Association of total mixed ration particle fractions retained on the Penn State Particle Separator with milk, fat, and protein yield lactation curves at the cow level

G.Impoco ^a,∗, L. Tuminello^b

a -Image Analysis Group– CoRFiLaC (ConsorzioRicercaFilieraLattiero-Casearia), Ragusa,Italy
b -Microscopy Group– CoRFiLaC (ConsorzioRicercaFilieraLattiero-Casearia), Ragusa,Italy

Supervised learning approaches to image segmentation receive considerable interest due to their power and flexibility. However, the training phase is not painless, often being long and tedious. Accurate image labelling can takes ever al hours of expert operators’ valuable time. User interfaces are often specifically designed to assist the user for the task at hand. This is clearly unfeasible for most application domains. We propose a simple segmentation framework based on classification and supervised incremental learn-ing. A statistical model of pixel classes is learnt by incrementally adding new sample image patches to automatically-learned probability functions. Learning is iterated and refined in a number of steps rather than being executed in a one-shot training phase. We show that one-shot training and incremental labelling tend to produce similar statistical models, as the number of iterations grows. Comparable classification results are thus obtained with considerably less human effort.

Keywords: Image analysis, Segmentation, Incremental larning, Image labelling, Microscopy, Microstructure

Laura Tuminello

Tecnico

tuminello@corfilac.it