Automated Defect Detection for Textile Fabrics Using Gabor Wavelet Networks PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download Automated Defect Detection for Textile Fabrics Using Gabor Wavelet Networks PDF full book. Access full book title Automated Defect Detection for Textile Fabrics Using Gabor Wavelet Networks by Pai Peng. Download full books in PDF and EPUB format.
Author: Pai Peng
Publisher: Open Dissertation Press
ISBN: 9781361476772
Category :
Languages : en
Pages :
Get Book Here
Book Description
This dissertation, "Automated Defect Detection for Textile Fabrics Using Gabor Wavelet Networks" by Pai, Peng, 彭湃, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled Automated Defect Detection for Textile Fabrics Using Gabor Wavelet Networks submitted by PENG PAI for the degree of Doctor of Philosophy at The University of Hong Kong in December 2006 This study seeks to develop efficient methodologies to facilitate automated detection of defects in textile fabrics. Its novelty consists in combining the practical implementation of feature extraction and learning techniques by using Gabor wavelet networks (GWNs) for object representation. The study develops three structure design algorithms to determine automatically the number of hidden nodes in a GWN. The first algorithm is based on a pyramid decomposition approach, and can be used to design wavelet networks. The second algorithm is based on two important properties of GWNs, and is developed specifically for designing GWNs to solve fabric defect detection problems. These properties, which are formally established in this study, indicate that: (1) the magnitude of the network weight associated with a wavelet of a GWN trained by using an objective function governs the contribution of the wavelet in reconstructing the function; and (2) in the network training process, the translation parameters of a wavelet in the network are likely to position at the edge region of the objective function being studied. The third algorithm is based on the concept of orthogonal forward selection, and can be used to design wavelet networks for solving small and medium sized problems. For larger problems, the algorithm can be used to supplement other structure design algorithms to reduce the size of the network. A new defect detection scheme which employs 2D GWNs is proposed in this study. A superwavelet is used to ensure correct alignment between a template image and the corresponding sample images. However, the complexity analysis of the proposed scheme indicates that it is computationally demanding. To overcome this limitation, a 1D version of the above scheme which does not employ a superwavelet is developed to speed up the detection process. The scheme's good defect detection performance is confirmed by using offline experiments and by using real time experiments conducted with the prototyped automated inspection system developed in this study. The deployment of a GWN to extract features from a non-defective fabric image for the purpose of designing "optimal" Gabor filters and "optimal" morphological filters is investigated. These "optimal" filters are then used to design three defect detection schemes for textile fabrics. Another filter design method based on a real Gabor wavelet network is also proposed. The method automatically tunes the real parts of the Gabor functions to match the texture being studied. Based on these tuned-matched Gabor wavelets, a new defect detection scheme for textile fabrics is developed. The performances of all schemes are evaluated offline and in real time by using a variety of homogeneous textile fabric images. The study also proposes a complex-valued wavelet network (CVWN), which employs complex-valued multi-dimensional Gabor wavelets as the transfer functions. The feasibility and effectiveness of the CVWN are shown by solving a complicated feature extraction problem. Indeed, it can be noticed that a CVWN can be separated into two real-valued wavelet networks, namely a Gabor wavelet network and a real Gabor wavelet network. DOI: 10.5353/th_b387661
Author: Pai Peng
Publisher: Open Dissertation Press
ISBN: 9781361476772
Category :
Languages : en
Pages :
Get Book Here
Book Description
This dissertation, "Automated Defect Detection for Textile Fabrics Using Gabor Wavelet Networks" by Pai, Peng, 彭湃, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled Automated Defect Detection for Textile Fabrics Using Gabor Wavelet Networks submitted by PENG PAI for the degree of Doctor of Philosophy at The University of Hong Kong in December 2006 This study seeks to develop efficient methodologies to facilitate automated detection of defects in textile fabrics. Its novelty consists in combining the practical implementation of feature extraction and learning techniques by using Gabor wavelet networks (GWNs) for object representation. The study develops three structure design algorithms to determine automatically the number of hidden nodes in a GWN. The first algorithm is based on a pyramid decomposition approach, and can be used to design wavelet networks. The second algorithm is based on two important properties of GWNs, and is