Julien Fouilloux's List: Professional : Papers

•  Soongsil University, 1-1 Sangdo-dong, Dongjak-Gu, Seoul 156-743, Korea, yjeong@ssu.ac.kr<script type="text/javascript"><!--  var u = "yjeong", d = "ssu.ac.kr"; document.getElementById("em0").innerHTML = '<a href="mailto:' + u + '@' + d + '">' + u + '@' + d + '<\/a>'//--></script> 

       The image analysis technique for finding the directions of warp and weft is introduced. The method is applied to some adorned fabrics like stripe, check, and dogtooth as well as unadorned fabrics of basic weaves to verify its efficiency and accuracy. Also, the parameters affecting the results are discussed. The method predicts well the warp and weft directions for adorned fabric as well as unadorned fabrics of basic weaves. Thus, the method can be useful in the image analysis area for textiles, especially, for determining fabric density.

• Cheng-Chih Tsai 

    Intelligence Control and Simulation Laboratory, Department of Polymer Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

       This paper proposes an approach to texture analysis that could be used to recognize the fabric nature and type of the main weaving texture. First, the color scanner captures the fabric image and saves it as a digital image and then the wavelet transformation is used to display the image texture. The co-occurrence matrix is then be applied to calculate the texture characteristics, such as angular second moment, entropy, homogeneity, and contrast, and finally, the learning vector quantization networks (LVQN) are adopted as a classifier to categorize the fabric nature and the type of weaving texture. The experimental result showed that this approach could automatically and accurately classify the fabric nature, including woven fabric, knitted fabric and non-woven fabric, and the type of its main weaving texture, such as plain, twill or satin weave, single or double knitted and non-woven fabric.

• A new method for knitted fabric defect detection and classification using image analysis and neural networks is presented. Images of six different induced defects were obtained and used in the analysis. Statistical procedures and Fourier Transforms were utilized as two different approaches in the feature extraction effort and neural networks were used to detect and classify the defects. The results showed success in detection and classification of most defects especially when the Fourier transforms technique was utilized.

• Chung-Feng Jeffrey Kuo 

    Chung-Yang Shih 

    Jiunn-Yih Lee 

    Intelligence Control and Simulation Laboratory, Department of Polymer Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China; jeff@tx.ntust.edu.tw<script type="text/javascript"><!--  var u = "jeff", d = "tx.ntust.edu.tw"; document.getElementById("em0").innerHTML = '<a href="mailto:' + u + '@' + d + '">' + u + '@' + d + '<\/a>'//--></script>  

       A novel approach of repeat pattern segmentation is proposed for printed fabrics. Printed fabrics are captured by a color scanner and converted into full color digital files. To classify the color segmentation and pattern elements a fuzzy C-means (FCM) clustering algorithm and a specific cluster-validity criterion are used to obtain the pattern image of printed fabric. Then, the repeat pattern segmentation of the pattern image is established by Hough transform. The experimental results have shown that this systematic method is very suitable for the analysis of the repeat pattern of printed fabrics.

• Chung-Feng Jeffrey Kuo 

    Intelligence Control and Simulation Laboratory, Department of Polymer Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

    Chung-Yang Shih 

    Intelligence Control and Simulation Laboratory, Department of Polymer Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

    Chih-Yuan Kao 

    Intelligence Control and Simulation Laboratory, Department of Polymer Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

    Jiunn-Yih Lee 

    Intelligence Control and Simulation Laboratory, Department of Polymer Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

       In this paper, a novel approach to color and pattern analysis is proposed for printed fabric. An unsupervised analysis method is developed using a fuzzy C-means (FCM) clustering algorithm and a specific cluster-validity criterion (SC criterion). First, the printed fabric is captured by a color scanner and converted into full color digital files, then the mean filter is used to smooth the color of the image. The search for good cluster numbers is made by the SC criterion, and the corresponding color clusters are obtained based on the FCM clustering algorithm. Finally, the color and pattern features of the printed fabric are calculated. The experimental results show that this approach is very suitable for analyzing the colors and patterns of printed fabrics.

