In this paper we present a novel wrinkle evaluation method that uses modified wavelet coefficients and optimized support-vector-machine (SVM) classifications to characterize and classify the wrinkling appearance of fabric. Fabric images were decomposed with the wavelet transform, and five parameters were defined, based on the modified wavelet coefficients, to describe wrinkling features, such as orientation, hardness, density, and contrast. These parameters were also used as the inputs of optimized SVM classifiers to obtain overall wrinkle grading in accordance with the standard American Association of Textile Chemists and Colorists smoothness appearance (SA) replicas. The SVM classifiers, based on a linear kernel and a radial-basis-function kernel, were used in the study. The effectiveness of this evaluation method was tested by 300 images of five selected fabrics that had different fiber contents, weave structures, colors, and laundering cycles. The cross-validation tests on the SA classifications indicated that the SA grades of more than 75% of these diversified samples could be recognized correctly. The extracted wrinkle parameters provided useful information for textile, appliance, and detergent manufactures to inspect wrinkling behaviors of fabrics.

White specks are a specific type of fiber defect that result in high financial losses to the cotton industry. Fiber entanglements are called neps. Neps that involve immature fibers do not dye properly and appear as white specks on the dyed fabric. Studies to predict white specks from bale fiber measurements are underway. Initially a reliable method for measuring white specks is needed. Several systems have been evaluated and are reported here. The systems accuracy was compared using fiber from the US Extreme Variety Study, which was grown specifically to have different levels of white specks. This paper sets out the experimental work and analysis undertaken to develop and validate a system for reliably quantifying the amount of white specks in a woven fabric. Four image analysis systems are compared. This includes two industrial imaging systems (Cambridge and Optimas)2 and two systems specifically developed for white speck analysis (Cotton Incorporated’s prototype and AutoRate). The Cambridge system was too sensitive for this application, and the Cotton Incorporated system was found to have drift in the data over time so that the problem could not be identified. The Optimas system is time consuming and not accurate enough for this application. The AutoRate system gives the most accurate measurements of white specks in the minimal amount of time, with minimal operator error, of all of the systems studied and is currently being used in developing prediction of white specks from bale fiber properties.

No abstract

Smooth leaf and hairy leaf Midsouth cottons were processed through minimum, two intermediate, and maximum gin cleanings. Based on fiber measurements after mill cleaning, minimum gin cleaning yielded the best fiber quality ( i.e., length, low short fiber content). Fiber damage as a result of maximum gin cleaning was evident through all levels of mill cleaning. There were also differences in nonlint content as a result of cleaning machinery and cotton varieties.

Seed coat fragments (SCFs) reduce the marketability of cotton fiber, yarns and fabrics. It is particularly important to measure SCF content in the fabric because they cause severe dyeing and appearance defects. SCF content is greatly affected by cotton varieties, environmental conditions during crop development, and mechanical processing, but studying their effects in fabrics can be very tedious and time consuming. In this paper, we present an image analysis system for accurate and fast measurement of SCFs in greige fabrics, and the conditions of using the system for reliable and repeatable data. In the study, four different US cotton varieties were selected, and processed with regular manufacturing facilities. The relationship between sample size and precision of the image analysis system was determined through statistical analysis. It was found that the minimum sample size for each variety should consist of five camera images with a minimum of four fabric samples per variety with three replications, which gives a least significant difference (LSD) of 64.54 for dark speck count. Dark speck counts for the four fabrics tested ranged from 267.7 to 659.9. Increasing sample size will lower the LSD.