Strip segmentation of oceanic internal waves in SAR images based on TransUNet
作者机构:College of Ocean Science and EngineeringShanghai Maritime UniversityShanghai 201306China Translational Research Institute of Brain and Brain-Like IntelligenceShanghai Fourth People’s HospitalSchool of MedicineTongji UniversityShanghai 200434China
出 版 物:《Acta Oceanologica Sinica》 (海洋学报(英文版))
年 卷 期:2023年第42卷第10期
页 面:67-74页
核心收录:
学科分类:083002[工学-环境工程] 0830[工学-环境科学与工程(可授工学、理学、农学学位)] 081802[工学-地球探测与信息技术] 0707[理学-海洋科学] 08[工学] 0818[工学-地质资源与地质工程] 0816[工学-测绘科学与技术] 081602[工学-摄影测量与遥感]
基 金:The National Natural Science Foundation of China under contract No.51679132 the Science and Technology Commission of Shanghai Municipality under contract Nos.21ZR1427000 and 17040501600
主 题:oceanic internal waves deep learning stripe segmentation synthetic aperture radar TransUNet
摘 要:The development of oceanic remote sensing artificial intelligence has made possible to obtain valuable information from amounts of massive *** internal waves play a crucial role in oceanic *** obtain oceanic internal wave stripes from synthetic aperture radar(SAR)images,a stripe segmentation algorithm is proposed based on the TransUNet framework,which is a combination of U-Net and Transformer,which is also *** adjusting the number of Transformer layer,multi-layer perceptron(MLP)channel,and Dropout parameters,the influence of over-fitting on accuracy is significantly weakened,which is more conducive to segmenting lightweight oceanic internal *** results show that the optimized algorithm can accurately segment oceanic internal wave ***,the optimized algorithm can be trained on a microcomputer,thus reducing the research *** proposed algorithm can also change the complexity of the model to adapt it to different date ***,TransUNet has immense potential for segmenting oceanic internal waves.