128results about How to "Reduce modeling costs" patented technology

The invention provides a cultivation method of Cyp gene knocked-out rats, and a preparation method of liver microsome of the rats. The Cyp gene knock-out herein includes Cyp single gene knock-out and Cyp multiple gene knock-out. In the method, firstly a Cyp gene knocked-out rat is constructed by means of a CRISPR / Cas system, which includes selection of a knocked-out target site, in-vitro synthesis and transcription of sg RNA and Cas9m RNA, preparation of a pseudopregnant female rat, in-vitro micro-injection and transplanting of a single-cell embryo, and cultivation of the rat, and finally, a homozygote Cyp gene knocked-out rat can be cultured. Furthermore, the liver of the Cyp gene knocked-out rat is extracted and is subjected to homogenization and differential centrifugation to prepare the liver microsome of the rat in Cyp gene deletion. The invention also provides an application of the Cyp gene knocked-out rats and the liver microsome thereof in study on drug metabolism.

The invention discloses a method for fast constructing a high-accuracy personalized face model on the basis of facial images, which belongs to the cross field of computer vision and computer graphics, and aims at achieving the purposes of fast reconstructing a personalized 3D geometrical facial model and a high-accuracy 2D facial texture mapping on the basis of real facial images. The method comprises the following steps: firstly, shooting five facial images from multiple angles; secondly, marking positions and corresponding relationships of forty-one characteristic points on the facial images and a neutral facial model; thirdly, aligning the projection of the neutral facial model on a virtual camera focal plane with the facial images; fourthly, carrying out hierarchical geometrical deformation on the neutral facial model to obtain the personalized geometrical facial model; and finally, synthesizing the high-accuracy facial texture mapping according to the five facial images. The face modeling method has the advantages of convenience for data acquisition, high modeling speed and modeling accuracy, and the like, and achieves higher practical values in the fields of video communication, electronic games, man-machine interaction, and the like.

The invention provides a near-infrared universal model detection method for quality indexes of fruits with similar optical and physical properties. According to the near-infrared universal model detection method, a near-infrared universal model is established; the near-infrared universal model is utilized for determining two parts of fruit quality indexes; the near-infrared universal model is established based on the similarity of near-infrared spectrums between the different types of the fruits based on the similar optical and physical properties. The basic modeling concept is that after the spectrums are collected and are pre-processed, a common characteristic wavelength range of all the types of the fruits is screened by using a moving window partial least square method; common characteristic wavelength points are further extracted in the common characteristic wavelength range by using an SPA (Super Pairwise Alignment) algorithm; finally, a PLS model or an MLR (Multiple Linear Regression) model is established by using existing software. According to the method, the determination accuracy is high and the feasibility is strong; the disadvantages that the different types of the fruits need to be detected by classes in an existing near-infrared detection technology are overcome; the modeling cost is reduced, the working efficiency is improved and the number of the modeling wavelength points is few; the near-infrared universal model detection method is applicable to a common optical filter type near-infrared instrument.

The invention relates to a three-dimensional visualization method for virtual crowd sports. It includes the following steps: analyzing the becoming virtual crowd, and founding modeling board model base and material base; capturing or generating by handcraft the need body sports data and storing it; composing the sports data; processing drawing preprocessing for modeling board model base and the sports data; inputting simple script data described crowd sports; generating virtual individual; driving it sports to generate dynamic virtual crowd; realize population playing up. The method has the advantages of high efficiency, common use, and low modeling cost, realizing real time vivid virtual crowd sports with 30000 peoples scale in common PC.

The invention discloses a general blood glucose prediction method based on data modeling and model transplanting. The method includes analyzing a subcutaneous blood glucose measurement signal of a human body, insulin infused into the human body and food taken by the human body, extracting the relation between the signal, the insulin and the food and future blood glucose, applying the relation to a new object short of data size, adopting a traditional lease square method and a latent variable method to build a self-regression blood glucose prediction model with an external source, conducting comparison, and selecting a base model to achieve model transplanting based on the latent variable method with better prediction performance. In actual application, a user does not need to wait for acquisition of a sufficient blood glucose measurement signal to conduct remodeling aiming at a new object, a base model can be directly transplanted to the new object, real-time blood glucose prediction is achieved according to an imitative effect by adjusting model coefficients on line, and building workload and complexity are greatly simplified.

