• Journal of The Korean Society of Grassland and Forage Science
• /
• v.39 no.2
• /
• pp.114-120
• /
• 2019

This study was carried out to explore the accuracy of near infrared spectroscopy(NIRS) for the prediction of moisture content and chemical parameters on winter annual forage crops. A population of 2454 winter annual forages representing a wide range in chemical parameters was used in this study. Samples of forage were scanned at 1nm intervals over the wavelength range 680-2500nm and the optical data was recorded as log 1/Reflectance(log 1/R), which scanned in intact fresh condition. The spectral data were regressed against a range of chemical parameters using partial least squares(PLS) multivariate analysis in conjunction with spectral math treatments to reduced the effect of extraneous noise. The optimum calibrations were selected based on the highest coefficients of determination in cross validation($R^2$) and the lowest standard error of cross-validation(SECV). The results of this study showed that NIRS calibration model to predict the moisture contents and chemical parameters had very high degree of accuracy except for barely. The $R^2$ and SECV for integrated winter annual forages calibration were 0.99(SECV 1.59%) for moisture, 0.89(SECV 1.15%) for acid detergent fiber, 0.86(SECV 1.43%) for neutral detergent fiber, 0.93(SECV 0.61%) for crude protein, 0.90(SECV 0.45%) for crude ash, and 0.82(SECV 3.76%) for relative feed value on a dry matter(%), respectively. Results of this experiment showed the possibility of NIRS method to predict the moisture and chemical composition of winter annual forage for routine analysis method to evaluate the feed value.

• Korean Journal of Soil Science and Fertilizer
• /
• v.42 no.6
• /
• pp.522-528
• /
• 2009

This study focused on establishing prediction models using visible-near infrared spectrum to simultaneously detect multiple components of soils and enhancing the performance quality by suitably transformed input spectra and classification of soil spectral types for prediction model input. The continuum-removed spectra showed significant result for all cases in terms of soil properties and classified or bulk predictions. The prediction model using classified soil spectra at an absorption peak area around 500nm and 950nm efficiently indicating soil color showed slightly better performance. Especially, Ca and CEC were well estimated by the classified prediction model at $R^{2}$ > 0.8. For organic carbon, both classified and bulk prediction model had a good performance with $R^{2}$ > 0.8 and RPD> 2. This prediction model may be applied in global soil mapping, soil classification, and remote sensing data analysis.

• Korean Journal of Remote Sensing
• /
• v.37 no.3
• /
• pp.409-418
• /
• 2021

The land cover map is a very important data that is used as a basis for decision-making for land policy and environmental policy. The land cover map is mapped using remote sensing data, and the classification results may vary depending on the acquisition time of the data used even for the same area. In this study, to overcome the classification accuracy limit of single-period data, multi-series satellite images were used to learn the difference in the spectral reflectance characteristics of the land surface according to seasons on a U-Net model, one of the deep learning algorithms, to improve classification accuracy. In addition, the degree of improvement in classification accuracy is compared by comparing the accuracy of single-period data. Seoul, which consists of various land covers including 30% of green space and the Han River within the area, was set as the research target and quarterly Sentinel-2 satellite images for 2020 were aquired. The U-Net model was trained using the sub-class land cover map mapped by the Korean Ministry of Environment. As a result of learning and classifying the model into single-period, double-series, triple-series, and quadruple-series through the learned U-Net model, it showed an accuracy of 81%, 82% and 79%, which exceeds the standard for securing land cover classification accuracy of 75%, except for a single-period. Through this, it was confirmed that classification accuracy can be improved through multi-series classification.

• Korean Journal of Plant Resources
• /
• v.30 no.1
• /
• pp.38-49
• /
• 2017

The objective of this research was to develop Near-Infrared Reflectance Spectroscopy (NIRS) model for amylose and protein contents analysis of large accessions of rice germplasm. A total of 511 accessions of rice germplasm were obtained from National Agrobiodiversity Center to make calibration equation. The accessions were measured by NIRS for both brown and milled brown rice which was additionally assayed by iodine and Kjeldahl method for amylose and crude protein contents. The range of amylose and protein content in milled brown rice were 6.15-32.25% and 4.72-14.81%, respectively. The correlation coefficient ($R^2$), standard error of calibration (SEC) and slope of brown rice were 0.906, 1.741, 0.995 in amylose and 0.941, 0.276, 1.011 in protein, respectively, whereas $R^2$, SEC and slope of milled brown rice values were 0.956, 1.159, 1.001 in amylose and 0.982, 0.164, 1.003 in protein, respectively. Validation results of this NIRS equation showed a high coefficient determination in prediction for amylose (0.962) and protein (0.986), and also low standard error in prediction (SEP) for amylose (2.349) and protein (0.415). These results suggest that NIRS equation model should be practically applied for determination of amylose and crude protein contents in large accessions of rice germplasm.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1032-1032
• /
• 2001

