There are two sides to every story of visual saliency modeling in the frequency domain. On the one hand, image saliency can be effectively estimated by applying simple operations to the frequency spectrum. On the other hand, it is still unclear which part of the frequency spectrum contributes the most to popping-out targets and suppressing distractors. Toward this end, this paper tentatively explores the secret of image saliency in the frequency domain. From the results obtained in several qualitative and quantitative experiments, we find that the secret of visual saliency may mainly hide in the phases of intermediate frequencies. To explain this finding, we reinterpret the concept of discrete Fourier transform from the perspective of template-based contrast computation and thus develop several principles for designing the saliency detector in the frequency domain. Following these principles, we propose a novel approach to design the saliency detector under the assistance of prior knowledge obtained through both unsupervised and supervised learning processes. Experimental results on a public image benchmark show that the learned saliency detector outperforms 18 state-of-the-art approaches in predicting human fixations.
Disclosed is a method for processing a green tea with delicate chestnut aromas. The production method comprises: using fresh leaves of lobular varieties with the tenderness of 1 bud 1 leaf or 1 bud 2 leaves in early development as the raw material, using a specific facility and tea leaves spreading technology to carry out a spreading operation, and after the three-stage fixation treatment by an electromagnetic fixation device, destroying the cell tissues in the leaves by using a suitable light rolling or medium rolling process, and then carrying out a drying operation by a stoving plus baking process to obtain a green tea product.
Provided is a heat management structure, comprising: a heat dissipation layer; an accommodation member for accommodating at least one heat source; and a heat pipe having two ends respectively connected to the heat dissipation layer and the accommodation member. Also provided is an unmanned aerial vehicle. The present invention has a compact structure and excellent heat dissipation performance, and does not consume energy and increase load, thereby ensuring operation stability and safety of an unmanned aerial vehicle.
A polymorph of (R)-methyl-2-(3-aminopiperidin-1-yl)-3-(2-cyanobenzyl)-4-carbonyl-3,4-dihydrothiophene [3,2-d] pyrimidine-6-carboxylic acid maleate and a preparation method therefor. The polymorph is a crystal having high stability and low hygroscopicity, and the crystal from is selected from among crystal form A, crystal form B and crystal form C. Furthermore, the crystal form has a strong in-vivo hypoglycemic activity and can be used for the preparation of a novel drug for the treatment or prevention of diabetes mellitus type H and/or complications of diabetes mellitus type H.
Provided are a fatty acid compound, a preparation method therefor and use thereof. The fatty acid compound is of a structure represented by general formula I. The compound has an AMPK activating activity and a capability of inhibiting glucose production in mouse primary hepatocytes, and can be used for preparing drugs for treating obesity or diabetes.
The invention relates to a method for computerizing delineation and/or multi-label classification of an ECG signal, comprising: applying a neural network to said ECG. labelling the ECG. and optionally displaying the labels according to time with the ECG signal.
The present invention relates to the crystal form A of a compound. The present invention further discloses a preparation method and a pharmaceutical composition of the crystal form A of the compound. The crystal form A has a relative strong hypoglycemic activity in vivo, and is hoped to be a new pharmaceutically active ingredient for treating or preventing type II diabetes and/or the complications of type II diabetes.
A data processing method, associated apparatus, and data storage system (100) are provided to enhance flexibility of data storage. The data processing method comprises: a first storage management apparatus (140) receives a data write request generated by a first server (110) (301), wherein the first storage management apparatus (140) is disposed in the first server, and the data write request comprises data to be written; the first storage management apparatus (140) determines, according to the data write request, scheduling information corresponding to the data write request (302), wherein the scheduling information corresponding to the data write request indicates a distributed storage pool (150) or a local data storage apparatus (120) of the first server; and the first storage management apparatus (140) processes, according to the scheduling information corresponding to the data write request, the data to be written (303). A storage management apparatus (500), a computing apparatus, and a data storage system (100) are also provided to enhance flexibility of data storage.
Provided is a method for adjusting the threshold voltage of a multi-grid structure device, which is characterized by preparing a multi-grid structure device to form a channel impurity distribution of high doping on the surface and low doping in the interior; and using impurity doping to reduce the influence of coulomb impurity scattering on a carrier as far as possible while adjusting a threshold voltage, so as to enable that the carrier mobility is maintained at a higher level. Firstly, the solution can enable a multi-grid device to obtain a multi-threshold voltage with a larger range, and facilitate different demands of IC designers for the device in the process of circuit design. Secondly, in the process of introducing impurity doping to adjust a threshold voltage, the influence of coulomb impurity scattering on a channel carrier is reduced as far as possible to enable that the carrier mobility is maintained at a higher level and ensure that the device has a higher drive current. Finally, the solution can be achieved with a technological method compatible with the traditional CMOS, thereby having the potential of mass production.