Location: Cotton Structure and Quality ResearchTitle: Preliminary study of NIR model transfer in the determination of cotton trash components Author
Submitted to: Proceedings of the American Society of Agricultural and Biological Engineers International (ASABE)
Publication Type: Proceedings
Publication Acceptance Date: 6/21/2013
Publication Date: 7/21/2013
Citation: Liu, Y., Thibodeaux, D., Foulk, J., Rodgers III, J.E. 2013. Preliminary study of NIR model transfer in the determination of cotton trash components. Proceedings of the American Society of Agricultural and Biological Engineers International (ASABE). Paper No: 131594540. Interpretive Summary: Presence of trash in commercial cotton bales at varying amounts degrades the market values, requires additional cleaning process, and impacts the end-use qualities for yarn and fabric products. In order to determine the trash amount, a number of methods have been developed. In general, these methods only produce the amount of total trash, instead of the content for individual or targeted trash components. Near infrared (NIR) spectroscopy, as a rapid and low-cost method that can be used, away from the laboratory, in places such as ginning sites, has been examined for its potential in the prediction of specific and unique cotton trash components (leaves, seed coats, hulls, stems, and sand/soil) in grounded samples. Next, these NIR models were transferred to regular bulky samples with various instrumental leaf grade or trash levels. The accumulated knowledge could be of value as a rapid analytical tool to cotton breeders for cotton variety enhancement and also to cotton ginning engineers for effective trash-removal cleaning devices. The outcome provides cotton fiber / textile engineers, researchers, ginners, and regulators a new way in the determination of cotton trash components.
Technical Abstract: To assess the potential of a rapid and low-cost method that can be used, away from the laboratory, in places such as ginning sites, near infrared (NIR) spectroscopy has been applied to perform the qualitative classification and quantitative prediction of various cotton quality indices, including cotton trash. It is well-known that current-in-use trash measuring devices only generate the trash amounts in some aspects, instead of the content for individual trash components. This difficulty comes from the complexity of the co-existence of different cotton-plant related trash types, for example, leaves (leaf and bract), seed coats, hulls, and stems. To this regard, mixtures of known trash components (e.g., leaves, seed coats, hulls, stems, and sand/soil) with cut lint fibers were prepared physically and then their NIR spectra were correlated with the respective trash contents. Then the NIR models were examined for the feasibility of determining the trash amounts in commercial and uncut cotton fibers. One of great challenges is how to calibrate or validate the results from NIR model transfer by appropriate sample standards.