Artificial Neural Network (ANN) calibration techniques have been used commercially for agricultural applications since the mid-nineties. Global models, based on transmission data from 850 to 1050 nm, are used routinely to measure protein and moisture in wheat and barley and also moisture in triticale, rye, and oats. These models are currently used commercially in approx. 15 countries throughout the world. Results concerning earlier European ANN models are being published elsewhere. Some of the findings from that study will be discussed here. ANN models have also been developed for coarsely ground samples of compound feed and feed ingredients, again measured in transmission mode from 850 to 1050 nm. The performance of models for pig- and poultry feed will be discussed briefly. These models were developed from a very large data set (more than 20,000 records), and cover a very broad range of finished products. The prediction curves are linear over the entire range for protein, fat moisture, fibre, and starch (measured only on poultry feed), and accuracy is in line with the performance of smaller models based on Partial Least Squares (PLS). A simple bias adjustment is sufficient for calibration transfer across instruments. Recently, we have investigated the possible use of ANN for a different type of NIR spectrometer, based on reflectance data from 1100 to 2500 nm. In one study, based on data for protein, fat, and moisture measured on unground compound feed samples, dedicated ANN models for specific product classes (cattle feed, pig feed, broiler feed, and layers feed) gave moderately better Standard Errors of Prediction (SEP) compared to modified PLS (MPLS). However, if the four product classes were combined into one general calibration model, the performance of the ANN model deteriorated only slightly compared to the class-specific models, while the SEP values for the MPLS predictions doubled. Brix value in molasses is a measure of sugar content. Even with a huge dataset, PLS models were not sufficiently accurate for commercial use. In contrast an ANN model based on the same data improved the accuracy considerably and straightened out non-linearity in the prediction plot. The work of Mr. David Funk (GIPSA, U. S. Department of Agriculture) who has studied the influence of various types of spectral distortions on ANN- and PLS models, thereby providing comparative information on the robustness of these models towards instrument differences, will be discussed. This study was based on data from different classes of North American wheat measured in transmission from 850 to 1050 nm. The distortions studied included the effect of absorbance offset pathlength variation, presence of stray light bandwidth, and wavelength stretch and offset (either individually or combined). It was shown that a global ANN model was much less sensitive to most perturbations than class-specific GIPSA PLS calibrations. It is concluded that ANN models based on large data sets offer substantial advantages over PLS models with respect to accuracy, range of materials that can be handled by a single calibration, stability, transferability, and sensitivity to perturbations.

• Journal of Satellite, Information and Communications
• /
• v.1 no.2
• /
• pp.1-7
• /
• 2006

GOCi (Geostationary Ocean Color Imager) is one of the COMS payloads that KARI is currently developing and scheduled to be in operation from around 2008. Its primary objective is to monitor the Korean coastal water environmental condition. We report the current progress in development of the integrated optical model as one of the key analysis tools for the GOCI in-orbit performance verification. The model includes the Sun as the emitting light source. The curved Earth surface section of 2500 km x 2500 km includingthe Korean peninsular os defined as a Lambertian scattering surface consisted of land and sea surface. From its geostationary orbit, the GOCI optical system observes the reflected light from the surfaces with varying reflectance representing the changes in its environmental conditions. The optical ray tracing technique was used to demonstrate the GOCI in-orbit performances such as red tide detection. The computational concept, simulation results and its implications to the on-going development of GOCI are presented.

• Aerospace Engineering and Technology
• /
• v.13 no.2
• /
• pp.80-86
• /
• 2014

In this study, we the estimated surface normalized difference vegetation index by using the KOrea Multi-Purpose SATellite-3 (KOMPSAT-3) multi-spectral images for comparative analysis. The estimated NDVI from KOMPSAT-3 is used as for comparison with the high resolution GeoEye products. The geometry conditions for atmospheric effects are selected from meta files of KOMPSAT-3 bundle data. The used geometry conditions are consist of solar zenith angle, solar azimuth angle, viewing zenith angle, viewing azimuth angle, and date. And, Atmospheric effects such as attenuation, scattering and absorption were physically simulated from water vapor, ozone and aerosol information. Generally, although ground measurements are important for accurate information, in this study, MODIS atmospheric products are used as atmospheric constituents. The surface reflectance from radiative transfer model is utilized for estimating vegetation index. The present study, to reduce atmospheric and geometry conditions between KOMPSAT-3 and GeoEye having difference observation characteristics, data acquisition time is carefully determined for reliable vegetation spectral characteristics.