The invention provides a ship electric system failure recovery method based on a multi-agent technology. A ship comprehensive electric system failure recovery algorithm is designed through a figure track analysis and multi-agent technology. A figure track analysis method is used for establishing a ship comprehensive electric system figure model, each side in the figure model corresponds to a component in a physical electric system, each main component has an agent of the main component, and component tracks are used for managing an electric system figure model topological structure and traversing other component agents. On the basis, the method uses the functions of information exchanging, interaction effect, cooperation, coordination work and communication of the agents as a foundation, and a ship electric system failure recovery task is completed in a coordination mode. The method can process the multi-priority failure recovery problem of an electric system with a loop and a load of the electric system, and under the premise that power supply recovery is carried out on a high-priority component preferentially, power supply recovery is carried out on outage components as many as possible.

The invention discloses a three-dimensional face modeling method and system based on multi-angle two-dimensional face data, and the method comprises the steps: recognizing face regions in a pluralityof face images with different shooting angles, and extracting face key feature information points of the face regions; performing corresponding matching adjustment on the face key feature informationpoints and corresponding feature information points of a standard face three-dimensional model; according to the plurality of face images, performing face size and shape matching adjustment on the standard face three-dimensional model to obtain a three-dimensional face grid model; and performing texture mapping on the three-dimensional face grid model based on the face texture map irrelevant to the viewpoint so as to generate a three-dimensional face image. According to the method, the cost of traditional three-dimensional face modeling can be reduced, and the precision of three-dimensional face recognition is improved.

The invention discloses a three-dimensional building model construction method applied to three-dimensional real estate information management. The invention belongs to the technical field of real estates. According to the invention, the SHP format two-dimensional house surface is generated through conversion; constructing an association relationship between a two-dimensional house and a two-dimensional house meter house of the three-dimensional house model; taking the basic level surface as the geographic space information reference of the two-dimensional house surface corresponding to the house layer; Construction of three-dimensional building model, assigning a value to the three-dimensional building model, and outputting and applying the three-dimensional building model to produce three three-dimensional building models with different data organization modes so as to adapt to browsing and query of the three-dimensional building models with multiple scales and different LOD levels at a Web end, and realizing rapid construction of the three-dimensional building model with geographic space information so as to be applied to two-dimensional and three-dimensional integrated management of the real estate.

The invention belongs to the field of oncobiology and relates to a method of establishing a tumor multi-drug resistant cell mode in vitro. The method comprises the step of continuously culturing a sensitive tumor cell by taking ROS as an incubation drug to enable the sensitive tumor cell to generate multi-drug resistance. Moreover, the invention further relates to a human breast cancer multi-drug resistant cell strain established by virtue of the method. The cell strain is named MCF-7 / ROS and preserved in CCTCC (China Center for Type Culture Collection), wherein the address is 430072, Wuhan University, Wuhan, China; the preservation date is 28th, January, 2015; the preservation number is CCTCC-C201520. The method provided by the invention is simple in process and stable in modeling condition, the modeling cost can be remarkably lowered, and the method has wide applicability. The MCF-7 / ROS cell strain established by the method can serve as an advantageous experimental mode for researching related mechanism, target spot exploration and screening of new drugs.

The invention relates to the method for increasing gear hob processing accuracy. Operation steps comprise the steps that (1) aiming at gear hob technologies of a processed gear, gear hob technological parameters which affect gear processing accuracy are extracted, a gear hob technological parameter combination is designed, gear hob processing experiments are conducted, and a gear face outline error indexes of the processed gear are obtained; (2) a gear face outline error digital model is established through a response curve face nonlinear fitting method according to the gear hob technological parameter combination and the gear face outline error indexes of the processed gear; (3) based on an expected processing accuracy value corresponding to each gear face outline error index, a weight ratio is applied to each gear face outline error model, and a digital model of gear face comprehensive outline errors is established; and (4) based on a genetic algorithm, the gear hob technological parameters are optimized and optimal gear hob technological parameters are obtained. According to the invention, the processing accuracy of a gear hob is increased; and a theoretical base for selection of actual gear hob technological parameters and increase of gear accuracy.