developed specifically for designing GWNs to solve fabric defect detection problems. These properties, which are formally established in this study, indicate that: (1) the magnitude of the network weight associated with a wavelet of a GWN trained by using an objective function governs the contribution of the wavelet in reconstructing the function; and (2) in the network training process, the translation parameters of a wavelet in the network are likely to position at the edge region of the objective function being studied. The third algorithm is based on the concept of orthogonal forward selection, and can be used to design wavelet networks for solving small and medium sized problems. For larger problems, the algorithm can be used to supplement other structure design algorithms to reduce the size of the network. A new defect detection scheme which employs 2D GWNs is proposed in this study. A superwavelet is used to ensure correct alignment between a template image and the corresponding sample images. However, the complexity analysis of the proposed scheme indicates that it is computationally demanding. To overcome this limitation, a 1D version of the above scheme which does not employ a superwavelet is developed to speed up the detection process. The scheme's good defect detection performance is confirmed by using offline experiments and by using real time experiments conducted with the prototyped automated inspection system developed in this study. The deployment of a GWN to extract features from a non-defective fabric image for the purpose of designing "optimal" Gabor filters and "optimal" morphological filters is investigated. These "optimal" filters are then used to design three defect detection schemes for textile fabrics. Another filter design method based on a real Gabor wavelet network is also proposed. The method automatically tunes the real parts of the Gabor functions to match the texture being studied. Based on these tuned-matched Gabor wavelets, a new defect detection scheme for textile fabrics is developed. The performances of all schemes are evaluated offline and in real time by using a variety of homogeneous textile fabric images. The study also proposes a complex-valued wavelet network (CVWN), which employs complex-valued multi-dimensional Gabor wavelets as the transfer functions. The feasibility and effectiveness of the CVWN are shown by solving a complicated feature extraction problem. Indeed, it can be noticed that a CVWN can be separated into two real-valued wavelet networks, namely a Gabor wavelet network and a real Gabor wavelet network. DOI: 10.5353/th_b387661
Author: Rajkishore Nayak
Publisher: Woodhead Publishing
ISBN: 0081011334
Category : Technology & Engineering
Languages : en
Pages : 428
Get Book Here
Book Description
Automation in Garment Manufacturing provides systematic and comprehensive insights into this multifaceted process. Chapters cover the role of automation in design and product development, including color matching, fabric inspection, 3D body scanning, computer-aided design and prototyping. Part Two covers automation in garment production, from handling, spreading and cutting, through to finishing and pressing techniques. Final chapters discuss advanced tools for assessing productivity in manufacturing, logistics and supply-chain management. This book is a key resource for all those engaged in textile and apparel development and production, and is also ideal for academics engaged in research on textile science and technology. Delivers theoretical and practical guidance on automated processes that benefit anyone developing or manufacturing textile products Offers a range of perspectives on manufacturing from an international team of authors Provides systematic and comprehensive coverage of the topic, from fabric construction, through product development, to current and potential applications
Author: Calvin Wong
Publisher: Woodhead Publishing
ISBN: 0081012187
Category : Technology & Engineering
Languages : en
Pages : 314
Get Book Here
Book Description
Applications of Computer Vision in Fashion and Textiles provides a systematic and comprehensive discussion of three key areas that are taking advantage of developments in computer vision technology, namely textile defect detection and quality control, fashion recognition and 3D modeling, and 2D and 3D human body modeling for improving clothing fit. It introduces the fundamentals of computer vision techniques for fashion and textile applications, also reviewing computer vision techniques for textile quality control, including chapters on wavelet transforms, Gibor filters, Fourier transforms, and neural network techniques. Final sections cover recognition, modeling, retrieval technologies and advanced human shape modeling techniques. The book is essential reading for scientists and researchers working in the field of fashion production, quality assurance, product development, textiles, fashion supply chain managers, R&D professionals and managers in the textile industry. Explores computer vision technology with reference to improving budget, quality and schedule control in textile manufacturing Provides a thorough understanding of the role of computer vision in developing intelligent systems for the fashion and textiles industries Elucidates the connections between human body modeling technology and intelligent manufacturing systems
