卷积神经网络对SAR目标识别性能分析

A Performance Analysis of Convolutional Neural Network Models in SAR Target Recognition

近年来,以卷积神经网络(Convolutional Neural Network,CNN)为代表的深度学习方法在图像识别领域取得了巨大进展,但尚未在SAR目标识别领域得到广泛应用。基于此,将具有代表性的LeNet,AlexNet,VGGNet,GoogLeNet,ResNet,DenseNet,SENet等卷积神经网络模型应用到SAR图像目标识别上,并依据识别精度、模型尺寸、运行时间等指标在公开SAR数据集MSTAR上对9类目标进行识别实验。详细对比分析了不同CNN模型的综合性能,验证了利用CNN网络模型进行SAR图像目标识别的优越性,同时也为该领域的后续工作提供了参考基准。

In recent years,the deep learning methods represented By convolutional neural network(CNN)have made great progress in the field of image recognition,but still have not found wide applications in SAR target recognition.in this paper,the representative convolution neural network models,such as LeNet,AlexNet,VGGNet,GoogLeNet,ResNet,DenseNet,and SENet,are applied to SAR image target recognition.Recognition experiment of 9 types of target is carried out on the public dataset MSTAR.According to accuracy,model size,training time and other indicators,the performances of different CNN models are analyzed and compared.The superiority of CNN models in SAR image target recognition is verified.The study also provides a reference for the follow-up work in this field.