Automated Detection of Anemia in Small Ruminants Using Non-Invasive Visual Analysis Based on BIC Descriptor.

This study presents an automated method for the detection of anemia in small ruminants using non-invasive visual analysis of the ocular conjunctiva. The method involves image processing techniques and machine learning algorithms to extract relevant features and classify animals into different infection levels. The dataset consisted of photographs taken from real animals, with each animal evaluated based on the FAMACHA scale and hemoglobin level measurements. A total of 114 images were analyzed, including both sheep and goats, representing various stages of infection and distinct fur types. BIC (Border/Interior Classification) descriptor was used to extract information from the conjunctiva and two different classifiers were tested: SVM and KNN. The experimental results showed promising outcomes.