Processamento digital de imagens para análise do comportamento e bem-estar de frangos de corte

Abstract

The present research was conducted with the main objective of using computational vision techniques to develop a program that could identify, qualify and quantify the frequency of ingestive behavior expressed by broilers. And analyze the differences between the ingestive behavior regarding the factors age, production environment, sexing and lighting system, using stochastic tools of multivariate analysis. The validation of the software was performed by the result obtained from the visual analysis of a specialist, by means of linear regression analysis by the R platform. Based on the analyzes, it was verified that the coefficient of determination (R²) ranged from 0.74 to 0.97, for all validation events of the software, which shows good characterization of the ingestive behavior of broiler chickens by the computational view. Feeding behavior of broilers was influenced by the environment, especially in the morning and afternoon. The night was most evident the influence of the white LED lighting system, by the greater weight gain presented by the birds. The birds submitted to the lighting systems did not present behavioral changes during the day, because the incidence of solar radiation is greater than the illuminance promoted by the lighting system. The variation of the weight gain and live weight of the birds presented better response when the birds were exposed to a temperature of 27ºC, that is, 3 °C above the comfort condition recommended, for the 5th week of the breeding cycle.