Classification of Tank Images Using Convolutional Neural Network
Ying Liu, Yongbin Yu, Lin Wang, Nyima Tashi, Nima Zhaxi, Hang Huang, Quanxin Deng · 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) · 2020
Recent years, deep learning has made huge progress in various fields including image classification. Battlefield target recognition is the foundation of battlefield situation research and tank is one of the most important targets of land battlefield. Tracked tank and wheeled tank which also known as wheeled vehicles are two kinds of tank, but there are a few researches on classification of tank. In this paper, a convolutional neural network (CNN) is utilized to classify tank images. The contribution of this work are as follows: First of all, we collected tank images and tag them. Secondly, we used CNN to classify wheeled tank and tracked tank, and the experiments show that the test accuracy achieves 0.93 with accuracy of tracked tank is 0.9268 and wheeled tank is 0.9368, reaching a good performance.