Home > Study finds FOSS Infratec 1241 grain analysers extremely accurate, stable

Study finds FOSS Infratec 1241 grain analysers extremely accurate, stable

Supplier News
article image These grain analysers produce incredibly accurate results
A five year ring test study has found that when FOSS Infratec grain analysers are combined with the FOSS global ANN prediction model WB003034, they predict the moisture and protein content of grain in a highly accurate manner.

FOSS has conducted collaborative studies for the European Grain Network (EGN) on an annual basis for over fifteen years.

The most recent ring test was conducted in 2011 and involved an analysis of protein and moisture content in whole kernels.

Twelve wheat and nine barley samples from the 2010 harvest were used in this particular study. The results were as follows:

  • Protein content ranged between 8.3% - 15.3%.
  • The reference method produced a mean result of 11.87%
  • Infratec grain analysers using the FOSS global ANN calibration gave a mean of 11.88%.
  • The Infratec with other locally adjusted ANN prediction models found the mean to be 11.92%.
  • So while locally adjusted calibration models deviated from the reference method by 0.05, the FOSS ANN calibration deviated by a mere 0.01.
  • In terms of reproducibility (standard deviation), both the reference methods and other ANN prediction models measured 0.21, while the FOSS ANN calibration was just 0.12, meaning results from the Infratec using the FOSS ANN calibration are even more reproducible than the reference method.  
  • Moisture content ranged between 12.2% - 15.0% with the reference method giving a mean of 13.87%.  
  • The adjusted ANN prediction models found the mean moisture to be 13.69, deviating from the reference methods by 0.18
  • The FOSS ANN calibration found this to be 13.74, 0.13 from the reference method 
  • The reproducibility (standard deviation) of the reference method for moisture was found to be 0.11, while the locally adjusted ANN prediction models found it to be 0.15, and  the FOSS ANN calibration registered this at 0.12.
The Infratec grain analysers and FOSS ANN calibration technology take into account a range of variables, including geography, grain type, temperature and instruments, and they cover a wide range of protein and moisture values.

It is these features that make the FOSS ANN calibration particularly robust and more effective than ANN prediction models that have been adjusted in an attempt to improve performance in a specific geographic region.

Newsletter sign-up

The latest products and news delivered to your inbox