Classification of Animal based on FootPrint Using DeepLearning

Abstract: An emerging area in machine vision is a real dog biometric system that can identify and describe animal life in images and videos these programs offer methods for classifying animals using computer vision CNN features a well-liked deep learning technique are the foundation of the current system for classifying animal faces. Here, the suggested system analyses photos of animal footprints to categorise them using deep learning. Using a clever method, the footprint photos are pre-processed and turned into grayscale boundaries. Gabor filter are used to extract features of segmented image. The dimensionality reduction is carried out based on unsupervised model, Principal Component Analysis (PCA). The classification model is then fed with reduced feature vectors. The categorization and identification of the animal class is done using probabilistic neural networks, or CNNs.Footprints 0 dataset of five different animal categories of 100 images is to be used for classification. The performance analysis of the system is evaluated using the measure accuracy, precision, recall and fl-measure.

Keywords: Probabilistic Neural Network, precision, recall, F1score