The commercialization of Brazil nuts, seeds of Bertholletia excelsa Bonpl. (Lecythidaceae), represents one of the main income-generation activities for local and indigenous people from the Brazilian Amazon region. Because trees of B. excelsa grow and bear fruit almost exclusively in natural forests, information on their spatial distribution is crucial for nut harvest planning. However, this information is difficult to obtain with traditional approaches such as ground-based surveys. Here, we show the potential of convolutional neural networks (CNNs) and WorldView-3 satellite images (pixel size = 30 cm) to map individual tree crowns (ITCs) and groves of B. excelsa in Amazonian forests. First, we manually outlined B. excelsa ITCs in the WorldView-3 images using field-acquired geolocation information. Then, based on ITC boundaries, we sequentially extracted image patches and selected 80% of them for training and 20% for testing. We trained the DeepLabv3+ architecture with three backbones: ResNet-18, ResNet-50, and MobileNetV2. The average producer's accuracy was 93.87 ± 0.85%, 93.89 ± 1.6% and 93.47 ± 3.6% for ResNet-18, ResNet-50 and MobileNetV2, respectively. We then developed a new random patch extraction training strategy and assessed how a reduction in the percentage of training patches impacted the classification accuracy. To illustrate the robustness of the new training strategy, similar F1-scores were achieved whether 80% or 10% of the total number of patches were used to train the CNN model. By analyzing the feature maps derived from ResNet-18, we found that the shadow of emergent B. excelsa trees are important for their discrimination. Geometric distortions in the WorldView-3 images resulting from extreme off-nadir viewing angles compromise the presence of shadows, thus potentially hampering B. excelsa detection. Our results show that ITCs and groves of B. excelsa can be mapped by integrating CNNs and very-high-resolution (VHR) satellite images, paving the way for monitoring this important tree species in large tracts of Amazonian forests.

巴西坚果，Bertholletia excelsa Bonpl的种子的商业化。豆科（Lecythidaceae）是巴西亚马逊地区为当地和土著人民创造收入的主要活动之一。由于B. excelsa的树木几乎只能在天然林中生长并开花结果，因此有关其空间分布的信息对于坚果收获的计划至关重要。但是，使用地面调查等传统方法很难获得此信息。在这里，我们展示了卷积神经网络（CNN）和WorldView-3卫星图像（像素大小= 30厘米）可用于绘制亚马逊森林中B. excelsa的单个树冠（ITC）和树林的潜力。首先，我们手动概述B. excelsaWorldView-3图像中的ITC使用的是现场获取的地理位置信息。然后，基于ITC边界，我们依次提取了图像补丁，并选择了其中的80％用于训练和20％用于测试。我们使用三个主干网对DeepLabv3 +架构进行了培训：ResNet-18，ResNet-50和MobileNetV2。ResNet-18，ResNet-50和MobileNetV2的平均生产者准确度分别为93.87±0.85％，93.89±1.6％和93.47±3.6％。然后，我们开发了一种新的随机补丁提取训练策略，并评估了减少的训练补丁百分比如何影响分类准确性。为了说明新训练策略的鲁棒性，无论使用补丁总数的80％还是10％训练CNN模型，都获得了相似的F1分数。通过分析从ResNet-18派生的特征图，B. excelsa树对于它们的区分很重要。WorldView-3图像中的极低天底视角导致的几何失真会损害阴影的存在，从而可能妨碍B. excelsa的检测。我们的结果表明，可以通过整合CNN和超高分辨率（VHR）卫星图像来绘制B. excelsa的ITC和树丛，从而为在亚马逊大片森林中监测这一重要树种铺平道路。