Author: Tin-Chi Lee
Publisher: Open Dissertation Press
ISBN: 9781374716933
Category : Technology & Engineering
Languages : en
Pages : 172
Get Book Here
Book Description
This dissertation, "Fabric Defect Detection by Wavelet Transform and Neural Network" by Tin-chi, Lee, 李天賜, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled Submitted by LEE Tin Chi for the degree of Master of Philosophy at The University of Hong Kong in July 2004 Textile inspection plays an important role in maintaining the quality of products. In this thesis, three methods which utilize matched masks, wavelet transform and neural network are proposed for fabric defect detection. An evaluation of the performance of the methods was conducted on eight classes of fabric defects (Broken End, Dirty Yarn, Mispick, Netting Multiples, Slack End, Thick Bar, Thin Bar, and Wrong Draw). In the first method, a multi-channel filtering bank equipped with five matched masks was used. Matched masks are 2-D filters that characterize specific texture properties. They are designed to emphasize the Wrong Draw texture, the Mispick texture, the horizontal edges, the bars structure and the filled regions on fabric images. At the filter outputs, segmentation by thresholds is applied, followed by a logical OR operation. The total number of pixels exceeding the threshold on the resulting image determines whether the fabric image is defective or defect-free. Using this method, 96% of fabric defects were successfully detected, and the false alarm rate was 6%. The method achieved a 90% - 100% detection rate for most fabric defects, though the detection rate for Thin Bar defects was only 75%. The second method employed wavelet transform to decompose fabric images into multi-scales and orientations. During the training stage, the parameters to be optimized include the rotation angles and the two thresholds applied on the horizontal and vertical transformed images. The variation in rotation angles determines the selection of wavelet bases. During the detection stage, the discrimination criterion is based on the total number of defect windows. Using this method, only 76% of fabric defects were identified, and the false alarm rate was 7%. The detection rate for Dirty Yarn was high, but much lower for Broken End and Wrong Draw defects. The last method took advantage of the fault tolerance and learning ability of neural networks. We explored the texture structure of defect-free images so that feature extraction was conducted on repeating units with proper selection of locations. For defect images, similar feature vectors were extracted and passed to the neural network. Using this method, the detection rate was as high as 92% and the false alarm rate was 6%. Dirty Yarn, Netting Multiples, Mispick, Thin Bar and Wrong Draw defects were completely identified, while 75% of Broken End and Slack End defects were detected. However, only 73% of Thin Bar defects were detected. The method employing matched masks proved the most effective in detecting fabric defects. The neural network method was next best. The wavelet transform method was the least effective, because it was only able to detect effectively certain classes of fabric defects. Dirty Yarn, Netting Multiples, Mispick and Slack End defects are relatively easy to identify correctly. Wrong Draw and Thin Bar defects are less easy to detect and Broken End and Thick Bar defects are the most difficult to detect. DOI: 10.5353/th_b2928728 Subjects: Wavelets (Mathematics) Neural networks (Computer science) Textile fabrics - Testing
Author:
Publisher:
ISBN:
Category :
Languages : en
Pages :
Get Book Here
Book Description
The purpose of this research project is to show that a computerized system based on image processing software is capable of identifying defects in woven fabrics. Current defect detection is carried out through use of visual inspection of fabric rolls after the rolls have been doffed from the production machinery, which adds a substantial lag between defect creation and detection. Existing methods for automatic defect detection rely on methods that suffer from substantial analysis time or a low percentage of detection. The method described in this thesis represents a quick and accurate approach to automatic defect detection and is capable of identifying defects such as lines, tears, and spots. Utilizing a Genetic Algorithm (GA) as the primary means of solving the wavelet filter equations with respect to a fabric image proved adequate in the construction of a wavelet filter that was capable of removing large amounts of the fabric texture from the image, thus allowing defect segmentation algorithms to run more effectively. Although a real-time system is not developed, suggestions for constructing such a system are presented. This work provides a foundation for the development of a real-time automated defect detector based on the algorithms and methodologies employed in this work.
