2021 Fiscal Year Annual Research Report
Development of an integrated sugarcane quality management system using combined use of optical sensors
| Project/Area Number |
21J14214
|
|---|
| Research Institution | Kagoshima University |
|---|
Principal Investigator |
Aparatana Kittipon 鹿児島大学, 連合農学研究科, 特別研究員(DC2)
|
|---|
| Project Period (FY) |
2021-04-28 – 2023-03-31
|
| Keywords | Sugarcane / Sugar / Near infrared / Spectroscopy / Multivariate analysis |
|---|
| Outline of Annual Research Achievements |
We were able to successfully develop calibration models for both a benchtop visible near-infrared (Vis-NIR) spectrometer using sugarcane juice spectra and sugar quality measured by the conventional method, and a portable Vis-NIR spectrometer using sugarcane stalk spectra and sugar quality measured by the benchtop Vis-NIR spectrometer.
For the benchtop spectrometer, the best model for Brix calibration was developed using standard normal variate (SNV) spectral pretreatment, resulting in a coefficient of determination of 0.99 and an RMSEP of 0.2 percent Brix. For Pol calibration, the second derivative was the most effective spectral pretreatment, producing a coefficient of determination of 0.99 and an RMSEP of 0.3 percent Pol.
As for the portable Vis-NIR spectrometer, the second derivative was the best model for Brix calibration, resulting in a coefficient of determination of 0.70 and an RMSEP of 1.4 percent Brix. For Pol calibration, the second derivative combined with SNV pretreatment was the most effective, with a coefficient of determination of 0.70 and an RMSEP of 1.4 percent Pol.
Overall, our work demonstrated that it is feasible to develop reliable calibration models for both benchtop and portable Vis-NIR spectrometers using sugarcane juice and stalk spectra, respectively. Our approach is not only cost-efficient, but it also reduces the processing time required for analysis.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
As the pandemic situation worsened, quarantine restrictions prevented us from traveling and conducting new experiments. To continue our work, we had to rely on data from the previous year and focus on editing the dataset, conducting analysis, developing calibration models, and performing validation. In addition, we had to adapt our lives to communicate and work online in response to the limitations of physical meetings and interactions.
|
| Strategy for Future Research Activity |
Our plan is to conduct experiments on Minamidaito; Island in Okinawa, Japan, during the growth or ripening stage of sugarcane (October-December). To measure sugar quality in the sugarcane field, we intend to use a portable Vis-NIR spectrometer. We will scan the internode surface of the sugarcane stalk, once on the bottom and once on the top, and then take an average. We will also investigate the spectral data of the sugarcane stalk.
Furthermore, we plan to collect multispectral images using an unmanned aerial vehicle equipped with a multispectral camera (DJI Phantom 4 multispectral) to calculate vegetation indices. By checking the correlation between sugar quality measured by a portable Vis-NIR spectrometer and vegetation indices, we will be able to evaluate the effectiveness of the multispectral images in estimating sugar quality.
|
