Prediction performances of portablenearinfraredinstrumentsforat farm forage analysis
The objective of this study was to evaluate the useofNearInfrared Spectroscopy (NIRS) to analyze maize silage withaportable instrument. The instrument was a Zeiss Corona45workingbetween 960 and 1700 nm which was used in Italy,CzechRepublic andPoland. Best prediction performances were obtainedusing theItalian data set. Prediction error were 1.0,0.16and0.4respectivelyfor DM, CP and NDF on a as is basis.With theinstrument from Polandand Czech Republic there were lower accuracyof prediction comparedto the Italian dataset,probablyfortheirlimited (less than 100samples) calibration dataset. Mergingall the data set improvedprediction accuracy for CP but not DM.Itwould appear that someform of instrument standardization is neededbefore merging dataset.
作者使用Zeiss Corona 45 應用NIRS光譜法來分析及預測青貯飼料-玉米，作者分別在波蘭、捷克采樣、義大利。
