Use of Very High Spatial Resolution Commercial Satellite Imagery and Deep Learning to Automatically Map Ice-Wedge Polygons across Tundra Vegetation Types.
Use of Very High Spatial Resolution Commercial Satellite Imagery and Deep Learning to Automatically Map Ice-Wedge Polygons across Tundra Vegetation Types.
复制标题
DOI：
10.3390/jimaging6120137
复制
发表时间：
2020-12-11
影响因子：
3.2
通讯作者：
Liljedahl AK
中科院分区：
文献类型：
--
作者：
Bhuiyan MAE;Witharana C;Liljedahl AK
我们开发了一种高通量测绘工作流程，以高性能分布式计算资源上的深度学习 (DL) 卷积神经网络 (CNN) 算法为中心，自动表征亚米分辨率商业卫星图像中的冰楔多边形 (IWP)。我们应用基于区域的 CNN 对象实例分割算法（即 Mask R-CNN）来自动检测和分类阿拉斯加北坡的 IWP。我们研究的中心目标是系统地阐述 DLCNN 模型在不同苔原类型（莎草、草丛莎草和非草丛莎草）和图像场景复杂性之间的互操作性，以加深对区域尺度测绘应用的机遇和挑战的理解。我们验证了定量误差统计数据以及详细的目视检查，以衡量 IWP 检测的准确性。我们发现对于具有不同苔原类型的所有候选图像场景，模型性能很有前景（检测精度：89% 至 96%，分类精度：94% 至 97%）。绘图工作流程通过表现出较低的绝对平均相对误差 (AMRE) 值 (0.17–0.23) 来识别 IWP。结果进一步表明，在实践迁移学习策略以绘制跨异质苔原覆盖类型的 IWP 时，增加训练样本的可变性非常重要。总体而言，我们的研究结果证明了 IWP 测绘工作流程在多个苔原景观中的稳健性能。
We developed a high-throughput mapping workflow, which centers on deep learning (DL) convolutional neural network (CNN) algorithms on high-performance distributed computing resources, to automatically characterize ice-wedge polygons (IWPs) from sub-meter resolution commercial satellite imagery. We applied a region-based CNN object instance segmentation algorithm, namely the Mask R-CNN, to automatically detect and classify IWPs in North Slope of Alaska. The central goal of our study was to systematically expound the DLCNN model interoperability across varying tundra types (sedge, tussock sedge, and non-tussock sedge) and image scene complexities to refine the understanding of opportunities and challenges for regional-scale mapping applications. We corroborated quantitative error statistics along with detailed visual inspections to gauge the IWP detection accuracies. We found promising model performances (detection accuracies: 89% to 96% and classification accuracies: 94% to 97%) for all candidate image scenes with varying tundra types. The mapping workflow discerned the IWPs by exhibiting low absolute mean relative error (AMRE) values (0.17–0.23). Results further suggest the importance of increasing the variability of training samples when practicing transfer-learning strategy to map IWPs across heterogeneous tundra cover types. Overall, our findings demonstrate the robust performances of IWPs mapping workflow in multiple tundra landscapes.
登录
查看更多内容
影响因子：
10.6
作者：
Borkar, Tejas S.;Karam, Lina J.
通讯作者：
Karam, Lina J.
影响因子：
3.2
作者：
Bhuiyan MAE;Witharana C;Liljedahl AK;Jones BM;Daanen R;Epstein HE;Kent K;Griffin CG;Agnew A
通讯作者：
Agnew A
影响因子：
4.9
作者：
TURNER, FJ;GRIGGS, DT;HEARD, H
通讯作者：
HEARD, H
影响因子：
4.5
作者：
Galleguillos, Carolina;Belongie, Serge
通讯作者：
Belongie, Serge
影响因子：
5
作者：
Buchhorn, Marcel;Walker, Donald A.;Schwieder, Marcel
通讯作者：
Schwieder, Marcel