Author: Dash, Sujata
Publisher: IGI Global
ISBN: 152252858X
Category : Computers
Languages : en
Pages : 567
Get Book Here
Book Description
The digital age is ripe with emerging advances and applications in technological innovations. Mimicking the structure of complex systems in nature can provide new ideas on how to organize mechanical and personal systems. The Handbook of Research on Modeling, Analysis, and Application of Nature-Inspired Metaheuristic Algorithms is an essential scholarly resource on current algorithms that have been inspired by the natural world. Featuring coverage on diverse topics such as cellular automata, simulated annealing, genetic programming, and differential evolution, this reference publication is ideal for scientists, biological engineers, academics, students, and researchers that are interested in discovering what models from nature influence the current technology-centric world.
Author: Wai Keung Wong
Publisher: Springer
ISBN: 3319996959
Category : Technology & Engineering
Languages : en
Pages : 329
Get Book Here
Book Description
The book includes the Proceedings of the Artificial Intelligence on Fashion and Textiles conference 2018 which provides state-of-the-art techniques and applications of AI in the fashion and textile industries. It is essential reading for scientists, researchers and R&D professionals working in the field of AI with applications in the fashion and textile industry; managers in the fashion and textile enterprises; and anyone with an interest in the applications of AI. Over the last two decades, with the great advancement of computer technology, academic research in artificial intelligence (AI) and its applications in fashion and textile supply chain has been becoming a very hot topic and has received greater attention from both academics and industrialists. A number of AI-related techniques has been successfully employed and proven to handle the problems including fashion sales forecasting, supply chain optimization, planning and scheduling, textile material defect detection, fashion and textile image recognition, fashion image and style retrieval, human body modeling and fitting, etc.
Author: George Bebis
Publisher: Springer Science & Business Media
ISBN: 3642240305
Category : Computers
Languages : en
Pages : 776
Get Book Here
Book Description
The two volume set LNCS 6938 and LNCS 6939 constitutes the refereed proceedings of the 7th International Symposium on Visual Computing, ISVC 2011, held in Las Vegas, NV, USA, in September 2011. The 68 revised full papers and 46 poster papers presented together with 30 papers in the special tracks were carefully reviewed and selected from more than 240 submissions. The papers of part I (LNCS 6938) are organized in computational bioimaging, computer graphics, motion and tracking, segmentation, visualization; mapping modeling and surface reconstruction, biomedical imaging, computer graphics, interactive visualization in novel and heterogeneous display environments, object detection and recognition. Part II (LNCS 6939) comprises topics such as immersive visualization, applications, object detection and recognition, virtual reality, and best practices in teaching visual computing.
Author: Jinlian Hu
Publisher: Elsevier
ISBN: 0857093606
Category : Computers
Languages : en
Pages : 393
Get Book Here
Book Description
Computer technology has transformed textiles from their design through to their manufacture and has contributed to significant advances in the textile industry. Computer technology for textiles and apparel provides an overview of these innovative developments for a wide range of applications, covering topics including structure and defect analysis, modelling and simulation, and apparel design. The book is divided into three parts. Part one provides a review of different computer-based technologies suitable for textile materials, and includes chapters on computer technology for yarn and fabric structure analysis, defect analysis and measurement. Chapters in part two discuss modelling and simulation principles of fibres, yarns, textiles and garments, while part three concludes with a review of computer-based technologies specific to apparel and apparel design, with themes ranging from 3D body scanning to the teaching of computer-aided design to fashion students. With its distinguished editor and international team of expert contributors, Computer technology for textiles and apparel is an invaluable tool for a wide range of people involved in the textile industry, from designers and manufacturers to fibre scientists and quality inspectors. Provides an overview of innovative developments in computer technology for a wide range of applications Covers structure and defect analysis, modelling and simulation and apparel design Themes range from 3D body scanning to the teaching of computer-aided design to fashion students
Author: Mohan S
Publisher: Springer Science & Business Media
ISBN: 813221000X
Category : Technology & Engineering
Languages : en
Pages : 620
Get Book Here
Book Description
The proceedings includes cutting-edge research articles from the Fourth International Conference on Signal and Image Processing (ICSIP), which is organised by Dr. N.G.P. Institute of Technology, Kalapatti, Coimbatore. The Conference provides academia and industry to discuss and present the latest technological advances and research results in the fields of theoretical, experimental, and application of signal, image and video processing. The book provides latest and most informative content from engineers and scientists in signal, image and video processing from around the world, which will benefit the future research community to work in a more cohesive and collaborative way.
