The Significance Of Artificial Intelligence In Arabian Horses Identification System

Over the past decade, increasing attention has been drawn to the application of artificial intelligence (AI) in agriculture market and veterinary-sciences. This paper introduces a new approach for Arabian-horses identification system using their iris significant features and artificial intelligence technology. These features are the retina cogina and the pupil shape. Arabian horses are identified here by using the segmented retina cogina with the pupil region out of iris images using color based k-means clustering. The identification system has been done using different AI algorithms including Artificial Neural Networks (ANN), and Support Vector Machines (SVM) with Scale Invariant Feature Transform (SIFT) and Speeded Up Robust Features (SURF) were tested in the experiments. Moreover, deep learning approach of deep convolutional neural network (CNN) was also implemented. However, segmented features images were given as input to deep convolutional neural network. The experiments were conducted on collected data set of horses' eyes images of 145 Arabian horses. The results of the conducted experiments shows that SVM with SIFT features achieved an accuracy of 97.6%.