When investigating the vortex-induced vibration(VIV)of marine risers,extrapolating the dynamic response on the entire length based on limited sensor measurements is a crucial step in both laboratory experiments and fa...When investigating the vortex-induced vibration(VIV)of marine risers,extrapolating the dynamic response on the entire length based on limited sensor measurements is a crucial step in both laboratory experiments and fatigue monitoring of real risers.The problem is conventionally solved using the modal decomposition method,based on the principle that the response can be approximated by a weighted sum of limited vibration modes.However,the method is not valid when the problem is underdetermined,i.e.,the number of unknown mode weights is more than the number of known measurements.This study proposed a sparse modal decomposition method based on the compressed sensing theory and the Compressive Sampling Matching Pursuit(Co Sa MP)algorithm,exploiting the sparsity of VIV in the modal space.In the validation study based on high-order VIV experiment data,the proposed method successfully reconstructed the response using only seven acceleration measurements when the conventional methods failed.A primary advantage of the proposed method is that it offers a completely data-driven approach for the underdetermined VIV reconstruction problem,which is more favorable than existing model-dependent solutions for many practical applications such as riser structural health monitoring.展开更多
Human motion recognition is a research hotspot in the field of computer vision,which has a wide range of applications,including biometrics,intelligent surveillance and human-computer interaction.In visionbased human m...Human motion recognition is a research hotspot in the field of computer vision,which has a wide range of applications,including biometrics,intelligent surveillance and human-computer interaction.In visionbased human motion recognition,the main input modes are RGB,depth image and bone data.Each mode can capture some kind of information,which is likely to be complementary to other modes,for example,some modes capture global information while others capture local details of an action.Intuitively speaking,the fusion of multiple modal data can improve the recognition accuracy.In addition,how to correctly model and utilize spatiotemporal information is one of the challenges facing human motion recognition.Aiming at the feature extraction methods involved in human action recognition tasks in video,this paper summarizes the traditional manual feature extraction methods from the aspects of global feature extraction and local feature extraction,and introduces the commonly used feature learning models of feature extraction methods based on deep learning in detail.This paper summarizes the opportunities and challenges in the field of motion recognition and looks forward to the possible research directions in the future.展开更多
基金financially supported by the National Natural Science Foundation of China(Grant Nos.51109158,U2106223)the Science and Technology Development Plan Program of Tianjin Municipal Transportation Commission(Grant No.2022-48)。
文摘When investigating the vortex-induced vibration(VIV)of marine risers,extrapolating the dynamic response on the entire length based on limited sensor measurements is a crucial step in both laboratory experiments and fatigue monitoring of real risers.The problem is conventionally solved using the modal decomposition method,based on the principle that the response can be approximated by a weighted sum of limited vibration modes.However,the method is not valid when the problem is underdetermined,i.e.,the number of unknown mode weights is more than the number of known measurements.This study proposed a sparse modal decomposition method based on the compressed sensing theory and the Compressive Sampling Matching Pursuit(Co Sa MP)algorithm,exploiting the sparsity of VIV in the modal space.In the validation study based on high-order VIV experiment data,the proposed method successfully reconstructed the response using only seven acceleration measurements when the conventional methods failed.A primary advantage of the proposed method is that it offers a completely data-driven approach for the underdetermined VIV reconstruction problem,which is more favorable than existing model-dependent solutions for many practical applications such as riser structural health monitoring.
基金2021 Scientific research funding project of Liaoning Provincial Education Department(Research and implementation of university scientific research information platform serving the transformation of achievements).
文摘Human motion recognition is a research hotspot in the field of computer vision,which has a wide range of applications,including biometrics,intelligent surveillance and human-computer interaction.In visionbased human motion recognition,the main input modes are RGB,depth image and bone data.Each mode can capture some kind of information,which is likely to be complementary to other modes,for example,some modes capture global information while others capture local details of an action.Intuitively speaking,the fusion of multiple modal data can improve the recognition accuracy.In addition,how to correctly model and utilize spatiotemporal information is one of the challenges facing human motion recognition.Aiming at the feature extraction methods involved in human action recognition tasks in video,this paper summarizes the traditional manual feature extraction methods from the aspects of global feature extraction and local feature extraction,and introduces the commonly used feature learning models of feature extraction methods based on deep learning in detail.This paper summarizes the opportunities and challenges in the field of motion recognition and looks forward to the possible research directions in the future.
