Detection of Bovine Mastitis in Raw Milk, Using a Low-Cost NIR Spectrometer and k-NN Algorithm

Among the bovine diseases, mastitis causes high economic losses in dairy production system. Nowadays, detection under field conditions is mainly performed by California Mastitis Test, which considered de facto standard. However, this method presents with problems of slowness and expensiveness chemical-reactive process, deeply dependent on an expert’s trained eye and, consequently, highly imprecise. The aim work to propose a new for conditions. proposed uses low-cost, smartphone-connected NIR spectrometer solves aforementioned slowness, expert dependency disposability chemical methods. This spectra combination two k-Nearest Neighbors models. first model used detect presence while second classifies positive cases into weak strong. resulting was validated using leave-one-out technique where ground truth obtained Test. achieved accuracy 92.4%, one classifying severity showed 95%.