Classification of color texture feature patterns extracted from cells in histological images of fish ovary

TítuloClassification of color texture feature patterns extracted from cells in histological images of fish ovary
AutoresM. Fernández-Delgado, E. Cernadas and M. Pérez-Ortiz
TipoCapítulo de libro
Fonte Ensemble Classification Methods with Applications in R, John Wiley and Sons, pp. 119-134 , 2018.
ISBN978-1-119-42109-2
AbstractThe assessment of fecundity is fundamental in the study of biology and to define the management of sustainable fisheries. Stereometry (Emerson et al., 1990) is an accurate method to estimate fecundity from histological images, although it is rarely developed because it requires a lot of time from specialized technicians. Figure 6.1 shows some histological images of fish species Merluccius merluccius (also named European hake). The fecundity estimation requires the measurement of the diameter of matured cells (oocytes), so they must be detected and classified according to the presence or absence of nucleus, and according to the stage of development: cortical alveoli, hydrated or vitellogenic/atretic. The software Govocitos1 (González-Rufino et al., 2013) supports technicians in fecundity studies using histological images of fish ovary, allowing oocyte detection (which can be interactively corrected by the user) and classification (accord- ing to the presence/absence of nucleus and the development stage), and using the oocyte measurements to automatically estimate the fish fecundity.
Palabras chaveAdabag, Adaboost, texture analysis, ordinal classification, fish oocytes, ensembles