DINO(self-distillation with no label)首次将自监督学习与transformer结合。为结合卷积网络的局部性优势,提出一个三分支的网络模型DINO+,即在DINO的基础上添加一个卷积蒸馏模块,给DINO中的transformer进行知识蒸馏,从而将卷积网络与t...DINO(self-distillation with no label)首次将自监督学习与transformer结合。为结合卷积网络的局部性优势,提出一个三分支的网络模型DINO+,即在DINO的基础上添加一个卷积蒸馏模块,给DINO中的transformer进行知识蒸馏,从而将卷积网络与transformer结合。蒸馏后的ViT(vision transformer)在STL-10、CIFAR-10数据集上的分类准确率分别上升了5.7%、4.8%,且优于其他自监督模型,证明了提出方法的有效性。展开更多
Aminopeptidase N(APN/CD13), a Zn<sup>2+</sup>-dependent ectopeptidase localized on the cell surface, is widely considered to influence the invasion of tumor cells. We found that boroleucine and dino-leucin...Aminopeptidase N(APN/CD13), a Zn<sup>2+</sup>-dependent ectopeptidase localized on the cell surface, is widely considered to influence the invasion of tumor cells. We found that boroleucine and dino-leucine borate exhibited a strong inhibitory effect on the enzyme activity of aminopeptidase N. The tested assay indicated that both compounds had an anti-proliferative effect on triple-negative breast cancer cells. Wound healing assay, migration test and matrigel-coated transwell assay showed that both boroleucine and dino-leucine borate inhibited the migration and invasion of breast cancer cells. Immunoblot analysis showed that both compounds down-regulated the expression of matrix metalloproteinase-2/9. In the capillary tube formation assay of human umbilical vein endothelial cells (HUVECs), dino-leucine borate showed better antiangiogenic activity than ubenimex even at a low concentration (10 μM). Moreover, compared with ubenimex, the anti-metastatic activity of dino-leucine borate in vivo was similar to or even better than that of ubenimex in the H22 pulmonary metastasis mouse model. In this paper, we found the novel APN inhibitors to markedly suppress the enzyme activity of APN and inhibit the migration and invasion of tumor cells in vitro and in vivo.展开更多
The phytoplankton group’s composition is an excellent indicator of the state of coastal waters. The aim of this study is to explore spatial variability of phytoplankton in shallow coastal areas from different regions...The phytoplankton group’s composition is an excellent indicator of the state of coastal waters. The aim of this study is to explore spatial variability of phytoplankton in shallow coastal areas from different regions(East, Center and West) of the Algerian coast. The quantitative and qualitative study of the phytoplankton population was conducted on samples taken during the 2012–2013 period by the research vessel GRINE BELKACEM. The qualitative results show a significant diversity of diatoms and dinoflagellates, which rank first and second in the total flora(Fr > 50%). This finding is also confirmed quantitatively by the abundance values in both 2013 and 2012 samples which largely exceeds the average value in most other areas of the western region(56%) and slightly exceeding 52% in all central areas which represent an equi-distribution between diatoms and dinoflagellates. The Dinoflagellates contribute significantly to the quantitative richness in the eastern region(>60%). The minimum value of Dia/Dino index, recorded in the eastern region confirms the dominance of the dinoflagellates especially in Skikda(0.31)(Skikda).However, our results reveal the presence, in smaller proportion, of other groups such as the cocolithophoridae and euglenophyceae.展开更多
Understanding the dynamics of phytoplankton communities in coastal zones is crucial for the management and conservation of coastal ecosystems.Previous research indicated that the phytoplankton community structure and ...Understanding the dynamics of phytoplankton communities in coastal zones is crucial for the management and conservation of coastal ecosystems.Previous research indicated that the phytoplankton community structure and dominant taxa in the Bohai Sea(BHS)have exhibited significant shifts from the 1990s to the early 2010s in response to environmental changes,especially the change in nutrient structure.This study comprehensively investigated the variations in net-collected phytoplankton(>76μm)community structure,diversity,and environmental factors in the BHS during the late summers of 2011-2020,aiming to understand the recent trend in phytoplankton community structure and to explore the interactions between the communities and the environment.During the study period,the nutrient status in the BHS was characterized by a decrease in dissolved inorganic nitrogen(DIN)concentration,an increase in dissolved inorganic phosphorus(DIP)concentration,and