RFID-based human activity recognition(HAR)attracts attention due to its convenience,noninvasiveness,and privacy protection.Existing RFID-based HAR methods use modeling,CNN,or LSTM to extract features effectively.Still...RFID-based human activity recognition(HAR)attracts attention due to its convenience,noninvasiveness,and privacy protection.Existing RFID-based HAR methods use modeling,CNN,or LSTM to extract features effectively.Still,they have shortcomings:1)requiring complex hand-crafted data cleaning processes and 2)only addressing single-person activity recognition based on specific RF signals.To solve these problems,this paper proposes a novel device-free method based on Time-streaming Multiscale Transformer called TransTM.This model leverages the Transformer's powerful data fitting capabilities to take raw RFID RSSI data as input without pre-processing.Concretely,we propose a multiscale convolutional hybrid Transformer to capture behavioral features that recognizes singlehuman activities and human-to-human interactions.Compared with existing CNN-and LSTM-based methods,the Transformer-based method has more data fitting power,generalization,and scalability.Furthermore,using RF signals,our method achieves an excellent classification effect on human behaviorbased classification tasks.Experimental results on the actual RFID datasets show that this model achieves a high average recognition accuracy(99.1%).The dataset we collected for detecting RFID-based indoor human activities will be published.展开更多
Due to the rapid urbanization and industrialization,and excessive consumption of fossil fuel,haze weather characterized by PM2.5 has become a severe pollution problem in major Chinese cities recently,which has harmful...Due to the rapid urbanization and industrialization,and excessive consumption of fossil fuel,haze weather characterized by PM2.5 has become a severe pollution problem in major Chinese cities recently,which has harmful effect on the air quality,visibility,clime system and human health.To indicate suitable directions for the prevention and control of haze pollution,the pollutant source apportionment and formation mechanism of urban haze should be figured out firstly.In this work,we briefly review the frequently-used methods for PM2.5 source apportionment and formation mechanism analysis of urban haze based on normal perspective in recent years.Furthermore,based on the new perspective of systematic methodology,the utilization of fault tree approach for the causation mechanism analysis of urban haze is significantly introduced and discussed.Finally,the recent progress on controlling strategies of urban haze in China is also synoptically introduced and discussed.It is expected that more effective tool/method can be found,developed and employed for the causation mechanism analysis and risk management of urban haze in China.展开更多
Objective This study elucidated the function and role of SUMOylation in type Ⅰ endometrial carcinoma.Methods Fifty type Ⅰ endometrial carcinoma cases and para-cancer tissue samples were collected.The expression leve...Objective This study elucidated the function and role of SUMOylation in type Ⅰ endometrial carcinoma.Methods Fifty type Ⅰ endometrial carcinoma cases and para-cancer tissue samples were collected.The expression levels of ubiquitin-conjugating enzyme E2I(Ube2 i,Ubc9) and small ubiquitin-like modifier 1(SUMO1)/sentrin-specific peptidase 1(SENP1) proteins were examined using immunohistochemistry and the correlation with clinicopathological parameters was analyzed.Results Ubc9 expression in type Ⅰ endometrial carcinoma tissues was significantly higher than that in the para-cancer tissues;in contrast,the expression of the SENP1 protein was markedly lower than that in the para-cancer tissues.Ubc9 and SENP1 expression levels were negatively correlated and were associated with tumor differentiation,but not age,depth of invasion,tumor stage,and lymph node metastasis.Conclusion SUMOylation modification plays a major role in the pathogenesis and development of type Ⅰ endometrial carcinoma.Thus,it could be a potential target for the treatment of endometrial cancer.展开更多
Split ring resonators (SRRs)-based broadband metamaterial filters have attracted considerable attention due to their great prospect of practical applications. These filters had been usually obtained by stacking mult...Split ring resonators (SRRs)-based broadband metamaterial filters have attracted considerable attention due to their great prospect of practical applications. These filters had been usually obtained by stacking multiple different-sized metallic patterns, making their fabrication quite troublesome. Herein, we presented a simple design of broadband filter composed of two nested SRRs. The resonance bandwidth of the metamatefial filter gradually increased with the decrease of the ann length of the inner SRR. The increase in the resonance bandwidth was attributed to the increase in the radiation of the entire structure. Moreover, the bandwidth of the metamaterial can be further broadened by decreasing the period of the structure. The proposed filter provides a meaningful way toward expanding the bandwidth operating range from narrowband to broadband in an effective way.展开更多
