Machine learning study demonstrates 71.66% accuracy in classifying dorsal fin types from blue whale images, highlighting potential for automated photo-identification.
The process of photo-identification images of the blue whale (Balaenoptera musculus), is made manually; this process classifies images blue whale through its dorsal fin characteristics.The features are extracted visually, which can generate errors at the moment of classified.In this work an image classifier blue whale is presented, which have features such as color pigmentation, background image, type of dorsal fin, among others; these common characteristics generate high statistical dependence.This statistical dependence causes the data extracted through a segmented image of the blue whale, are to be observed through a hyperplane.Using statistical techniques and the K-Nearest Neighbors classifier, the classification of three types of dorsal fin is obtained with an accuracy of 71.66%, assessing the value of K = 7, taking reference catalog of species CICIMAR-IPN.
No takes yet. Share an insight, caveat, or question.
Ramos-Arredondo et al. (2014) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: