Classification of Venomous and Non-venomous Snakes Using Transfer Learning with MobileNetV2

Classification of snakes is the categorisation of various snake species based on their behaviour and physiological, morphological and genetic traits. It includes the organisation of snakes into groups of taxonomy and providing them scientific names based on their standardised classification system. Classification of snakes depends on numerous methods and data sources, including genetic analysis, behaviour observations, morphological examination, distribution patterns and ecological traits. Sequencing in DNA and advancement in molecular techniques have largely added to the improvement of the taxonomy of snakes. Classification of snakes is a fundamental feature of herpetology and grows the knowledge of evolution, snakebite incidents, snake diversity and conservation. It is a difficult task as there are a large number of snake species present. The paper consists of a machine learning approach that helps to classify snakes using Transfer Learning with MobileNetV2 and data augmentation. The dataset consists of training and testing data, giving an accuracy of 88.28%. There are several images in variable size in .jpeg format.