Submitted to: ASABE Annual International Meeting
Publication Type: Proceedings
Publication Acceptance Date: 7/30/2011
Publication Date: 8/24/2011
Citation: Liu, Y., Gamble, G.R., Thibodeaux, D.P. 2011. Improving NIR model for the prediction of cotton fiber strength. 2011 ASABE Annual International Meeting, August 7-11, 2011, Louisville, KY. p. 1-10. Interpretive Summary: One of essential end-use qualities in cotton fibers is strength component, which is related to the manufacturing of quality goods for consumers. In cotton industry, fiber strength has been mostly measured by automation-based high volume instrument (HVI) system. An alternative method to HVI testing is near infrared (NIR) spectroscopy, a useful technique due to the speed, ease of use, and adaptability to on-line or off-line implementation. Although previous NIR models have suggested the promising in accurate and reliable prediction of strength quality, in this research we have reported a much improved NIR model for HVI strength if the calibration samples were pre-screened. As a different and complementary approach, the FT-IR model for Stelometer strength revealed that the capability of FT-IR model on Stelometer strength is in good agreement with that of NIR model on HVI strength, verifying the acceptability of NIR technique in the robust and reliable model development for determining the cotton strength property. The outcome provides cotton fiber / textile engineers, researchers and regulators a new sight in applying both HVI and IR/NIR spectroscopy for rapid and routine determination of cotton strength quality.
Technical Abstract: Cotton fiber strength is an important quality characteristic that is directly related to the manufacturing of quality consumer goods. Currently, two types of instruments have been implemented to assess cotton fiber strength, namely, the automation oriented high volume instrument (HVI) and the laboratory based Stelometer. Each has unique merits but correlation between two readings was reported to be relatively low. Nevertheless, two strength values from two independent testing should be highly correlated. This concept could be applied to determine the appropriate samples in calibration set, prior to develop NIR model on HVI strength. For the first time, we proposed a pre-screening procedure to determine appropriate calibration samples in NIR models. As a validation and complementary approach, the FT-IR spectra were collected on small broken Stelometer specimens (~0.4 mg) and correlated with cotton Stelometer strength. The results suggested that the capability of NIR model on HVI strength is in good agreement with that of FT-IR model on Stelometer strength, verifying the potential of NIR technique in robust, reliable and quantitative determination of HVI strength.