ANN is the ideal technology for generating robust NIR calibrations. ANN technology has been tried, tested and successfully used by the grain industry for many years. USDA-GIPSA has officially approved this technology, and made it the benchmark for grain analysis for payment systems.
At FOSS we know that calibration development never stops; it is an ongoing effort with continuous calibration updates. Our many upgrades have resulted in very wide, robust calibration models that will save significant costs, right from day one and for as long as you care to utilise them. The ability to cope with non-linear relationships makes ANN particularly suitable for complex applications such as feed, pet food and forage.
ANN features include:
Improved accuracy of up to 10-15% compared to PLS calibrations
Fewer calibrations needed to cover a wide range products and raw materials
Increased robustness and less work in adjusting and updating calibrations
Calibrations can be used from day one with only minor adjustments
Easier transferability between instruments
Accurate and reliable global calibrations