Location: Sugarbeet and Bean Research
2009 Annual Report
Data analysis will involve determining portions of the spectra and the mode, or combinations of spectra and modes, capable of best predicting density and sugars levels.
The progression of the research would include the first year of broad spectral measurement of whole tubers and tissue samples and critical analysis of data. A second year would be important (similar budget) to validate first year findings and, in parallel, focus on a particular sensing configuration based on what we learned from year 1. This sensing configuration would be a step toward a prototype device incorporating dedicated electronics for real-time in-field measurements.
Different spectra pretreatment methods were evaluated and used to remove the variations in the spectra caused by unknown sources that tend to increase errors in calibration models. Partial least squares regression (PLSR) analysis was used to develop calibration models for each of the constituents per cultivar. After the calibration equations were established, samples in the validation group were used to assess the prediction accuracy of the calibration equations, as measured by the standard error of prediction (SEP), correlation coefficient (R), bias and the ratio of standard deviation of reference data in the prediction sample set to SEP (RPD).
Excellent predictions of leaf primordial was achieved using the interactance method on the sliced samples, with the R values of 0.97 and 0.95, the RPD of 4.03 and 3.90, and the SEP of 2.76 and 1.71 for cultivars FL and RB, respectively. These results show the potential of VIS/NIR technique as a means for nondestructive measurement of leaf primordial of potato with reasonable accuracy. Relatively good predictions of glucose was achieved using the interactance method on the sliced samples, with the R values of 0.85 and 0.92, RPD of 1.70 and 2.64, and SEP of 0.02 and 0.18 for FL and RB, respectively. Hence the VIS/NIR technique is potentially useful for nondestructive measurement of potato glucose concentration; however more studies are recommended to increase the method accuracy. Relatively poor predictions of sucrose were achieved using the interactance method on sliced samples, with the R values of 0.67 and 0.60, the RPD of 1.57 and 1.19, and the SEP of 0.01 and 0.13 for FL and RB, respectively. This demonstrated the limited potential of VIS/NIR technique for nondestructive measurement of potato sucrose concentration; however, further studies are recommended to possibly increase the method’s accuracy and/or validate this finding. Very poor prediction results were obtained in determining specific gravity and soluble solids, with the R values being lower than 0.5. This indicates that the evaluated techniques could not effectively measure potato soluble solids and specific gravity. Project progress was monitored via meetings, emails, and joint sessions on planning, execution, and analysis of research with the collaborating researchers.