Automated Methods to Identify Snake Species in Sri Lanka: A Review

Abstract

Snake bites in Sri Lanka cause death to nearly 100 people each year. Among the many reasons for this condition is the inability of people to identify the snake type which prevents administering the appropriate anti venom treatment. Misidentification of snakes also causes threats to the existence of harmless snakes that contributes to the biodiversity of reptile species. A conducted with 223 participants to ascertain the ability of people to correctly identify the snake type when an image of a snake is available revealed that the majority the participants was unable to recognize the snake type This paper presents some of the survey results and review of various methods such as k-nearest (KNN), Support Vector Machines (SVM), Image Processing techniques, Probabilistic Graphical Models, Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN) which are used to automatically identify objects birds, marine species, humans and animals that could be applied for snake recognition to assist people identifying snake types which can contribute in reducing morbidity and mortality due to snake bites as well as to minimize the harm caused to innocent snake types