Reading text in images automatically has become an attractive research topic in computer vision.Specifically,end-to-end spotting of scene text has attracted significant research attention,and relatively ideal accuracy...Reading text in images automatically has become an attractive research topic in computer vision.Specifically,end-to-end spotting of scene text has attracted significant research attention,and relatively ideal accuracy has been achieved on several datasets.However,most of the existing works overlooked the semantic connection between the scene text instances,and had limitations in situations such as occlusion,blurring,and unseen characters,which result in some semantic information lost in the text regions.The relevance between texts generally lies in the scene images.From the perspective of cognitive psychology,humans often combine the nearby easy-to-recognize texts to infer the unidentifiable text.In this paper,we propose a novel graph-based method for intermediate semantic features enhancement,called Text Relation Networks.Specifically,we model the co-occurrence relationship of scene texts as a graph.The nodes in the graph represent the text instances in a scene image,and the corresponding semantic features are defined as representations of the nodes.The relative positions between text instances are measured as the weights of edges in the established graph.Then,a convolution operation is performed on the graph to aggregate semantic information and enhance the intermediate features corresponding to text instances.We evaluate the proposed method through comprehensive experiments on several mainstream benchmarks,and get highly competitive results.For example,on the SCUT-CTW1500,our method surpasses the previous top works by 2.1%on the word spotting task.展开更多
Gill morphological changes and physiological responses in juvenile large yellow croaker(Larimichthys crocea)were examined upon exposure to low salinity after indoor culture of the fish at salinities of 2,4,6,8,and 24...Gill morphological changes and physiological responses in juvenile large yellow croaker(Larimichthys crocea)were examined upon exposure to low salinity after indoor culture of the fish at salinities of 2,4,6,8,and 24‰(control group).The thickness of the lamellae was significantly higher in the low-salinity groups than in the control group;in contrast,the interlamellar space was significantly lower in the low-salinity groups than in the control group.Additionally,a significant negative correlation was found between the thickness of lamellae and interlamellar spaces(P<0.01).Mitochondria-rich cells(MRCs)were mainly found in the filament at 24‰salinity and proliferated in the lamellae at lower salinities,suggesting that filament and lamellar MRCs are responsible for ion secretion and absorption,respectively.Meanwhile,the activity of Na+/K+-ATPase(NKA)was significantly elevated with the decrease in salinity from 24‰to 4‰(P<0.05),which was consistent with MRCs proliferation.Finally,the activity of NKA declined at 2‰salinity(P<0.05),indicating the limit of osmoregulation,which was consistent with the degeneration and necrosis of the lamellae.Additionally,different levels of gill histopathological lesions,including pavement cell(PVC)exfoliation,lamellar epithelial lifting,edema,fusion,aneurism,and necrosis,were observed from salinities of 24 to 4‰,likely reducing the respiratory efficiency and compromising the health of juvenile fish.In conclusion,large yellow croaker juveniles could improve the osmoregulatory capacity by increasing lamellar MRCs and NKA activity with the decrease in salinity from 24 to 4‰.However,the associated histopathological lesions are likely to negatively influence the fish by affecting respiration and osmoregulation,especially when the salinity is below 4‰.展开更多
基金the Strategic Priority Research Program of Chinese Academy of Sciences(Grant No.XDC02040300)for this study.
文摘RFID-based human activity recognition(HAR)attracts attention due to its convenience,noninvasiveness,and privacy protection.Existing RFID-based HAR methods use modeling,CNN,or LSTM to extract features effectively.Still,they have shortcomings:1)requiring complex hand-crafted data cleaning processes and 2)only addressing single-person activity recognition based on specific RF signals.To solve these problems,this paper proposes a novel device-free method based on Time-streaming Multiscale Transformer called TransTM.This model leverages the Transformer's powerful data fitting capabilities to take raw RFID RSSI data as input without pre-processing.Concretely,we propose a multiscale convolutional hybrid Transformer to capture behavioral features that recognizes singlehuman activities and human-to-human interactions.Compared with existing CNN-and LSTM-based methods,the Transformer-based method has more data fitting power,generalization,and scalability.Furthermore,using RF signals,our method achieves an excellent classification effect on human behaviorbased classification tasks.Experimental results on the actual RFID datasets show that this model achieves a high average recognition accuracy(99.1%).The dataset we collected for detecting RFID-based indoor human activities will be published.
基金Supported by the National Natural Science Foundation of China(21706029,21736004,21775022)the Research Fund Program of Guangdong Provincial Key Lab of Green Chemical Product Technology(GC201817)
文摘Due to the rapid urbanization and industrialization,and excessive consumption of fossil fuel,haze weather characterized by PM2.5 has become a severe pollution problem in major Chinese cities recently,which has harmful effect on the air quality,visibility,clime system and human health.To indicate suitable directions for the prevention and control of haze pollution,the pollutant source apportionment and formation mechanism of urban haze should be figured out firstly.In this work,we briefly review the frequently-used methods for PM2.5 source apportionment and formation mechanism analysis of urban haze based on normal perspective in recent years.Furthermore,based on the new perspective of systematic methodology,the utilization of fault tree approach for the causation mechanism analysis of urban haze is significantly introduced and discussed.Finally,the recent progress on controlling strategies of urban haze in China is also synoptically introduced and discussed.It is expected that more effective tool/method can be found,developed and employed for the causation mechanism analysis and risk management of urban haze in China.
