scholarly journals Dry Matter Estimation of Standing Corn with Near-Infrared Reflectance Spectroscopy

2021 ◽  
Vol 37 (5) ◽  
pp. 775-781
Author(s):  
Matthew F. Digman ◽  
Jerry H. Cherney ◽  
Debbie J. Cherney
Keyword(s):  
Moisture Content ◽  
Prediction Model ◽  
Near Infrared ◽  
Reflectance Spectroscopy ◽  
Infrared Reflectance ◽  
Near Infrared Reflectance Spectroscopy ◽  
Near Infrared Reflectance ◽  
Infrared Reflectance Spectroscopy ◽  
Whole Plant ◽  
Plant Moisture

HIGHLIGHTSQuadratic relationships were established to relate ear moisture or stover moisture to whole plant moisture, and they explained 90% and 84% of whole plant moisture, respectively. Based on our observations, the moisture content of a corn field can be estimated within +1% w.b. in 19 out of 20 fields by sampling 5-10 plants. The calibration offered by SCiO was successful at predicting oven-dried moisture content based on traditional NIRS metrics of R2 = 0.92, RMSE = 3.6, RPD = 3.2, and RER = 15. However, the 95% prediction bands were +6.9% w.b., which would indicate little utility in estimating ear moisture content. Based on a prediction model that was developed using the data collected for this study, a significant instrument-to-instrument bias was observed, indicating the necessity of including multiple SCiO devices in calibration spectra collection. ABSTRACT. Determining the appropriate time to harvest whole-plant corn is an essential factor driving the successful preservation via anaerobic fermentation (ensiling). The current options for timely on-farm monitoring of corn moisture in the field include selecting a set of representative plants, chopping and drying a subsample, or harvesting a portion of the field using a harvester equipped with an on-board moisture sensing system. Both methods are time-consuming and expensive, limiting their practicality for harvest decision-making. This work’s objective was to develop a practical solution that utilizes the moisture content of the ear to estimate whole-plant moisture. An improvement of this method was also considered that utilized a hand-held near-infrared reflectance spectroscopy (NIRS) device to predict ear moisture in situ. Based on the data collected during this work, a quadratic relationship was developed where ear moisture explained 90% of the variability in whole-plant corn moisture. However, based on our observations, the hand-held NIRS evaluated would have little utility in predicting whole-plant corn moisture with either the calibration developed here or provided by the manufacturer. The manufacturer’s prediction model yielded the best result with an R2 of 0.92, and a ratio of performance to deviation of 3.19. However, the 95% prediction band was +6.85% w.b. Finally, we determined that for a corn field uniform in appearance, sampling five to ten plants is likely to provide a reasonable estimate of field moisture. Keywords: Corn silage, Forage analysis, Harvest timing, Moisture content, NIRS.

scholarly journals Prediksi Kadar Salinitas, pH dan C-Organik Tanah Menggunakan Near Infrared Kecamatan Baitussalam Kabupaten Aceh Besar

2020 ◽  
Vol 4 (4) ◽  
pp. 542-551
Author(s):  
Riska Nurul Saputri ◽  
Ichwana Ichwana ◽  
Agus Arip Munawar
Keyword(s):  
Prediction Model ◽  
Near Infrared ◽  
Reflectance Spectroscopy ◽  
Least Square ◽  
Organic Soils ◽  
Infrared Reflectance ◽  
Near Infrared Reflectance Spectroscopy ◽  
Near Infrared Reflectance ◽  
Infrared Reflectance Spectroscopy ◽  
Spectrum Correction

