Bird Species Classification from an Image Using VGG-16 Network

Abstract

Birds are an integral part of any environment and they are of the utmost importance to nature. Considering this, it is clear how necessary it is to be able to identify birds in the wilderness. This paper proposes a Machine Learning approach to identify Bangladeshi birds according to their species. We used VGG-16 network as our model to extract the features from bird images. In order to perform the classification, we used a data set that contains pictures of different bird species of Bangladesh which were used as they are, without any annotation. We then used various classification methods, where each method gave us different results. However, compared to other classification methods such as Random Forest and K-Nearest Neighbor (KNN), Support Vector Machine (SVM) gave us the maximum accuracy of 89%.