1094results about How to "Eliminate negative effects" patented technology

An active matrix driven electro-optical device, such as a liquid crystal device, is provided which is able to add sufficient storage capacitance to pixel electrodes and to decrease the diameter of contact holes connecting with pixel electrodes, even when a fine pixel pitch is employed. The liquid crystal device has TFTs, data lines, scanning lines, storage capacitor lines, and pixel electrodes provided on a TFT array substrate. Each of the pixel electrodes is electrically connected to one of the TFTs by two contact holes through a barrier layer. A part of a semiconductor layer and each of the capacitor lines sandwich a first dielectric film and constitute a first storage capacitor, while a part of the barrier layer and each of the capacitor lines sandwich a second dielectric film and constitute a second storage capacitor.

The invention relates to an individual network news recommending method based on multitheme tracing, which comprises the following steps: obtaining a news webpage browsed by a user and dividing the news webpage into a plurality of user interest submodels; dynamically updating a multitheme interest model of a user according to the condition that whether the user reads a news report recommended by the interest submodels or not; judging whether the quantity of the interest submodels exceeds a preset threshold value, if so, searching and deleting the interest submodel which is farthest from the interest of the user; searching the maximal similarity between news to be recommended and all the interest submodels; calculating sequencing values of the news report and sequencing the sequencing values from the large value to the small value; and recommending a sequenced news list to the user. The invention can cover various interest characteristics of the user and has the advantages of high recommending accuracy and lighter following maintenance burden of a system.

When light generated in a light emitting layer of a display device is emitted through insulating films such as a planarizing film, an interlayer insulating film and a gate insulating film of a transistor, diffused reflection is caused due to slight-rough surfaces of the insulating films every time light passes through each insulating film. Accordingly, problems are caused in which stray light is caused, an outline of a pixel becomes indistinct, and a characteristic of the transistor is deteriorated. A display device according to the invention includes a transistor formed over a substrate, an insulating film with a light shielding property formed on the transistor, an opening for transmitting light therethrough formed in the insulating film with the light shielding property, and a light emitting element formed overlapping the opening.

The invention discloses a target tracking method based on a twin neural network and a parallel attention module, and belongs to the field of machine vision. The method comprises the following steps: 1, cutting a template image and a search area image according to the position and size of a target in a video sequence picture to form a training data set; 2, constructing a twin network, wherein a basic skeleton of the twin network adopts a finely tuned residual network; 3, embedding parallel attention modules into template branches of the twin network, wherein the parallel attention modules comprise a channel attention module and a space attention module which are parallel to each other; 4, based on the training set, constructing an adaptive focus loss function, training a twin network with aparallel attention module, and obtaining a network model with training convergence; 5, performing online tracking by using the trained network model. In the tracking process, the problems of target appearance change and the like can be effectively solved, and the tracking precision is improved.

The invention relates to relates to a method for preparing a composition containing 1,3-di-unsaturated fatty acyl-2-saturated fatty acyl glyceryl ester, particularly containing 1,3-di-oleoyl-2-palmitoyl glyceryl ester (OPO), which comprises the steps of: carrying out random ester exchange on an ester rich in saturated fatty acid residue and an ester rich in unsaturated fatty acid residue serving as reaction substrates, and carrying out enzymatic oriented ester exchange on the reaction mixture and unsaturated fatty acid or unsaturated fatty acid ester. The invention also provides application of the composition to preparing formula milk powder for infant and preparing synthetic food for infant, and the formula milk powder for infant and the synthetic food for infant containing the composition.

The present invention discloses a troposphere atmosphere delay error correction method and correction system. The method comprises the steps of (1) obtaining GNSS observation data through a GNSS foundation reinforcement system network reference stations, (2) sending the GNSS observation data to a CORS server baseline solution system, wherein the step (2) concretely comprises the steps of (a) decomposing troposphere delay into a troposphere dry component and a troposphere wet component according to the GNSS observation data and (b) carrying out regional linear interpolation model processing on the troposphere wet component to obtain a wet component correction value, and using a troposphere prior model to carry out altitude component correction of the troposphere dry component to obtain a dry component correction value, (C) combining the wet component correction value and the dry component correction value to obtain a troposphere correction value total. According to the method, the improvement of troposphere delay error correction is facilitated.

The invention relates to an energy transfer type power battery quick balancing system and a control method thereof. The invention relates to the battery voltage balancing system and further relates to the control method of battery pack quick balancing process, which solves the problem that a balanced battery single body and an energy storing element are impacted through pulse current by a currentbalancer in the prior art. The system comprises a power battery pack single body battery voltage and flying capacitor voltage detecting network, a current sensor and a signal modulating circuit, a single chip microcomputer, an electric power transfer channel, a first filter circuit, a two-way buck-boost converter, a second filter circuit and a flying capacitor. After starting to work, the system firstly detects the battery pack single body battery voltage, the flying capacitor voltage and the charging/discharging current; and if the dropout voltage between the highest voltage of the battery single body and the lowest voltage of the single body exceeds a set value, the balancing operation is started.

The disclosed embodiments include an apparatus and method of using a laser to scan the ground or a target from an airborne or ground-based platform. In certain embodiments, the apparatus and method produces a 3-D elevation model of the terrain. In some embodiments, the apparatus includes a pulsed laser, a receiver to detect and amplify the pulse after being reflected by objects on the ground (or the ground itself), and electronics which measures the time of flight of the optical pulse from which the slant range to the target is calculated. Technical advantages of the disclosed embodiments include avoiding blind zones to ensure that no laser shots are wasted. In certain embodiments for airborne applications, the apparatus may also be configured to maintain a constant swath width or constant spot spacing independent of aircraft altitude or ground terrain elevation.

Generally speaking, the method of the present invention is performed by making various ultrasonic scans at preselected orientations along the length of a material being tested. Data from the scans are then plotted together with various calculated parameters that are calculated from this data. Lines or curves are then fitted to the respective plotted points. Review of these plotted curves allows the location and severity of defects within these sections to be determined and quantified. With this information various other decisions related to how, when or whether repair or replacement of a particular portion of a structure can be made.

The invention relates to a multienzyme microorganism and plant growth promotion additive. An active regulator containing multienzyme microorganism and leavening, amylase, protease, cellulose, pectase and polyase are used as base stocks, are added with chitin, various elements and vitamins in micro quantity, amino acid and are added with a microorganism and leavening inoculum Em bacterial solution, are mixed and are fermented to prepare the multienzyme microorganism and plant growth promotion additive. The additive can effectively improve soil, strength land capacity, reduce plant diseases and insect pests, fully excavate the soil potential, bring organic nutritive compositions for plants, promote the healthy and rapid growth of the plants, remove negative influence of pesticide and fertilizer, protect soil and entironment, is beneficial for society and people, adopt waste utilization, and has low investment, low cost, rapid effect and good effect.

The present invention relates to a plug-type connection for releasably connecting two bodies wherein the first body includes a first connection zone having a plurality of coaxially aligned first connection segments and the second body includes a plurality of coaxially aligned second connection segments. The first connection segment includes at least one segment with a non-circular cross-section and at least one segment having an axially tapering cross-section. The second connection segment includes at least one segment with a non-circular cross section sized and configured for rotatively engaging the non-circular cross section of the first connection segment and at least one segment sized and configured to mate with the at least one tapering section, wherein the tapered section of the first connection segment and the mating section of the second connection segment form an axially non-positive engagement for guiding the first body into engagement with the second body. The axially tapering cross-section of the first connection segment is preferably made from a deformable material so that the axially tapering cross section deforms when it mates with the second connection segment.

A method for the indirect application of printing liquid onto a printing material provides an intermediate carrier, preferably a circulating belt, a liquid conditioning medium including a first substance applied onto the intermediate carrier and a printing liquid, in particular an inkjet ink, including a second substance applied onto the conditioning medium on the intermediate carrier. The printing liquid is situated as droplets or a layer substantially on the conditioning medium and the droplets or the layer form a contact region on their underside with the conditioning medium. The printing liquid is heated, preferably by way of a dryer and the printing liquid is transferred from the intermediate carrier onto the printing material.