a return of the nitrogen-to-phosphorus(N/P)molar ratio(hereinafter referred to as N/P ratio)to the Redfield ratio since 2016.The eutrophication index(EI)in the BHS remained stable and was generally at a low level(<1).The Dia/Dino index fluctuated but did not show an obvious trend.Overall,the eutrophication,the imbalance in nutrient ratio,and the shift in phytoplankton community structure did not continue during the study period.The increased abundance of phytoplankton was strongly associated with elevated concentrations of DIN,as well as higher N/P and nitrogen-to-silicon(N/Si)ratios,whereas the greater diversity was strongly linked to higher concentrations of DIP.Diatoms and dinoflagellates showed significant differences in their interactions with the environment,and their relative dominance was related to water column depth and stratification intensity;their impacts on the phytoplankton community diversity were also significantly different.The variations of certain dominant species,i.e.,Skeletonema costatum,Paralia sulcata,and Tripos longipes,exhibited strong links to the changes in nutrient structure in the BHS.The findings of this study contribute to understanding the regional environmental changes and provide insights into the adaptive strategies of coastal ecosystems in response to environmental shifts and fluctuations.展开更多
Real-time and accurate traffic light status recognition can provide reliable data support for autonomous vehicle decision-making and control systems.To address potential problems such as the minor component of traffic...Real-time and accurate traffic light status recognition can provide reliable data support for autonomous vehicle decision-making and control systems.To address potential problems such as the minor component of traffic lights in the perceptual domain of visual sensors and the complexity of recognition scenarios,we propose an end-to-end traffic light status recognition method,ResNeSt50-CBAM-DINO(RC-DINO).First,we performed data cleaning on the Tsinghua-Tencent traffic lights(TTTL)and fused it with the Shanghai Jiao Tong University’s traffic light dataset(S2TLD)to form a Chinese urban traffic light dataset(CUTLD).Second,we combined residual network with split-attention module-50(ResNeSt50)and the convolutional block attention module(CBAM)to extract more significant traffic light features.Finally,the proposed RC-DINO and mainstream recognition algorithms were trained and analyzed using CUTLD.The experimental results show that,compared to the original DINO,RC-DINO improved the average precision(AP),AP at intersection over union(IOU)=0.5(AP50),AP for small objects(APs),average recall(AR),and balanced F score(F1-Score)by 3.1%,1.6%,3.4%,0.9%,and 0.9%,respectively,and had a certain capability to recognize the partially covered traffic light status.The above results indicate that the proposed RC-DINO improved recognition performance and robustness,making it more suitable for traffic light status recognition tasks.展开更多
文摘DINO(self-distillation with no label)首次将自监督学习与transformer结合。为结合卷积网络的局部性优势,提出一个三分支的网络模型DINO+,即在DINO的基础上添加一个卷积蒸馏模块,给DINO中的transformer进行知识蒸馏,从而将卷积网络与transformer结合。蒸馏后的ViT(vision transformer)在STL-10、CIFAR-10数据集上的分类准确率分别上升了5.7%、4.8%,且优于其他自监督模型,证明了提出方法的有效性。
文摘Aminopeptidase N(APN/CD13), a Zn<sup>2+</sup>-dependent ectopeptidase localized on the cell surface, is widely considered to influence the invasion of tumor cells. We found that boroleucine and dino-leucine borate exhibited a strong inhibitory effect on the enzyme activity of aminopeptidase N. The tested assay indicated that both compounds had an anti-proliferative effect on triple-negative breast cancer cells. Wound healing assay, migration test and matrigel-coated transwell assay showed that both boroleucine and dino-leucine borate inhibited the migration and invasion of breast cancer cells. Immunoblot analysis showed that both compounds down-regulated the expression of matrix metalloproteinase-2/9. In the capillary tube formation assay of human umbilical vein endothelial cells (HUVECs), dino-leucine borate showed better antiangiogenic activity than ubenimex even at a low concentration (10 μM). Moreover, compared with ubenimex, the anti-metastatic activity of dino-leucine borate in vivo was similar to or even better than that of ubenimex in the H22 pulmonary metastasis mouse model. In this paper, we found the novel APN inhibitors to markedly suppress the enzyme activity of APN and inhibit the migration and invasion of tumor cells in vitro and in vivo.
