ABSTRACT

Poplars are one of the most widespread fast-growing tree species used for forest plantations. Owing to their distinct features (fast growth and short rotation) and the dependency on the timber price market, poplar plantations are characterized by large inter-annual fluctuations in their extent and distribution. Therefore, monitoring poplar plantations requires a frequent update of information – not feasible by National Forest Inventories due to their periodicity – achievable by remote sensing systems applications. In particular, the new Sentinel-2 mission, with a revisiting period of 5 days, represents a potentially efficient tool for meeting this need.

In this paper, we present a deep learning approach for mapping poplar plantations using Sentinel-2 time series. A reference dataset of poplar plantations was available for a large study area of more than 46,000 km² in Northern Italy and served as training and testing data. Two classification methods were compared: (1) a fully connected neural network (also called multilayer perceptron), and (2) a traditional logistic regression. The performance of the two approaches was estimated through bootstrapping procedure with a confidence interval of 99%. Results indicated for deep learning an omission error rate of 2.77%±2.76%, showing improvements compared to logistic regression, omission error rate = 8.91%±4.79%.

摘要

杨树是用于人工林的最广泛的速生树种之一。由于其独特的特征（快速生长和短轮作）以及对木材价格市场的依赖，杨树人工林的面积和分布具有较大的年际波动。因此，监测杨树人工林需要频繁更新信息——由于其周期性，国家森林清单不可行——可通过遥感系统应用实现。特别是新的 Sentinel-2 任务，重访期为 5 天，是满足这一需求的潜在有效工具。

在本文中，我们提出了一种使用 Sentinel-2 时间序列绘制杨树种植园的深度学习方法。意大利北部超过 46,000 km ²的大型研究区提供了杨树种植园的参考数据集，可用作训练和测试数据。比较了两种分类方法：（1）完全连接的神经网络（也称为多层感知器），以及（2）传统的逻辑回归。这两种方法的性能是通过引导程序估计的，置信区间为 99%。结果表明，深度学习的遗漏错误率为 2.77%±2.76%，与逻辑回归相比有所改进，遗漏错误率为 8.91%±4.79%。