科研动态Research News ｜ 刘亦书教授在遥感领域TOP期刊IEEE TGRS发表学术论文

以下文章来源于华南师范大学地理科学学院 ，作者地理科学学院

刘亦书教授在遥感领域TOP期刊

IEEE TGRS发表学术论文

 Yishu Liu published academic papers in IEEE TGRS, a top journal in Geoscience and remote sensing

近日，华南师范大学地理科学学院刘亦书教授以第一作者身份在遥感领域TOP期刊《IEEE Transactions on Geoscience and Remote Sensing》发表题为“Integrating Knowledge Distillation With Learning to Rank for Few-Shot Scene Classification”的研究论文(https://ieeexplore.ieee.org/abstract/document/9487012)。

小样本学习在遥感图像自动解译中具有重要的意义，该论文研究小样本遥感场景分类问题。目前大多数前沿的小样本学习模型均采用所谓的“情景训练”模式——通过将训练样本分成支撑图像和查询图像，模仿了真实的小样本分类情景。然而，这些模型只判断支撑图像和查询图像是否来自同一个类别。这种二值预测过于粗略，故作者设立一个更为苛刻的训练目标，“强迫”网络模型根据支查相似度（即支撑图像和查询图像之间的相似度）对所有支撑图像进行排序，因而赋予模型更强的泛化能力。更具体地说，作者提出“保序知识蒸馏”方法，鼓励学生网络模仿教师网络的排序行为；进而利用Plackett–Luce模型，构造出一种新的蒸馏损失函数，并提出一种新的小样本学习模型——排序网络。作者在两个公开的遥感场景图像库上进行深入全面的实验，实验结果表明，排序网络的分类性能优于现有方法。

排序网络架构图

该研究得到国家自然科学基金项目（61673184）的资助。《IEEE TGRS》是地球科学和遥感领域的顶级期刊，也是IEEE地球科学与遥感协会（GRSS）的会刊，最新影响因子为8.125。

Few-shot learning (FSL) has great potential for automatic interpretation of remote sensing (RS) images. In this article, we make a study of few-shot RS scene classification. Many recently proposed FSL models adopt a very effective episode-based training procedure, where each episode is contrived to mimic the few-shot task by dividing training examples into support images and query images. However, these models can only judge whether or not a support image has the same class membership as a given query image. Such a yes-or-no prediction is rather rough, so we set a more demanding training objective, enforcing FSL models to rank support images according to support-query similarity and, hence, endowing them with better generalization ability. To this end, first, we propose ranking-preserving knowledge distillation (KD), which encourages a student network to rank support images in the same way as teacher networks. Then, integrating multi-teacher KD with learning to rank, we construct a novel distillation loss using the Plackett–Luce distributions and build a novel few-shot classification model called ranking network. Extensive evaluation on two public RS datasets shows that the ranking network achieves the state of the art by a wide margin.

转自：“生态遥感前沿”微信公众号

如有侵权，请联系本站删除！