattern recognition but computing with Type-2 fuzzy sets (T2FSs) can require undesirably large amount of computations since it involves numerous embedded T2FSs.To reduce the complexity, interval type-2 fuzzy sets (IT2FSs) have been used, since the secondary memberships are all equal to one.
In this thesis, a modified method for textile defects recognition is proposed. Reducing error on identifying fabric defects requires more automotive and accurate inspection process, so this method is based on interval Type-2 fuzzy logic system .Type-2 fuzzy sets have been shown to manage uncertainty more effectively than T1 fuzzy sets (TIFSs) and IT2FSs are more practical than T2FSs. The set of features used are Area, Threshold, Mean and Standard Deviation. Experimental results for several data sets are given, which showed the effectiveness of the suggested technique for detecting fabric defects and also show the privilege and high accuracy when compared with the state-of-the-art defects recognition methods.