The invention discloses a multi-layer anomaly detection method based on network traffic. The invention is capable of detecting the small traffic attack behavior well with high detection accuracy, andmay adapt to different data sets. The invention comprises: firstly adopting a binary representation of symbol attributes in the data preprocessing stage to eliminate the negative influence of the traditional numerical size on the classification, and raising the attribute set of the data set to a relatively high dimension, so that the subsequent data classification effect is more accurate; then using the dimension reduction method to extract features and reduce the amount of data, so that the running speed is faster and the memory consumption is lower during the subsequent steps; subsequently,using the KNN outlier detection method and genetic algorithm combination method for data selection, so that different types of data are more balanced, each type of data is separated as far as possible, and the classification result is fairer; finally, using the constructed multi-layer classifier, thereby enabling more accurate identification of large-flow attacks and small-flow attacks.

The invention relates to a lithium battery SOH estimation and RUL prediction method based on an AST-LSTM neural network, and belongs to the field of lithium battery PHM. The method comprises the following steps: 1) collecting voltages, currents, temperatures and corresponding capacity values of multiple battery charging and discharging periods; 2) constructing a deep AST-LASTM model; and 3) carrying out lithium battery SOH estimation and RUL prediction on the basis of the AST-LSTM neural network. The battery capacity data can be obtained by only measuring voltage, current, temperature and timeof a lithium battery to be detected, the SOH and RUL of the lithium battery are estimated, the measurement process is simple, the error is low, and the precision is high.

The present invention relates to compositions and methods for treatment of ear infections and inflammation. Specifically, the method involves administering to a patient a composition containing an antibacterial, an anti-inflammatory, and an anti-septic that is substantially free of pramoxine to treat the occurrence or negative effects of otic infection and inflammation.

A method and an apparatus for detecting gas concentration with infrared absorption characteristics have the function of automatic calibration. The function of automatic calibration is accomplished in such a manner that a reference chamber filled with CO₂ gas of a known concentration is added in the apparatus; by switching the reference chamber in real-time or periodically into the detecting light path, the measured value of the concentration of CO₂ gas in the reference chamber can be obtained; then by comparing the measured value with the standard value of known concentration of CO₂ gas, the calibration factor is obtained; once finishing the calibration process, a detecting chamber is switched into the detecting light path so that a concentration of a gas can be detected accurately. In the apparatus of the present invention, the same detecting light path (e.g. the infrared light source, and the light filter, etc.), the same infrared light signal processing unit (e.g. the infrared sensor, the amplification circuit and the single-chip microcomputer system, etc.) are used for both the calibration process and the detection process. Therefore, the negative effect caused by difference of the characteristics of temperature drift of circuit components between different light paths in prior art can be eliminated, and the errors in the detected results by using the apparatus in the present invention is greatly reduced. In addition, the structure of the apparatus in the present invention is relatively simple, resulting in the lower production cost and no necessity for the manual maintenance.

The invention discloses a multi-person pose estimation method based on mask perceptual depth reinforcement learning. Firstly, a multi-person pose estimation model is constructed. The multi-person poseestimation model is composed of a detection network for acquiring a detection frame and a mask, a depth reinforcement learning network for improving the positioning accuracy, and a single-person poseestimation network. Then the multi-character pose estimation model is trained by training samples. During the test, the images to be detected are input into the trained multi-person pose estimation model to obtain the pose of all the detection frames of the images to be detected. The method of the invention introduces the mask information into the depth reinforcement learning network and the single-person attitude estimation network, improves the effects of the two stages, and solves the problems of gradient disappearance and gradient explosion by using the residual structure. The method of the invention is more competitive than other advanced multi-character posture estimation methods.

A substrate coating and method of applying the same in which the coating provides UV blocking and AR properties and in which at least one of the layers is formed of a combination of zinc oxide and a stabilizing secondary metal oxide.

The invention provides a method and device for pushing mobile internet content based on user preference for obtaining internet content preference information of a user according to access of the user and pushing pointed content to the user according to the preference information. The method for pushing the mobile internet content based on the user preference comprises the steps that first information to which the user has access is obtained according to the access behavior of the user; the network content relevant to the first information is classified to obtain second information; the preference of the user is graded according to the second information to obtain the preference information of the user, and the network content corresponding to the preference information is pushed to the corresponding user according to the preference information. By means of the technical scheme, the preference information of the user is obtained through the network, the potential customers can be located according to the preference information, and the efficiency of popularizing the network service is improved.

