科研动态Research News | 我院丁浒特聘副研究员在地球科学与遥感领域TOP期刊发表学术论文
澳门十大正规老牌网赌/十大网赌网址信誉官网2023-09-06 14:29:00来源:学院评论:0点击:收藏本文
我院丁浒特聘副研究员在地球科学与遥感领域TOP期刊International Journal of Digital Earth发表学术论文
Dr. Hu Ding has published his research in International Journal of Digital Earth, a top journal in Earth Science and Remote Sensing.
作者简介
丁浒,澳门十大正规老牌网赌特聘副研究员。主要研究领域包括数字地形分析,数字地貌制图等方面。主持国家自然科学基金,广东省自然科学基金,中国博士后科学基金等项目,近年来相关成果发表在Transactions in GIS, Remote Sensing, Geomorphology等国内外权威期刊。
近日,澳门十大正规老牌网赌丁浒特聘副研究员以通讯作者在地球科学和遥感领域TOP期刊International Journal of Digital Earth发表研究论文Height estimation from single aerial imagery using contrastive learning based multi-scale refinement network。
(https://doi.org/10.1080/17538947.2023.2225881)
当前,基于航空影像进行高度估计已在定位导航、地图生产和三维目标检测等应用中起到重要作用。目前,利用深度卷积神经网络从单目遥感影像进行高度信息预测已得到广泛的应用,但该类方法通常需要大量的训练数据,并且经常忽略正射影像中的形状特征,使得预测结果产生一定的误差。为了较好解决这些问题,本研究提出了一种基于梯度的自监督学习网络,加入了动量对比损失,以从预训练阶段的非标记图像中提取形状信息。同时,所提出的监督网络在多个解码阶段使用本地隐式约束层,以提炼高分辨率特征进行高度估计。此外,本研究利用结构感知损失用以提高网络对边界区域的位置偏移和针对轻微结构变化的鲁棒性。采用ISPRS基准数据集对所提出的网络进行效果评估,实验结果显示所提出的方法优于其他基准网络,Vaihingen数据集的最小MAE和RMSE分别为0.116和0.289,Potsdam数据集的最小MAE和RMSE分别为0.077和0.481。此外,所提出的方法在Potsdam数据集上显示出大约三倍的数据效率提升,并在Enschede数据集上进行了域泛化。实验结果证明了所提出的方法在基于单目遥感影像进行高度估计方面的有效性。
International Journal of Digital Earth是地球科学领域的国际权威期刊,2022年影响因子为5.1 (中科院1区TOP)。
Height map estimation from a single aerial image plays a crucial role in localization, mapping, and 3D object detection. Deep convolutional neural networks have been used to predict height information from single-view remote sensing images, but these methods rely on large volumes of training data and often overlook geometric features present in orthographic images. To address these issues, this study proposes a gradient-based self-supervised learning network with momentum contrastive loss to extract geometric information from non-labeled images in the pretraining stage. Additionally, novel local implicit constraint layers are used at multiple decoding stages in the proposed supervised network to refine high-resolution features in height estimation. The structural-aware loss is also applied to improve the robustness of the network to positional shift and minor structural changes along the boundary area. Experimental evaluation on the ISPRS benchmark datasets shows that the proposed method outperforms other baseline networks, with minimum MAE and RMSE of 0.116 and 0.289 for the Vaihingen dataset and 0.077 and 0.481 for the Potsdam dataset, respectively. The proposed method also shows around threefold data efficiency improvements on the Potsdam dataset and domain generalization on the Enschede datasets. These results demonstrate the effectiveness of the proposed method in height map estimation from single-view remote sensing images.
图1.本研究提出的高度估计网络框架
图2.本文方法在Potsdam数据集上的部分实验结果与其他基准网络实验结果对比(从左至右依次为:航空影像, U-IMG2DSM, DCNet, PLNet, 本文方法和参考真值)
图3.本文方法在Vaihingen数据集上的部分实验结果与其他基准网络实验结果对比(从左至右依次为:航空影像, U-IMG2DSM, DCNet, PLNet, 本文方法和参考真值)
文字 | 丁 浒
初审 | 袁亚娟
复审 | 陶 伟
终审 | 刘云刚
标签: