Automated Identification of Tree Species by Bark Texture Classification Using Convolutional Neural Networks

Abstract: Identification of tree species plays a key role in forestry related tasks like forest conservation, disease diagnosis and plant production. There had been debate regarding the part to be used for differentiation, whether it should leaves, fruits, flowers or bark. Studies have proven that bark is utmost importance as will present despite seasonal variations provides characteristic identity by structure. In this paper, deep learning based approach presented leveraging method computer vision classify 50 species, on basis texture using BarkVN-50 dataset. This maximum number trees being considered classification till now. A convolutional neural network(CNN), ResNet101 has implemented transfer-learning technique fine tuning maximise model performance. The produced an overall accuracy >94% during evaluation. performance validation done K-Fold Cross Validation testing unseen data collected from Internet, proved model's generalization capability real-world uses.