• Chang-Chiun Huang 

    Department of Textile Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

    Sun-Chong Liu 

    Department of Textile Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

    Wen-Hong Yu 

    Department of Textile Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan, Republic of China 

       A new image processing approach is proposed for identifying weave patterns of woven fabrics and automatically displaying harness drafts and chain drafts. Fabric counts are also measured. Based on the maximum and minimum gray-level sums of the horizontal and vertical pixel lines over an entire image, yarns and their crossover points are located. The decision rules for recognizing warp and weft floats are developed on the basis of geometric features of yarn distribution. The experimental materials include plain, twill, and satin weaves, each with twenty samples. Experimental results demonstrate that three basic weave patterns can be identified. The computer measurements of fabric counts show good agreement with manual measurements.

• Jeanette M. Cardamone 

    USDA, ARS, Eastern Regional Research Center, Wyndmoor, Pennsylvania 19038, U.S.A. 

    William C. Damert 

    USDA, ARS, Eastern Regional Research Center, Wyndmoor, Pennsylvania 19038, U.S.A. 

    John G. Phillips 

    USDA, ARS, Eastern Regional Research Center, Wyndmoor, Pennsylvania 19038, U.S.A. 

    William N. Marmer 

    USDA, ARS, Eastern Regional Research Center, Wyndmoor, Pennsylvania 19038, U.S.A. 

       Conventional methods of fabric inspection are tedious and require close association. In this work, digital image analysis for fabric assessment (DIAFA) is applied to modem and historic fabrics. Fabric images are divided into arrays of pixels with peak intensities distributed over gray scale ranges (histograms) and along constructed lines (line profiles). Fabric cover is measured by segmenting a histogram's B/w pixels, yarn spacing, and yarn thickness from moving point averages by calculating the breadth of a peak corresponding to an individual yarn at half the maximum peak height for a given line profile. The results show very good agreement with known values (r = 0.919, n = 30, p < 0.01). Fourier power spectra provide facile measurements of yarn parameters in close agreement with the results from line profiling. With DIAFA, a small tool kit is developed to assist in the objective analysis of the structural integrity of fabrics in the museum environment, and it could be very useful in advancing automated fabric inspection.

                                                   

• Abstract:

   

  Two solid state cameras are used in conjunction with an IBM personal computer and a PUMA 560 robot operating under VAL I to align accurately a fabric deposited on an aluminium work surface. The method adopted and the software used is of a modular design so as to use the end-effector of the robot to achieve the aligning of the fabric or alternatively to use a dedicated device such as an aligning table. Lighting is by ordinary fluorescent light. The software filters out noise such as the frayed edges of a fabric.

• Abstract:

   

  Purpose – The main aim of this paper is description of new apparatus and approach for contact less evaluation of surface roughness. For characterization of surface roughness, the procedures based on classical and non-classical (complexity) parameters are proposed.
  Design/methodology/approach – For obtaining the roughness profile in the selected direction (on the line transect of the surface), the special arrangements of textile bend around sharp edge is used. The image analysis is used for extraction of surface profile. The system of controlled movement allows one to obtain surface roughness profile in two dimensions.
  Findings – By using aggregation (cut length principle), the roughness resolution is decreased and roughness profile is created without local roughness variation. After application of cut length principle, the direct combination of slices leads to the creation of roughness surface.
  Research limitations/implications – There exists plenty of roughness characteristics based on standard statistics or analysis of spatial processes. For evaluation of suitability of these characteristics, it will be necessary to compare results from sets of textile surfaces.
  Practical implications – The measurement of fabric roughness by an RCM device is useful as simple tool for description of roughness in individual slices and in the whole rough plane. This method replaces the traditional contact stylus profiling methods
  Originality/value – The reconstruction of surface roughness from individual slices. The utilization of aggregation principle for creation of micro and macro roughness. The evaluation of roughness parameters based on the geometrical characteristics, harmonic analysis and complexity indices. 

• Abstract:

   

  Presents an overview of automated fabric flaw detection immediately after the knitting process. Considers the classification of fabric flaws and how image processing techniques can be applied to their classification, via an introductory example. Outlines problems associated with automating this inspection process and discusses possible flaw sensing systems and techniques. 

• Constrained Connectivity for Hierarchical Image Decomposition and Simplification
• Sequential Kernel Density Approximation and Its Application to Real-Time Visual Tracking
•     Multiscale Categorical Object Recognition Using Contour Fragments

1 more annotation...

• Joint Exact Histogram Specification and Image Enhancement Through the Wavelet Transform
• PKCS: A Polynomial Kernel Family With Compact Support for Scale- Space Image Processing