Supervised learning techniques are at the center of many tasks in remote sensing. Unfortunately, these methods, especially recent deep learning methods, often require large amounts of labeled data for training. Even though satellites acquire large amounts of data, labeling the data is often tedious, expensive and requires expert knowledge. Hence, improved methods that require fewer labeled samples are needed. We present MSMatch, the first semi-supervised learning approach competitive with supervised methods on scene classification on the EuroSAT and UC Merced Land Use benchmark datasets. We test both RGB and multispectral images of EuroSAT and perform various ablation studies to identify the critical parts of the model. The trained neural network achieves state-of-the-art results on EuroSAT with an accuracy that is up to 19.76% better than previous methods depending on the number of labeled training examples. With just five labeled examples per class, we reach 94.53% and 95.86% accuracy on the EuroSAT RGB and multispectral datasets, respectively. On the UC Merced Land Use dataset, we outperform previous works by up to 5.59% and reach 90.71% with five labeled examples. Our results show that MSMatch is capable of greatly reducing the requirements for labeled data. It translates well to multispectral data and should enable various applications that are currently infeasible due to a lack of labeled data. We provide the source code of MSMatch online to enable easy reproduction and quick adoption.
翻译:不幸的是,这些方法,特别是最近的深层学习方法,往往需要大量的标签数据来进行培训。即使卫星获得大量数据,但数据标签往往乏味、昂贵且需要专家知识。因此,需要改进方法,需要较少标签样本。我们介绍欧洲卫星组织和UC Merced Lands使用基准数据集的首个半监督学习方法,在现场分类方面,以监督方法进行竞争。我们测试欧洲卫星组织和UC Merced Landed Lands Uders 基准数据集的RGB和多谱段图像,并进行各种升级研究,以确定模型的关键部分。即使卫星获得大量数据,但经过培训的神经网络在EuroSAT上取得最先进的结果,其准确性达到19.76%,这取决于标签培训实例的数量。我们提供了第一批只有5个标签的例子,我们分别达到94.53%和95.86%的欧洲卫星组织RGB和多谱路段基准数据集的精确度。在UCUC Merced Luse数据集方面,我们比先前的工作要快到5.59 %,而我们的数据应用率要大大降低到90.71%。