| Abstrakt: |
Reflectance spectra of cotton fiber samples having different fiber quality levels have been measured with a high resolution spectrophotometer. The relationship between micronaire and spectral reflectance was evaluated with multiple regression models and wavelet analysis. Regression models indicate that the micronaire had a strong correlation (R2=0.89) with the reflectance values in seven 100-nm wavebands (1120, 1296, 1550, 1664, 1852, 2020, and 2340 nm). Micronaire could be predicted very well even with a model involving only two wavebands as independent variables (R2=0.88). Six wavelet regressors were identified and included in the calibration model during wavelet analysis. The calibration model also showed a very strong correlation between the spectral reflectance and micronaire (R2=0.97). A cotton fiber quality sensing system was designed based on the characteristics of the cotton fiber reflectance spectrum. The sensor will consist of VisGaAs camera and a data collecting and processing system with an on-line sampling device. Images at three NIR wavebands (1450, 1550, and 1605 nm) will be acquired and used with a predetermined algorithm to determine fiber micronaire. A singleboard- computer based system will be employed to control the on-line sampling device, change optical filters, acquire images, and synchronize the time. Fiber images along with their corresponding spatial data from GPS will be recorded and stored as cotton is harvested in the field for creating cotton quality maps. The sensor also has the potential being used for segregating cotton based on fiber quality during harvesting. [ABSTRACT FROM AUTHOR] |
