A CNN-Based Method of Vehicle Detection from Aerial Images Using Hard Example Mining
A CNN-Based Method of Vehicle Detection from Aerial Images Using Hard Example Mining
复制标题
DOI：
10.3390/rs10010124
复制
发表时间：
2018-01-01
期刊：
影响因子：
5
通讯作者：
Shibasaki, Ryosuke
中科院分区：
文献类型：
--
作者：
Koga, Yohei;Miyazaki, Hiroyuki;Shibasaki, Ryosuke
最近，深度学习技术在车辆检测中发挥了实际作用。虽然在将深度学习应用于车辆检测方面花费了大量精力，但尚未深入研究训练数据的有效使用，尽管它在改善训练结果方面具有很大的潜力，特别是在训练数据稀疏的情况下。在本文中，我们提出在卷积神经网络（CNN）的训练过程中使用硬示例挖掘（HEM）来进行航空图像中的车辆检测。我们将HEM应用于随机梯度下降（SGD），通过计算每批中的损失值并采用损失最大的示例来选择信息量最大的训练数据。我们在一次迭代中从500和1000个样本中挑选了100个样本进行训练，并测试了训练数据中正反样本的不同比例，以评估正反样本的平衡如何影响性能。在任何情况下，我们的方法总是优于普通SGD。来自纽约的图像的实验结果显示，与在普通SGD中训练的CNN相比，我们的方法的F1分数高出0.02。
Recently, deep learning techniques have had a practical role in vehicle detection. While much effort has been spent on applying deep learning to vehicle detection, the effective use of training data has not been thoroughly studied, although it has great potential for improving training results, especially in cases where the training data are sparse. In this paper, we proposed using hard example mining (HEM) in the training process of a convolutional neural network (CNN) for vehicle detection in aerial images. We applied HEM to stochastic gradient descent (SGD) to choose the most informative training data by calculating the loss values in each batch and employing the examples with the largest losses. We picked 100 out of both 500 and 1000 examples for training in one iteration, and we tested different ratios of positive to negative examples in the training data to evaluate how the balance of positive and negative examples would affect the performance. In any case, our method always outperformed the plain SGD. The experimental results for images from New York showed improved performance over a CNN trained in plain SGD where the F1 score of our method was 0.02 higher.