Application Of Deep Learning For Bird Species Classification
Bird species travel throughout the world and we find wide variety of bird species in our day-to-day life. Each Bird species has different size, colour, shape and sound in different scenarios. Classification of these species has been a challenge. Identifying the bird species using audio classification has generated satisfying results but we believe images show good variations to identify the bird species. In this paper Deep Convolution Neural Network is used for classification. The images are transformed into gray scale and multiple nodes of comparison are generated by using tensor flow. The score sheet is obtained by testing and comparing these different nodes. The bird species are predicted by using highest score. We are using dataset Caltech-UCSD birds 200 [CUB-200-2011] and satisfactory results are achieved by the algorithm with an accuracy of bird species identification ranging from 80% to 90%.