A method and apparatuses transform the electrical energy from multiple sources to grid compliant AC voltage. The apparatus according to one embodiment comprises: multiple energy sources (GES 1 to m in FIG. 1) from various power conversion devices; controllers (GER 1 to m in FIG. 1) pumping the maximum available electrical power to the DC inter-source bus; a hybrid inter-source DC bus carrying the electrical power from controllers to inverters and load level information among inverters; and inverters (CNT 1 to k) converting the DC power from the inter-source DC bus to grid compliant AC power.

The invention discloses a semantic approximate query method for an RDF knowledge map. The offline stage of the invention comprises the following steps: firstly, the RDF knowledge map is divided into RDF knowledge maps according to the semantic locality characteristics of entities and predicates of the RDF knowledge map, and the divided knowledge maps are generated into trainable text corpus; secondly, context-sensitive semantic learning is performed on the text corpus using the text embedding model, and the semantic vectors of entities and predicates are obtained. In the on-line phase: firstly, the syntax of SPARQL query submitted by users is analyzed, and the semantics of the predicates is extended; secondly, an approximate query based on predicate semantic similarity is carried out froma given entity, and the semantic approximate query results are obtained. The method utilizes semantic locality features to carry out context-sensitive semantic learning on the RDF knowledge map, thereby supporting fuzzy query application of the RDF knowledge map, and returning approximate query results satisfying user query intent in real time.

The invention discloses a composite coating of diamond/W-C gradient structure, and a preparation method thereof. The composite coating is characterized in that a composite coating which takes a W-C gradient structure as a transition layer and takes a diamond coating as a wearable coating is formed on a WC-Co hard alloy; the preparation method comprises the steps as follows: the hard alloy is used as a sample; a surface layer WC is dipped in Murakami reagent so as to form a Co-rich layer; subsequently, pure tungsten is taken as target material and argon and carbonaceous gas are used as reaction gases to prepare the W-C gradient structure of different structures as the transition layer by a reaction sputtering method by adjusting the concentration of carbonaceous active gas in the reaction gas, thus preparing the diamond wearable coating by a chemical vapor deposition technique. The composite coating solves the problem of the bonding strength of the film/substrate and ensures the surface smoothness of the diamond film.

The invention discloses a network intrusion detection method. The network intrusion detection method includes: searching network data to construct a test network data set; performing feature extraction on the test network data set by utilizing a kernel principal component analysis method; constructing a training data set, putting the training data set into a support vector machine classifier for training; obtaining feature datasets, obtaining an optimal feature subset from the feature data set by using a genetic algorithm; utilizing a firefly swarm optimization algorithm to obtain the overalllocal optimal feature subset and the optimal support vector machine parameters from the optimal feature subset, processing the training data set according to the overall local optimal feature subset,and inputting the training data set into a support vector machine classifier for classification modeling to obtain a network intrusion detection model. According to the method, the simplicity and convenience of the algorithm are improved, abnormal data can be more effectively found from samples, the detection accuracy of network intrusion is effectively improved, the missing report rate and the false report rate are reduced, and the overall performance of network intrusion detection is improved.

The invention discloses a collaborative filtering recommendation method integrating information entropy similarity and dynamic trust. The method is based on two similarity calculation methods of information entropy similarity and trust implicit similarity of score difference, and constructs a comprehensive similarity calculation model to alleviate the problem that similarity of cold start users isdifficult to calculate. Then, direct, indirect and global trust calculation models are constructed to reduce false recommendation of unreliable users by integrating the reliability of scoring and recommendation. Secondly, a fusion scoring prediction model of similarity and trust is constructed to complete the scoring prediction and personalized project recommendation for the target users. Finally, we evaluate the effectiveness of the recommendation user scoring, and propose a trust reward and punishment strategy to update the trust neighbor set dynamically for the target user, to suppress thenegative impact of users' random false scoring on the recommendation performance. The experimental results show that the method can improve the recommendation accuracy and reliability of the recommendation system, effectively alleviate the data reliability, data sparsity and cold start problems.