基金Aquatic Environment Monitoring and Quality Research Program Funded by the National Center for Research and Development of Fisheries and Aquaculture(AIEA-RAF 7012 Project)
文摘The phytoplankton group’s composition is an excellent indicator of the state of coastal waters. The aim of this study is to explore spatial variability of phytoplankton in shallow coastal areas from different regions(East, Center and West) of the Algerian coast. The quantitative and qualitative study of the phytoplankton population was conducted on samples taken during the 2012–2013 period by the research vessel GRINE BELKACEM. The qualitative results show a significant diversity of diatoms and dinoflagellates, which rank first and second in the total flora(Fr > 50%). This finding is also confirmed quantitatively by the abundance values in both 2013 and 2012 samples which largely exceeds the average value in most other areas of the western region(56%) and slightly exceeding 52% in all central areas which represent an equi-distribution between diatoms and dinoflagellates. The Dinoflagellates contribute significantly to the quantitative richness in the eastern region(>60%). The minimum value of Dia/Dino index, recorded in the eastern region confirms the dominance of the dinoflagellates especially in Skikda(0.31)(Skikda).However, our results reveal the presence, in smaller proportion, of other groups such as the cocolithophoridae and euglenophyceae.
基金The National Natural Science Foundation of China under contract No.42206161the Natural Science Foundation of Hebei Province under contract No.D2022407004+1 种基金the Science Research Project of Hebei Education Department under contract No.QN2022167the Open Fund Project of Hebei Key Laboratory of Ocean Dynamics,Resources and Environments under contract No.HBHY04.
文摘Understanding the dynamics of phytoplankton communities in coastal zones is crucial for the management and conservation of coastal ecosystems.Previous research indicated that the phytoplankton community structure and dominant taxa in the Bohai Sea(BHS)have exhibited significant shifts from the 1990s to the early 2010s in response to environmental changes,especially the change in nutrient structure.This study comprehensively investigated the variations in net-collected phytoplankton(>76μm)community structure,diversity,and environmental factors in the BHS during the late summers of 2011-2020,aiming to understand the recent trend in phytoplankton community structure and to explore the interactions between the communities and the environment.During the study period,the nutrient status in the BHS was characterized by a decrease in dissolved inorganic nitrogen(DIN)concentration,an increase in dissolved inorganic phosphorus(DIP)concentration,and a return of the nitrogen-to-phosphorus(N/P)molar ratio(hereinafter referred to as N/P ratio)to the Redfield ratio since 2016.The eutrophication index(EI)in the BHS remained stable and was generally at a low level(<1).The Dia/Dino index fluctuated but did not show an obvious trend.Overall,the eutrophication,the imbalance in nutrient ratio,and the shift in phytoplankton community structure did not continue during the study period.The increased abundance of phytoplankton was strongly associated with elevated concentrations of DIN,as well as higher N/P and nitrogen-to-silicon(N/Si)ratios,whereas the greater diversity was strongly linked to higher concentrations of DIP.Diatoms and dinoflagellates showed significant differences in their interactions with the environment,and their relative dominance was related to water column depth and stratification intensity;their impacts on the phytoplankton community diversity were also significantly different.The variations of certain dominant species,i.e.,Skeletonema costatum,Paralia sulcata,and Tripos longipes,exhibited strong links to the changes in nutrient structure in the BHS.The findings of this study contribute to understanding the regional environmental changes and provide insights into the adaptive strategies of coastal ecosystems in response to environmental shifts and fluctuations.
基金supported by the National Key R&D Program of China(2021YFB2501200)the Key Program of the National Natural Science Foundation of China(52131204)the Shaanxi Province Key Research and Development Program(2022GY-300).
文摘Real-time and accurate traffic light status recognition can provide reliable data support for autonomous vehicle decision-making and control systems.To address potential problems such as the minor component of traffic lights in the perceptual domain of visual sensors and the complexity of recognition scenarios,we propose an end-to-end traffic light status recognition method,ResNeSt50-CBAM-DINO(RC-DINO).First,we performed data cleaning on the Tsinghua-Tencent traffic lights(TTTL)and fused it with the Shanghai Jiao Tong University’s traffic light dataset(S2TLD)to form a Chinese urban traffic light dataset(CUTLD).Second,we combined residual network with split-attention module-50(ResNeSt50)and the convolutional block attention module(CBAM)to extract more significant traffic light features.Finally,the proposed RC-DINO and mainstream recognition algorithms were trained and analyzed using CUTLD.The experimental results show that,compared to the original DINO,RC-DINO improved the average precision(AP),AP at intersection over union(IOU)=0.5(AP50),AP for small objects(APs),average recall(AR),and balanced F score(F1-Score)by 3.1%,1.6%,3.4%,0.9%,and 0.9%,respectively,and had a certain capability to recognize the partially covered traffic light status.The above results indicate that the proposed RC-DINO improved recognition performance and robustness,making it more suitable for traffic light status recognition tasks.