文摘Objective This study elucidated the function and role of SUMOylation in type Ⅰ endometrial carcinoma.Methods Fifty type Ⅰ endometrial carcinoma cases and para-cancer tissue samples were collected.The expression levels of ubiquitin-conjugating enzyme E2I(Ube2 i,Ubc9) and small ubiquitin-like modifier 1(SUMO1)/sentrin-specific peptidase 1(SENP1) proteins were examined using immunohistochemistry and the correlation with clinicopathological parameters was analyzed.Results Ubc9 expression in type Ⅰ endometrial carcinoma tissues was significantly higher than that in the para-cancer tissues;in contrast,the expression of the SENP1 protein was markedly lower than that in the para-cancer tissues.Ubc9 and SENP1 expression levels were negatively correlated and were associated with tumor differentiation,but not age,depth of invasion,tumor stage,and lymph node metastasis.Conclusion SUMOylation modification plays a major role in the pathogenesis and development of type Ⅰ endometrial carcinoma.Thus,it could be a potential target for the treatment of endometrial cancer.
文摘Split ring resonators (SRRs)-based broadband metamaterial filters have attracted considerable attention due to their great prospect of practical applications. These filters had been usually obtained by stacking multiple different-sized metallic patterns, making their fabrication quite troublesome. Herein, we presented a simple design of broadband filter composed of two nested SRRs. The resonance bandwidth of the metamatefial filter gradually increased with the decrease of the ann length of the inner SRR. The increase in the resonance bandwidth was attributed to the increase in the radiation of the entire structure. Moreover, the bandwidth of the metamaterial can be further broadened by decreasing the period of the structure. The proposed filter provides a meaningful way toward expanding the bandwidth operating range from narrowband to broadband in an effective way.
文摘Reading text in images automatically has become an attractive research topic in computer vision.Specifically,end-to-end spotting of scene text has attracted significant research attention,and relatively ideal accuracy has been achieved on several datasets.However,most of the existing works overlooked the semantic connection between the scene text instances,and had limitations in situations such as occlusion,blurring,and unseen characters,which result in some semantic information lost in the text regions.The relevance between texts generally lies in the scene images.From the perspective of cognitive psychology,humans often combine the nearby easy-to-recognize texts to infer the unidentifiable text.In this paper,we propose a novel graph-based method for intermediate semantic features enhancement,called Text Relation Networks.Specifically,we model the co-occurrence relationship of scene texts as a graph.The nodes in the graph represent the text instances in a scene image,and the corresponding semantic features are defined as representations of the nodes.The relative positions between text instances are measured as the weights of edges in the established graph.Then,a convolution operation is performed on the graph to aggregate semantic information and enhance the intermediate features corresponding to text instances.We evaluate the proposed method through comprehensive experiments on several mainstream benchmarks,and get highly competitive results.For example,on the SCUT-CTW1500,our method surpasses the previous top works by 2.1%on the word spotting task.
基金This work was supported by the National Key R&D Program of China(2018YFD0900906)the AoShan Talents Cultivation Program Supported by the Qingdao National Laboratory for Marine Science and Technology(2017ASTCP-ES07)+3 种基金the Key Laboratory of Sustainable Development of Marine Fisheries,Ministry of Agriculture,P.R.China(2018HY-XKQ01)the STS project of Fujian Province(2017T3016)the Ningde Normal University Special Fund Program(2019ZX406)the Fujian Province Think Tank of Science and Technology(decision-making consulting)key projects(FJKX-A1618).
文摘Gill morphological changes and physiological responses in juvenile large yellow croaker(Larimichthys crocea)were examined upon exposure to low salinity after indoor culture of the fish at salinities of 2,4,6,8,and 24‰(control group).The thickness of the lamellae was significantly higher in the low-salinity groups than in the control group;in contrast,the interlamellar space was significantly lower in the low-salinity groups than in the control group.Additionally,a significant negative correlation was found between the thickness of lamellae and interlamellar spaces(P<0.01).Mitochondria-rich cells(MRCs)were mainly found in the filament at 24‰salinity and proliferated in the lamellae at lower salinities,suggesting that filament and lamellar MRCs are responsible for ion secretion and absorption,respectively.Meanwhile,the activity of Na+/K+-ATPase(NKA)was significantly elevated with the decrease in salinity from 24‰to 4‰(P<0.05),which was consistent with MRCs proliferation.Finally,the activity of NKA declined at 2‰salinity(P<0.05),indicating the limit of osmoregulation,which was consistent with the degeneration and necrosis of the lamellae.Additionally,different levels of gill histopathological lesions,including pavement cell(PVC)exfoliation,lamellar epithelial lifting,edema,fusion,aneurism,and necrosis,were observed from salinities of 24 to 4‰,likely reducing the respiratory efficiency and compromising the health of juvenile fish.In conclusion,large yellow croaker juveniles could improve the osmoregulatory capacity by increasing lamellar MRCs and NKA activity with the decrease in salinity from 24 to 4‰.However,the associated histopathological lesions are likely to negatively influence the fish by affecting respiration and osmoregulation,especially when the salinity is below 4‰.