Bugao Xu

Identifying Fabric Structures with Fast Fourier Transform Techniques

The Fast Fourier Transform (FFT) plays a very important role in image processing and pattern recognition. Since a woven fabric consists of regular repeating units, the FFT is particularly useful fo...The Fast Fourier Transform (FFT) plays a very important role in image processing and pattern recognition. Since a woven fabric consists of regular repeating units, the FFT is particularly useful for analyzing periodicity, directionality, and spacing of re peating units in the fabric. This paper describes procedures for applying FFT techniques in image processing to identify weave pattern, fabric count, yarn skewness, and other structural characteristics of woven fabrics. A color scanner is used to digitize fabric images (two-dimensional functions in a spatial domain), and a customized software package is used to apply the FFT to the images. A power spectrum image is derived from the FFT of an image, and considered a two-dimensional function in a frequency domain. Peaks in the power spectrum image stand for frequency terms of periodic elements, from which basic weave patterns (e.g., plain, twill, satin, etc.) can be rec ognized. A radial function and an angular function, derived from the power spectrum, are used to measure the coarseness and directionality of the periodic elements. By selecting power peaks in a certain direction to reconstruct the image, warp or weft elements can be extracted to facilitate fabric count and skewness measurements. Fourier filtering, that is, image filtering in the frequency domain, is used to suppress noise and to select features that have a certain range of frequencies. Fabrics with various weave patterns and yarn counts are tested using the FFT techniques.

Automatic inspection of pavement cracking distress

We present an image processing algorithm customized for high-speed, real-time inspection of pavement cracking. In the algorithm, a pavement image is divided into grid cells of 8×8 pixels, and each cell is classified as a noncrack or crack cell using the grayscale information of the border pixels. Whether a crack cell can be regarded as a basic element (or seed) depends on its contrast to the neighboring cells. A number of crack seeds can be called a crack cluster if they fall on a linear string. A crack cluster corresponds to a dark strip in the original image that may or may not be a section of a real crack. Additional conditions to verify a crack cluster include the requirements in the contrast, width, and length of the strip. If verified crack clusters are oriented in similar directions, they will be joined to become one crack. Because many operations are performed on crack seeds rather than on the original image, crack detection can be executed simultaneously when the frame grabber is forming a new image, permitting real-time, online pavement surveys. The trial test results show a good repeatability and accuracy when multiple surveys were conducted at different driving conditions.

Evaluation of a rotary laser body scanner for body volume and fat assessment

This paper reports the evaluation tests on the reliability and validity of a 3-dimensional (3D) laser body scanner for estimation of body volume and % fat. Repeated measures of body imaging were performed for reproducibility analysis. Validity of the instrument was assessed by comparison of measures of body volume by imaging to hydrodensitometry, and body fat was compared to hydrodensitometry and dual energy X-ray absorptiometry. Reproducibility analysis showed little difference between within-subjects measurements of volume (ICC ≥ 0.99, p < 0.01). Body volume estimations by laser body scanner and hydrodensitometry were strongly related (r = 0.99, p < 0.01), and agreement was high (ICC = 0.99, p < 0.01). Measurements of % body fat also agreed strongly with each other between methods (ICC = 0.86, p < 0.01), and mean % fat estimates by body imaging did not differ from criterion methods (p > 0.05). These findings indicate that the 3D laser body scanner is a reliable and valid technique for the estimation of body volume. Furthermore, body imaging is an accurate measure of body fat, as compared to dual energy X-ray absorptiometry. This new instrument is promising as a quick, simple to use, and inexpensive method of body composition analysis.

A real-time 3D scanning system for pavement distortion inspection

Pavement distortions, such as rutting and shoving, are the common pavement distress problems that need to be inspected and repaired in a timely manner to ensure ride quality and traffic safety. This paper introduces a real-time, low-cost inspection system devoted to detecting these distress features using high-speed 3D transverse scanning techniques. The detection principle is the dynamic generation and characterization of the 3D pavement profile based on structured light triangulation. To improve the accuracy of the system, a multi-view coplanar scheme is employed in the calibration procedure so that more feature points can be used and distributed across the field of view of the camera. A sub-pixel line extraction method is applied for the laser stripe location, which includes filtering, edge detection and spline interpolation. The pavement transverse profile is then generated from the laser stripe curve and approximated by line segments. The second-order derivatives of the segment endpoints are used to identify the feature points of possible distortions. The system can output the real-time measurements and 3D visualization of rutting and shoving distress in a scanned pavement.

