Published Paper Details:

Numerous human beings go to bird sanctuaries to get a relief from stress by seeing various birds and enjoying the beauty of colours and traits of the birds. For ensuring biodiversity, species information is a fundamental factor. In many scenarios, it is very difficult to identify the species of the birds due to the similarities existing in between the intraclass and interclass varieties of bird species. Recently, with the help of convolutional neural network (CNN), many state-of-the art algorithms on image classification achieved remarkable successes. This paper has been discussed to develop a deep learning platform to recognize the bird species by its image. Convolutional neural network is used for both the classification and prediction processes. In order to improve the feature extraction, a skip connection oriented neural network model is being introduced. The proposed method achieved 99.00% of classification accuracy for the training image. By using the proposed method, the amateur bird watchers can easily identify the bird species from the captured bird image.