The invention discloses a method for nondestructive testing of grape berries based on the near infrared spectrum technology, and particularly relates to a method for identifying table grape berries of different varieties and maturity degrees and under different insect attack conditions based on the near infrared spectrum technology. The spectral preprocessing method and the chemometrics modeling method are combined to build a quantitative analysis prediction model with grape inclusions as the test index on grape berries of different varieties and maturity degrees and under different insect attack conditions. The near infrared spectrum technology is adopted to identity the interior quality of different grape samples for test. The method has the advantages of being high in testing speed, high in efficiency and low in cost, is suitable for nondestructive testing of quality and safety of agricultural products, and can be used for effectively identifying grape berries of different varieties and maturity degrees and under different insect attack conditions. A series of studies suggest that the combination of the near infrared spectrum technology and the chemometrics method has great potential and development prospects in the fields of quick prediction of inclusions of grapes and variety identification.

The invention discloses a method and a system for network resource modeling. The method includes the steps of setting a network resource meta-model, and network resource modeling. The network resource modeling includes adding a target network resource model, and particularly includes the following steps: obtaining target network resource information, obtaining target network resource object attribute information, and adding the target network resource information and the target network resource object attribute information to the target network resource model which is corresponding to the network resource meta-model. Thus, the target network resource model built through the network resource modeling method has uniform data specification, and uniform validation logic and service processing logic are met, so that no physical table needs to be added even new network resource models are constantly added, and the modeling cost of the network resource model is greatly reduced.

A height-adjustable electromagnetic shielding structure is located above an electronic element of a circuit board and includes an electromagnetic screen member, a connecting portion connected to the electromagnetic screen member, and a height stacking member. The height stacking member includes a coupling portion which is formed in a stepped structure and perpendicular to the electromagnetic screen member and circuit board, and contains an upper coupling portion and a lower coupling portion. The upper coupling portion has one end connected to the electromagnetic screen member through the connecting portion and another end connected to the lower coupling portion. At least two height stacking members are stacked over one another through the upper and lower coupling portions to perform height adjustment. The electromagnetic screen member, connecting portion and height stacking member are formed integrally, hence merely one set of mold is needed to reduce the cost.

The invention discloses a noninvasive fabricating method for a pulmonary fibrosis animal model. The method comprises the steps: the method that the bacterium solution dropping is poured inside a lower trachea through an endoscope is adopted, and a nasal zero degree 4 # hard-type endoscope, a xenon lamp cold light source and a Panasonic KS-822 video monitoring system are utilized; a rat is placed in a dryer with 26 cm diameter, and the ether anesthesia is carried out on the rat, the 200 microlitre of the medicine bleomycin is ejected inside the trachea, the medicine is fully flowed into the bronchus along with the airflow when inhaling air, elastic fixings clamped on the four limbs of the rat are removed, the rat is placed into a rat cage to wait for the anabiosis, the anabiosis period ranges from 30 sec to 60 sec, and the rat is placed into a constant temperature animal house for feeding; the activity, the feeding and the spirit conditions of the inoculated rat are observed every day, and the corresponding data is recorded.

The invention discloses a low-noise real-time global illumination drawing method for a three-dimensional geometric scene. The low-noise real-time global illumination drawing method comprises the steps of: initializing drawing of a current frame image by a main node, including, drawing view-port segmentation, model differential transmission and drawing background chartlet, and distributing segmented view-port parameters, a three-dimensional model and a background chartlet to each sub-node; carrying out multi-scale view-port layered drawing for the received three-dimensional model and video-port parameters, and transmitting each layer of drawn images to the main node by the sub-node; synchronizing the image of the current frame drawn by each sub-node and displaying the images in the same layer in a splicing manner by the main node; counting the number of frames drawn by each sub-node every second and drawing the image of the next frame by the main node. As the global illumination way is adopted to draw a complex three-dimensional geometric scene, the low-noise real-time global illumination drawing method for the three-dimensional geometric scene can provide roaming and rapid browsing of fixed scenes and dynamic movable scenes for a user, and can reduce noise points in the drawing process, improve drawing quality and lower modeling cost of the three-dimensional scene.

The present invention provides a method for reflecting a long-span engineering project progress. The method comprises the steps of: employing a real scene modeling technology and a BIM technology to construct an actual progress scene and a planning progress scene, and performing model data superposition to reflect an engineering progress condition, improve the controllability of engineering economic cost and benefit and ensure the completion of the whole engineering according to a progress plan. The method for reflecting a long-span engineering project progress can be widely applied to management of industrial and civil construction bridge and water conservancy and hydropower domain engineering subject visual progress.

The invention discloses a building three-dimensional modeling method and a building three-dimensional modeling device. The building three-dimensional modeling method comprises the steps of: generatingan oblique photographic model of a target building based on oblique photographic data, and generating a vector model of the target building based on vector data, wherein the vector model comprises afloor household model and a natural building external contour model; and carrying out fusion processing on the oblique photographic model and the vector model to generate a three-dimensional fusion model of the target building. The building three-dimensional modeling method and the building three-dimensional modeling device utilize the rapid modeling means such as generation of the oblique photographic and vector models, save the modeling cost, reserve the sense of reality of the models, maintain the strong correlation between the models and real estate business data, solve the difficulties such as modeling cost, efficiency, individual-floor and individual-household individualization, real estate business attribute data correlation of rapid modeling of a three-dimensional real estate floorhousehold natural building, realize the rapid production of a three-dimensional real estate floor household natural building model, improve the working efficiency and using experience degree of users.

The invention provides a container environment anomaly detection method and device, a medium and computer equipment, and the method comprises the steps: obtaining the historical process data of all business containers in each host machine node during normal operation for the host machine nodes in a container cluster; for each mirror image, establishing a detection model based on historical processdata in each service container corresponding to the mirror image; when the current process data of each service container is received, detecting the current process data based on the corresponding detection model, and judging whether the environment of each service container is abnormal or not; according to the method and the device, for each mirror image, the number of the containers created bythe mirror image is multiple, so that the process data is diversified, and the detection precision of the detection model can be ensured; moreover, in the container cluster, the number of the mirror images is far smaller than that of the service containers, so that compared with the mode of modeling each container in the prior art, the mirror images are modeled, and the modeling cost can be greatly reduced.

The invention relates to a fault injection method based on remote procedure call (RPC), which comprises the following steps: a fault injection front end providing fault information, and the fault information comprising information about a target single machine and information about a fault type; the remote procedure call RPC module packaging the fault information into a first communication frame and sends the fault information to the satellite simulator through a network; analyzing, by the satellite simulator, the first communication frame into information about a target single machine and information about a fault type, and generating single machine data according to the information about the target single machine and the information about the fault type; and packaging the stand-alone data into the second communication frame and sending the stand-alone data to the fault injection front end through the network. The invention also relates to a corresponding system. According to the invention, various satellite faults can be rapidly and efficiently simulated in a universal manner.

The invention relates to a method for improving machining accuracy of internal honing wheel strong gear honing. The method comprises the following steps that 1, according to an internal honing wheel strong gear honing technology of a machined gear, the gear honing process parameters influencing the gear machining accuracy are extracted, the gear honing process parameter combination is designed, the gear honing machining experiment is conducted, and the tooth surface outline error indexes of the machined gear are obtained; 2, according to the gear honing process parameter combination and the tooth surface outline error indexes of the machined gear, the mathematical model of the tooth surface outline error is constructed through the response surface nonlinear fitting method; 3, on the basis of the anticipated machining precision value corresponding to each tooth surface outline error index, the weight ratio is applied to each tooth surface outline error model, and the tooth surface comprehensive outline error mathematical model is constructed; and 4, based on the artificial immune clone selection algorithm, the gear honing process parameters are optimized, and the optimal honing process parameters are obtained. According to the method, the machining accuracy of the internal honing wheel strong gear honing is improved, and the theoretical basis is provided for the practical selection of the internal honing wheel strong gear honing process parameters and the improvement of the gear precision.

The invention belongs to the technical field of KPI anomaly detection, and particularly relates to a large-scale anomaly detection system based on KPI clustering. The system comprises the following modules: a preprocessing module, a baseline extraction module, a clustering module, a classification module and a log module; the preprocessing module uses a linear interpolation method to fill the percentage of KPI missing according to data points adjacent to KPI; the baseline extraction module removes extreme values which may be abnormal; the clustering module executes density-based clustering on a sampling KPI baseline based on shape similarity by using a density-based clustering method; the classification module calculates the distance between the new data and each previous centroid, sorts the distances, determines the minimum class as the classified class, and determines that the new time sequence curve does not belong to any previous class when the minimum distance is greater than a certain threshold value; and the log module records curves which do not belong to any category.

The invention discloses a compressor outlet parameter prediction modeling method based on a width learning system. The method comprises the steps of collecting data and preprocessing the data; constructing an outlet parameter prediction model of the compressor by utilizing a width learning system, setting the number of characteristic node groups of the width learning system as n, the number of k characteristic nodes in each group, the number of enhanced node groups as m and q enhanced nodes in each group, and performing modeling; when a new training data input model exists, using an incremental learning algorithm for carrying out online updating on an original model, an input data matrix of new training data being set as Xa belonging to Ra * M, an output data matrix being set as Ya belonging to Ra * C, a representing the number of newly-added training data, and performing model updating; and verifying the constructed model by using the root-mean-square error. According to the method, the performance prediction model of the large compressor can be quickly established, and the development time and cost of the model can be effectively saved.

The invention discloses a Ph+ acute B lymphocytic leukemia KQBL-84 cell line and a construction method and an application thereof. The cell line is preserved in China Center for Type Culture Collection on July 25, 2018 with the preservation number of CCTCC NO: C2018164. The cell line is characterized in that an MSCV-P210 retrovirus carrying a human fusion gene BCR-ABL infects bone marrow cells ofa C57bl / 6 donor mouse to obtain a primary human cell line stably expressing the BCR-ABL human protooncogene, and screening and culture are then performed to obtain a highly carcinogenic and stably inherited precursor B cell line. Cells of the line established in vitro are injected into a C57bl / 6 homologous receptor mouse. The cell line can be used for efficiently constructing a primary B-ALL mousemodel. The model has higher malignancy, rapid disease modeling and low modeling cost, and is suitable for evaluation of treatment methods of refractory / relapsing B-ALL and pathogenesis study of patients with B-ALL in Philadelphia chromosome positive (Ph+).

The invention discloses a modeling method of a home parameterized model, which comprises the following steps: S1, modifying any parameter or part in a modeling tool; S2, error detection: detecting whether factors influencing model form construction exist in the current model state or not, if yes, prompting a user, and if not, performing model construction; S3, initializing a current global dictionary. According to the modeling method, a parameterized model which is consistent with a real world household product in form and can have any required building production process auxiliary informationcan be constructed, and the modeling method can be used for household 3D design and production of household categories such as cupboards, wardrobes, doors and windows and suspended ceilings.

The invention discloses a data model generation method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring various different customer data; performing data processing on the plurality of customer data to obtain a plurality of corresponding sample data; inputting the plurality of sample data into a deep neural network for training to obtain a universal customer data coding model; performing model post-processing on the universal customer data coding model; and adding a classifier on the processed general customer data coding model, and inputting customer response data of the corresponding task into the general customer data coding model for model fine tuning to obtain a data model of the corresponding task, so that the data modeling efficiency is improved, and the modeling cost is reduced.

The invention provides a nitrogen fertilizer real-time recommendation method based on canopy coverage and plant height of facility crops, which belongs to the technical field of agricultural plantingand comprises the following steps: subtracting the nitrogen application amount of an optimal nitrogen application planting area from the nitrogen application amount of a non-optimal nitrogen application planting area, counting as Delta F, and calculating the nitrogen fertilizer recommendation index R of the non-optimal nitrogen application planting area at the same time, performing linear regression analysis on the Delta F values and the corresponding R values of all the non-optimal nitrogen application planting areas to obtain a nitrogen fertilizer recommendation formula; after the crops areplanted, monitoring the daily average temperature of the crop growth environment, calculating the nitrogen fertilizer recommendation index R value of the crops before crop topdressing, and substituting the R value into a nitrogen fertilizer recommendation formula to obtain a corresponding nitrogen fertilizer difference Delta F; and adding the nitrogen fertilizer difference Delta F and the conventional nitrogen application amount of the optimal nitrogen application planting area during topdressing to obtain the real-time recommended amount of the nitrogen fertilizer. According to the method, the problem of inaccurate recommendation caused by relatively large error of a single feature is avoided, and the accuracy of a recommended fertilization result is improved; and the model establishmentperiod is greatly shortened.

The invention provides a three-dimensional point cloud scene updating method based on crowdsourcing images. The method comprises the steps of acquiring a three-dimensional point cloud scene image anda first point cloud of a three-dimensional point cloud scene corresponding to the image; preprocessing the first point cloud and the second point cloud to obtain a preprocessed first point cloud and apreprocessed second point cloud, and calculating an absolute scale of the first point cloud according to the preprocessed first point cloud and the preprocessed second point cloud; registering the first point cloud and the second point cloud to generate a third point cloud; and performing surface reconstruction on the third point cloud to generate an updated three-dimensional point cloud scene. The method has the beneficial effects that the modeling cost is reduced, the acquisition of the three-dimensional point cloud scene with measurable precision reaching centimeter level is realized, andthe three-dimensional point cloud scene can be quickly and accurately updated.