Convolutional neural networks (CNNs) have recently emerged as a popular topic for machine learning in various academic and industrial fields. It is often an important problem to obtain a dataset with an appropriate size for CNN training. However, the lack of training data in the case of remote image research leads to poor performance due to the overfitting problem. In addition, the back-propagation algorithm used in CNN training is usually very slow and thus requires tuning different hyper-parameters. In order to overcome these drawbacks, a new approach fully based on machine learning algorithm to learn useful CNN features from Alexnet, VGG16, VGG19, GoogleNet, ResNet and SqueezeNet CNN architectures is proposed in the present study. This method performs a fast and accurate classification suitable for recognition systems. Alexnet, VGG16, VGG19, GoogleNet, ResNet and SqueezeNet pretrained architectures were used as feature extractors. The proposed method obtains features from the last fully connected layers of each architecture and applies the ReliefF feature selection algorithm to obtain efficient features. Then, selected features are given to the support vector machine classifier with the CNN-learned features instead of the FC layers of CNN to obtain excellent results. The effectiveness of the proposed method was tested on the UC-Merced dataset. Experimental results demonstrate that the proposed classification method achieved an accuracy rate of 98.76% and 99.29% in 50% and 80% training experiment, respectively.

中文翻译：

---

具有融合深度学习架构的遥感图像识别的高效深度特征选择

卷积神经网络 (CNN) 最近已成为各种学术和工业领域机器学习的热门话题。获得合适大小的数据集用于 CNN 训练通常是一个重要的问题。然而，在远程图像研究的情况下缺乏训练数据会由于过拟合问题导致性能不佳。此外，CNN 训练中使用的反向传播算法通常非常慢，因此需要调整不同的超参数。为了克服这些缺点，本研究提出了一种完全基于机器学习算法的新方法，可以从 Alexnet、VGG16、VGG19、GoogleNet、ResNet 和 SqueezeNet CNN 架构中学习有用的 CNN 特征。该方法执行适用于识别系统的快速准确的分类。Alexnet, VGG16, VGG19, GoogleNet、ResNet 和 SqueezeNet 预训练架构被用作特征提取器。所提出的方法从每个架构的最后一个全连接层中获取特征，并应用 ReliefF 特征选择算法来获得有效的特征。然后，将选定的特征提供给具有 CNN 学习特征的支持向量机分类器，而不是 CNN 的 FC 层，以获得出色的结果。在 UC-Merced 数据集上测试了所提出方法的有效性。实验结果表明，所提出的分类方法在50%和80%的训练实验中分别达到了98.76%和99.29%的准确率。所提出的方法从每个架构的最后一个全连接层中获取特征，并应用 ReliefF 特征选择算法来获得有效的特征。然后，将选定的特征提供给具有 CNN 学习特征的支持向量机分类器，而不是 CNN 的 FC 层，以获得出色的结果。在 UC-Merced 数据集上测试了所提出方法的有效性。实验结果表明，所提出的分类方法在50%和80%的训练实验中分别达到了98.76%和99.29%的准确率。所提出的方法从每个架构的最后一个全连接层中获取特征，并应用 ReliefF 特征选择算法来获得有效的特征。然后，将选定的特征提供给具有 CNN 学习特征的支持向量机分类器，而不是 CNN 的 FC 层，以获得出色的结果。在 UC-Merced 数据集上测试了所提出方法的有效性。实验结果表明，所提出的分类方法在50%和80%的训练实验中分别达到了98.76%和99.29%的准确率。在 UC-Merced 数据集上测试了所提出方法的有效性。实验结果表明，所提出的分类方法在50%和80%的训练实验中分别达到了98.76%和99.29%的准确率。在 UC-Merced 数据集上测试了所提出方法的有效性。实验结果表明，所提出的分类方法在50%和80%的训练实验中分别达到了98.76%和99.29%的准确率。