Ship Model Recognition Based on Convolutional Neural Networks
Shihong Xing, Shaokang Zhang · 2018
Due to its advantages, the deep convolutional neural network has achieved good results in the field of target recognition and classification. It provides new approaches and methods for ship target recognition. Since the training of convolutional neural networks requires a large amount of data and high-performance hardware support, this paper proposes an efficient training scheme for convolutional neural networks training on the small capacity of ship data sets. The effectiveness of the scheme is verified through experiments, and the accuracy is improved. It can reach 99%. Finally, the feature map of convolutional neural network is analyzed, which further proves that the proposed scheme is effective.