• Journal of Environmental Impact Assessment
• /
• v.7 no.1
• /
• pp.81-91
• /
• 1998

The objective of this study was to use remotely sensed data, combined with in situ data, for the assessment of trophic state for Daecheong reservoir. Three Landsat TM(Thematic Mapper) imagery data were processed to portray trophic state conditions. The remotely sensed data and the measured data were obtained on 20 June 1995. Regression models have been developed between the chlorophyll-a concentration and reflectance which was converted to Landsat TM digital data. The regression model was determined based on the correlation coefficient which was higher than 0.7 and was applied to the entire study area to generate a distribution map of chlorophyll-a and trophic state. The equation, providing estimates of chlorophyll-a concentration, represented the year-to-year spatial variation of trophic zones in the reservoir. Satellite remote sensing data derived from Landsat TM had been successfully used for trophic slate mapping in Daecheong reservoir.

• Journal of Biosystems Engineering
• /
• v.34 no.5
• /
• pp.382-389
• /
• 2009

An in-situ probe-based spectrophotometer has been developed. This system used two spectrometers to measure soil reflectance spectra from 450 nm to 2200 nm. It collects soil electrical conductivity (EC) and insertion force measurements in addition to the optical data. Six fields in Kansas were mapped with the VIS-NIR (visible-near infrared) probe module and sampled for calibration and validation. Results showed that VIS-NIR correlated well with carbon in all six fields, with RPD (the ratio of standard deviation to root mean square error of prediction) of 1.8 or better, RMSE of 0.14 to 0.22%, and $R^2$ of 0.69 to 0.89. From the investigation of carbon variability within the soil profile and by tillage practice, the 0-5 cm depth in a no-till field contained significantly higher levels of carbon than any other locations. Using the selected calibration model with the soil NIR probe data, a soil profile map of estimated carbon was produced, and it was found that estimated carbon values are highly correlated to the lab values. The array of sensors (VIS-NIR, electrical conductivity, insertion force) used in the probe allowed estimating bulk density, and three of the six fields were satisfactory. The VIS-NIR probe also showed the obtained spectra data were well correlated with nitrogen for all fields with RPD scores of 1.84 or better and coefficient of determination ($R^2$) of 0.7 or higher.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1245-1245
• /
• 2001

Food adulteration is a serious consumer fraud and a matter of concern to food processors and regulatory agencies. A range of analytical methods have been investigated to facilitate the detection of adulterated or mis-labelled foods & food ingredients but most of these require sophisticated equipment, highly-qualified staff and are time-consuming. Regulatory authorities and the food industry require a screening technique which will facilitate fast and relatively inexpensive monitoring of food products with a high level of accuracy. Near infrared spectroscopy has been investigated for its potential in a number of authenticity issues including meat speciation (McElhinney, Downey & Fearn (1999) JNIRS, 7(3), 145-154; Downey, McElhinney & Fearn (2000). Appl. Spectrosc. 54(6), 894-899). This report describes further analysis of these spectral sets using a hierarchical approach and binary decisions solved using logistic regression. The sample set comprised 230 homogenized meat samples i. e. chicken (55), turkey (54), pork (55), beef (32) and lamb (34) purchased locally as whole cuts of meat over a 10-12 week period. NIR reflectance spectra were recorded over the wavelength range 400-2498nm at 2nm intervals on a NIR Systems 6500 scanning monochromator. The problem was defined as a series of binary decisions i. e. is the meat red or white\ulcorner is the red meat beef or lamb\ulcorner, is the white meat pork or poultry\ulcorner etc. Each of these decisions was made using an individual binary logistic model based on scores derived from principal component or partial least squares (PLS1 and PLS2) analysis. The results obtained were equal to or better than previous reports using factorial discriminant analysis, K-nearest neighbours and PLS2 regression. This new approach using a combination of exploratory and logistic analyses also appears to have advantages of transparency and the use of inherent structure in the spectral data. Additionally, it allows for the use of different data transforms and multivariate regression techniques at each decision step.