Department of Computer Science, Florida State University, Tallahassee, FL 32306-4530, USA.
Department of Geography, Florida State University, Tallahassee, FL 32306-2190, USA.
Neural Netw. 2018 Sep;105:346-355. doi: 10.1016/j.neunet.2018.05.019. Epub 2018 Jun 2.
Environmental sustainability research is dependent on accurate land cover information. Even with the increased number of satellite systems and sensors acquiring data with improved spectral, spatial, radiometric and temporal characteristics and the new data distribution policy, most existing land cover datasets are derived from a pixel-based, single-date multi-spectral remotely sensed image with an unacceptable accuracy. One major bottleneck for accuracy improvement is how to develop an accurate and effective image classification protocol. By incorporating and utilizing multi-spectral, multi-temporal and spatial information in remote sensing images and considering the inherit spatial and sequential interdependence among neighboring pixels, we propose a new patch-based recurrent neural network (PB-RNN) system tailored for classifying multi-temporal remote sensing data. The system is designed by incorporating distinctive characteristics of multi-temporal remote sensing data. In particular, it uses multi-temporal-spectral-spatial samples and deals with pixels contaminated by clouds/shadow present in multi-temporal data series. Using a Florida Everglades ecosystem study site covering an area of 771 square kilometers, the proposed PB-RNN system has achieved a significant improvement in the classification accuracy over a pixel-based recurrent neural network (RNN) system, a pixel-based single-image neural network (NN) system, a pixel-based multi-image NN system, a patch-based single-image NN system, and a patch-based multi-image NN system. For example, the proposed system achieves 97.21% classification accuracy while the pixel-based single-image NN system achieves 64.74%. By utilizing methods like the proposed PB-RNN one, we believe that much more accurate land cover datasets can be produced over large areas.
环境可持续性研究依赖于准确的土地覆盖信息。即使有越来越多的卫星系统和传感器获取具有改进的光谱、空间、辐射和时间特征的数据，以及新的数据分发政策，大多数现有的土地覆盖数据集仍然是基于像素的、单一日期的多光谱遥感图像，精度无法接受。提高精度的一个主要瓶颈是如何开发准确有效的图像分类协议。通过整合和利用遥感图像中的多光谱、多时相和空间信息，并考虑到相邻像素之间固有的空间和顺序相关性，我们提出了一种新的基于斑块的递归神经网络（PB-RNN）系统，专门用于分类多时相遥感数据。该系统是根据多时相遥感数据的独特特征设计的。特别是，它使用多时相-光谱-空间样本，并处理多时间序列数据中存在的云/阴影污染的像素。使用一个覆盖 771 平方公里的佛罗里达大沼泽地生态系统研究站点，与基于像素的递归神经网络（RNN）系统、基于像素的单图像神经网络（NN）系统、基于像素的多图像 NN 系统、基于斑块的单图像 NN 系统和基于斑块的多图像 NN 系统相比，所提出的 PB-RNN 系统在分类精度上有了显著提高。例如，所提出的系统实现了 97.21%的分类精度，而基于像素的单图像 NN 系统实现了 64.74%。通过利用像 PB-RNN 这样的方法，我们相信可以在更大的区域内生成更准确的土地覆盖数据集。
