基于对比学习的终端区相似气象场景识别OACSTPCD
Recognition of Similar Weather Scenarios in Terminal Area Based on Contrastive Learning
为了提高终端区相似气象场景(Similar weather scenarios,SWSs)的识别准确率,提出一种基于对比学习的SWS(SWS based on contrastive learning,SWS?CL)识别模型.首先,针对对流天气图像特点,设计了一种数据增强方法来增加天气图像样本的数量和质量.接着,设计了一种对比损失函数使向量表征空间中的正样本与锚点样本之间的距离更近,而负样本与锚点样本之间的距离更远,进而基于对比学习技术在无标记样…查看全部>>
In order to improve the recognition accuracy of similar weather scenarios(SWSs) in terminal area,a recognition model for SWS based on contrastive learning (SWS-CL) is proposed. Firstly,a data augmentation method is designed to improve the number and quality of weather scenarios samples according to the characteristics of convective weather images. Secondly,in the pre-trained recognition model of SWS?CL,a loss function is formulated to minimize the distance b…查看全部>>
陈海燕;刘振亚;周逸;袁立罡
南京航空航天大学计算机科学与技术学院,南京 211106,中国软件新技术与产业化协同创新中心,南京 210023,中国南京航空航天大学计算机科学与技术学院,南京 211106,中国南京航空航天大学民航学院,南京 211106,中国
信息技术与安全科学
空中交通管制终端区相似气象场景图像识别对比学习
air traffic controlterminal areasimilar weather scenarios(SWSs)image recognitioncontrastive learning
《南京航空航天大学学报（英文版）》 2022 (4)
425-433,9
This work was supported by the Fun?damental Research Funds for the Central Universities(NOS.NS2019054,NS2020045).
评论