Among six major types of tea(white,green,oolong,yellow,black,and dark teas)from Camellia sinensis,oolong tea,a semi-fermented tea,with its own unique aroma and taste,has become a popular consumption as indicated by th...Among six major types of tea(white,green,oolong,yellow,black,and dark teas)from Camellia sinensis,oolong tea,a semi-fermented tea,with its own unique aroma and taste,has become a popular consumption as indicated by the increasing production.Representing the characteristic flavonoids of oolong tea,theasinensins are dimeric flavan-3-ols.Many recent studies have indicated that oolong tea and theasinensins possess several health benefit properties.We consider it significant and necessary to have a comprehensive review in the recent advances of oolong tea.Therefore,the aim of the present review is to provide a new perspective on oolong tea and its characteristic phytochemicals,theasinensins associated with health benefits,molecular action pathway,and chemical mechanism of theasinensin formation from scientific evidences available on the literature.Furthermore,the chemical characterization of the oxidation products and the model oxidation system to the chemical changes of theasinensins are also discussed.展开更多
To solve the problem of low robustness of trackers under significant appearance changes in complex background,a novel moving target tracking method based on hierarchical deep features weighted fusion and correlation f...To solve the problem of low robustness of trackers under significant appearance changes in complex background,a novel moving target tracking method based on hierarchical deep features weighted fusion and correlation filter is proposed.Firstly,multi-layer features are extracted by a deep model pre-trained on massive object recognition datasets.The linearly separable features of Relu3-1,Relu4-1 and Relu5-4 layers from VGG-Net-19 are especially suitable for target tracking.Then,correlation filters over hierarchical convolutional features are learned to generate their correlation response maps.Finally,a novel approach of weight adjustment is presented to fuse response maps.The maximum value of the final response map is just the location of the target.Extensive experiments on the object tracking benchmark datasets demonstrate the high robustness and recognition precision compared with several state-of-the-art trackers under the different conditions.展开更多
基金This study was made possible by Naresuan University,Phisanulok,Thailand under the International Research University(IRU)program and Anhui Major Demonstration Project for Leading Talent Team on Tea Chemistry and Health,Anhui Department of Education,Hefei,China.
文摘Among six major types of tea(white,green,oolong,yellow,black,and dark teas)from Camellia sinensis,oolong tea,a semi-fermented tea,with its own unique aroma and taste,has become a popular consumption as indicated by the increasing production.Representing the characteristic flavonoids of oolong tea,theasinensins are dimeric flavan-3-ols.Many recent studies have indicated that oolong tea and theasinensins possess several health benefit properties.We consider it significant and necessary to have a comprehensive review in the recent advances of oolong tea.Therefore,the aim of the present review is to provide a new perspective on oolong tea and its characteristic phytochemicals,theasinensins associated with health benefits,molecular action pathway,and chemical mechanism of theasinensin formation from scientific evidences available on the literature.Furthermore,the chemical characterization of the oxidation products and the model oxidation system to the chemical changes of theasinensins are also discussed.
文摘To solve the problem of low robustness of trackers under significant appearance changes in complex background,a novel moving target tracking method based on hierarchical deep features weighted fusion and correlation filter is proposed.Firstly,multi-layer features are extracted by a deep model pre-trained on massive object recognition datasets.The linearly separable features of Relu3-1,Relu4-1 and Relu5-4 layers from VGG-Net-19 are especially suitable for target tracking.Then,correlation filters over hierarchical convolutional features are learned to generate their correlation response maps.Finally,a novel approach of weight adjustment is presented to fuse response maps.The maximum value of the final response map is just the location of the target.Extensive experiments on the object tracking benchmark datasets demonstrate the high robustness and recognition precision compared with several state-of-the-art trackers under the different conditions.