Abstrak. Akuisisi spektrum Near Infrared Reflectance Spectroscopy (NIRS) terkait kualitas dan kondisi tanah telah banyak dilakukan dalam berbagai penelitian. Pada penelitian ini menggunakan model prediksi Partileal Least Squares (PLS) dengan metode koreksi spektrum Mean Normalization (MN), Savitzky-Golay Smoothing, dan kombinasi Mean Normalization (MN) dan Savitzky-Golay Smoothing. Sampel tanah yang digunakan berasal dari Kecamatan Baitussalam Kabupaten Aceh Besar karena dianggap sesuai untuk prediksi kadar salinitas, pH dan C-Organik tanah. Hasil dari penelitian menunjukkan adanya korelasi antara prediksi Near Infrared Reflectance Spectroscopy (NIRS) dengan hasil aktual laboratorium setelah dilakukan pembangunan model prediksi Partileal Least Square (PLS) dan dievaluasi dengan parameter statistika; penggunaan pretreatment Mean Normalization (MN) merupakan metode terbaik atau pilihan, dimana dapat meningkatkan keakuratan hasil prediksi kadar salinitas, pH dan C-Organik tanah.Prediction of Salinity, pH and C-Organic Soils Level Using Near  in Baitussalam Regency, Aceh Besar RegencyAbstract. Near Infrared Reflectance Spectroscopy (NIRS) spectrum acquisition related to soil quality and condition has been carried out in various studies. This study used prediction model Partileal Least Squares (PLS) with the spectrum correction methods used are Mean Normalization (MN), Savitzky-Golay Smoothing, and Combination of Mean Normalization (MN) and Savitzky-Golay Smoothing. The soil samples used were from Baitussalam regency, Aceh Besar regency because they were considered suitable for the prediction of salinity, pH and C-Organic soils. The results of this study showed a correlation between the prediction of Near Infrared Reflectance Spectroscopy (NIRS) with the actual results of the laboratory after the construction of the prediction model Partileal Least Square (PLS) and and evaluated with statistical parameters; the use of pretreatment Mean Normalization (MN) is the best or preferred spectrum correction method, which can improve the accuracy of the predicted results of salinity, pH and C-Organic soil.

EFFECT OF GROWING YEAR AND APPLICATION OF A MULTI-YEAR CALIBRATION FOR PREDICTING QUALITY PARAMETERS BY NEAR INFRARED REFLECTANCE SPECTROSCOPY IN WHOLE-PLANT CORN FORAGE

1990 ◽  
Vol 70 (3) ◽  
pp. 747-755 ◽  
Author(s):  
E. V. VALDES ◽  
G. E. JONES ◽  
G. J. HOEKSTRA
Keyword(s):  
Dry Matter ◽  
Near Infrared ◽  
Reflectance Spectroscopy ◽  
Quality Parameters ◽  
Infrared Reflectance ◽  
Near Infrared Reflectance Spectroscopy ◽  
Near Infrared Reflectance ◽  
Test Set ◽  
Infrared Reflectance Spectroscopy ◽  
Whole Plant

Near infrared reflectance spectroscopy analysis (NIRA) to predict quality parameters in whole-plant corn forage was investigated. Quality parameters studied included acid detergent fiber (ADF), in vitro dry matter digestibility (IVDMD) and crude protein (CP). Samples of whole-plant corn forage were collected during three growing years (1984, 1985, 1986) across six geographical locations in Ontario, Canada and were harvested at an average 35% dry matter content. Samples were randomly divided into two sets: a calibration (CAL) set to develop NIRA equations and a testing (TEST) set to validate these equations. NIRA calibrations for ADF, IVDMD and CP percent were performed for each growing season across locations. A multi-year calibration was also developed with samples drawn from the three growing years. The accuracy of the NIRA predictions was assessed by the standard error of the estimate (SEE), bias or the mean difference between laboratory and NIRA data, the coefficient of determination (r2) and the slope (b) of the regressions between laboratory and NIRA data. The single year calibrations showed good predictions for all quality parameters in samples drawn within the year. The SEE values in the TEST sets for single year calibrations varied from 1.6 to 1.9% for ADF, 1.5 to 2.3% for IVDMD and 0.3 to 0.5% for CP, respectively. However, recalibration was necessary every year because calibrations based on a single year failed to account for variances introduced by samples drawn from other years. The multi-year calibration predicted ADF, IVDMD and CP percent accurately regardless of year. The SEE and bias for ADF, IVDMD and CP percent for the TEST set were 1.6 and 0.1, 2.2 and 0.3, and 0.5 and 0.1, respectively. The success of the multi-year calibration encourage the possibility to develop a standard calibration for whole-plant corn.Key words: Near infrared reflectance analysis, calibrations, whole-plant corn forage, year, multi-year quality parameters

Analysis of Rubber, Resin, and Moisture Content of Guayule by Near Infrared Reflectance Spectroscopy

1985 ◽  
Vol 58 (2) ◽  
pp. 304-313 ◽  
Author(s):  
L. T. Black ◽  
G. E. Hamerstrand ◽  
W. F. Kwolek
Keyword(s):  
Moisture Content ◽  
Near Infrared ◽  
Reflectance Spectroscopy ◽  
Infrared Reflectance ◽  
Near Infrared Reflectance Spectroscopy ◽  
Near Infrared Reflectance ◽  
Trade Off ◽  
Infrared Reflectance Spectroscopy

Abstract Although the overall error associated with the NIR analysis of guayule may be slightly higher than that of some other methods, the trade-off between the increased error and the very rapid aquistion of data by NIR is quite positive.

Sign in / Sign up

Export Citation Format

Share Document