Instrumental Evaluation of Fabric Pilling

Image analysis has been widely accepted as an objective method for evaluating fabric appearance, This paper presents the development of an image-analysis method that aims at characterizing and rating fabric-pilling appearance. The procedures involved in using the fast Fourier transform (FFT) and other techniques to correct image defects, such as non-uniform background and low contrast, are demonstrated. Examples are given. A novel approach that splits a fabric image into periodic and non-periodic structures is explained. A template-matching technique used for extracting pills from the non-periodic image is presented. Pill properties are characterized by density, size, and contrast, and measurements on the ASTM photographic standards are used to build the empirical grading equations of fabric-pilling.

Fiber Cross-Sectional Shape Analysis Using Image Processing Techniques:

. Quantitative characterization of fiber cross sections has attracted considerable in terlst, since cross-sectional size and shape have an important impact on the physical and mechanical properties of fibers, as well as the performance of end-use products. We present one application of automated measurement using image processing tech niques that extract basic shape features from fiber cross sections. Cross-sectional shapes are characterized with the aid of geometric and Fourier descriptors. Geometric de scriptors measure attributes such as area, roundness, and ellipticity. Fourier descriptors are derived from the Fourier series for the cumulative angular function of the cross- sectional boundary and are used to characterize shape complexity and other geometric attributes. Shape reconstruction based on Fourier coefficients is also discussed. We present the results of shape analysis for a wide variety of fiber types.

Instrumental Evaluation of Fabric Wrinkle Recovery

This paper describes the application of computer image analysis techniques to characterization and quantification of fabric wrinkling for the purpose of devising an automated rating system of fabric wrinkle recovery. Two variables, surface ratio and shade ratio, are defined to quantify wrinkled appearance. The techniques for measuring these variables are discussed. Based on the analysis of the AATCC wrinkle recovery replicas using these techniques, we found that each of these two variables approximately exhibits an exponential relationship with the grades presented by the replicas. Through non-linear regression, two empirical equations are formulated to serve as mathematical models for grading fabric wrinkle recovery. The wrinkle recovery of seven different fabrics is evaluated, respectively, by observers using AATCC test method 128 and by the image system. Results show that the computer grading is highly consistent with the visual grading, and is more accurate and reliable.

Clustering Analysis for Cotton Trash Classification

Raw cotton may contain various kinds of trash, such as leaf, bark, and seed coat particles. The content of each of these trash categories is useful information for finding more efficient cleaning processes and predicting the quality of the finished products. This paper addresses the importance of using chromatic and geometric features of trash for trash description, and presents three different clustering methods that automatically classify trash based on the feature measurements. Compared with the geometric attributes of trash, such as size and shape, color attributes are less changeable during harvesting and ginning of cotton and are therefore more reliable and descriptive in categorizing trash. Three clustering methods—sum of squares, fuzzy, and neural network—prove effective for trash classification. Sum of squares clustering and fuzzy clustering require iterative computations and generate comparable classification accuracy. Neural network clustering yields the highest accuracy, but it needs more computational time for network training.

Automatic segmenting and measurement on scanned human body

Purpose – This paper presents methods and algorithms to automatically segment and measure the human body.Design/methodology/approach – In the segmentation procedure, two different methods are designed to find the crotch point for the situation of non‐contacted thigh and contacted thigh, respectively. Three different methods: minimum distance algorithm, minimum inclination angle algorithm, and directional neighbor identification algorithm are introduced to search the branching points or triangle. In the body measurement procedure, a pre‐sorted circling method is designed for circumference measurement, and the basic principle of landmark acquisition has been discussed. These techniques are validated via testing over different type of scanned model.Findings – The results of automatic segmentation and body measurement have verified that our methods are efficient and versatile in processing different type of scanned body.Research limitations/implications – The accurate and automatic locating of wrist, ankle an...

Evaluating Carpet Appearance Loss: Surface Intensity and Roughness

Changes in the appearance of carpet pile due to mechanical wear may be evaluated with respect to changes in global image properties, i.e., those based on one-dimensional characterizations of the intensity surface. This paper describes the application of gray scale image analysis to the measurement of microtexture variation and roughness or relief. We attempt to characterize microtexture with intensity mean, variance, skewness, and spatial co-occurrence. Surface relief is quantified as normalized intensity area and fractal dimension. Our samples show that simple mechanical wear generally entails a loss of relief and variation, but carpet construction has an important effect on the observed trends. Intensity mean and fractal dimension are promising candidates for automated evaluation of carpet appearance.