277results about How to "Improve extraction speed" patented technology

The invention relates to a judgement document-based structured processing method. According to the method, a natural language processing technology and an advanced machine learning technology are adopted to automatically realize case type classification on the basis of keyword extraction of brief texts, so that structured processing is carried out through constructing case hierarchical structuresand designing an extraction rule. Through constructing and extending a related lexicon, segmenting a judgment document module, designing and determining a cluster number K and an initial cluster center, and taking an increment of a word weight as second feature selection, the improvement of a kmeans cluster algorithm is realized, and class labels of cases are obtained; and through creating different hierarchical frameworks according to different case types and combining the designed extraction rule, structured processing of judgement documents is obtained. The method is capable of rapidly realizing the structured processing of judgement documents.

The present invention provides an improved closed-loop detection algorithm-based mobile robot vision SLAM (Simultaneous Location and Mapping) method. The method includes the following steps that: S1,Kinect is calibrated through a using the Zhang Dingyou calibration method; S2, ORB feature extraction is performed on acquired RGB images, and feature matching is performed by using the FLANN (Fast Library for Approximate Nearest network); S3, mismatches are deleted, the space coordinates of matching points are obtained, and inter-frame pose transformation (R, t) is estimated through adopting thePnP algorithm; S4, structureless iterative optimization is performed on the pose transformation solved by the PnP; and S5, the image frames are preprocessed, the images are described by using the bagof visual words, and an improved similarity score matching method is used to perform image matching so as to obtain closed-loop candidates, and correct closed-loops are selected; and S6, an image optimization method centering cluster adjustment is used to optimize poses and road signs, and more accurate camera poses and road signs are obtained through continuous iterative optimization. With the method of the invention adopted, more accurate pose estimations and better three-dimensional reconstruction effects under indoor environments can be obtained.

The invention relates to coalbed gas mining technology, in particular to a method for extracting coalbed gases from coal mines by upper and lower combination. The method solves the problem of poor gas drainage effects through surface well drilling and hole drilling in the coalbed. The method comprises the following steps: constructing extraction drilled wells on the surface, constructing underground directional long drilled holes in the coalbed, crosslinking the fracture areas of the extraction drilled wells with the drainage areas of the underground directional long drilled holes, then effectively sealing the underground directional long drilled holes in the coalbed, then utilizing fracturing equipment on the surface to fracture the drainage areas drilled on the surface and the drainage areas of the long drilled holes in the coalbed via the extraction drilled wells and finally carrying out extraction. The method has the advantages of improving the extraction rate and efficiency of the coalbed gases and shortening the construction period of the coalbed gas extraction engineering.

The invention discloses a method for manufacturing a planar high-voltage ultrafast soft recovery diode. The method comprises the following steps of: oxidizing, and photoetching to form an active region and a field limiting ring; doping and pushing; manufacturing a polycrystalline silicon field plate; performing platinum diffusion; thinning; forming an N-type buffer layer; performing contact doping and annealing; and metalizing. The method can be used for manufacturing an ultrafast soft recovery diode chip which is low in cost and short in recovery time and has high-voltage resistance, low leakage current, low forward voltage drop and soft recovery characteristics.

The invention discloses a hierarchical screening method of violent videos based on a multiplex mode, belonging to the technical field of multimedia processing. The method comprises the following steps: firstly effectively eliminating some health videos by rapidly screening audio characteristics; selecting candidate scenes and videos which are possibly the violent videos; performing secondary screening on the candidate scenes by using video movement detection and face detection; extracting keyframes from the residual scenes; extracting image characteristics; and matching with a support vector machine so as to screen out the violent videos accurately. The method can effectively solve real-time requirements for network monitoring by using the characteristic for rapidly processing audio frequency, and the judgment time is less than 1% of video length; and the candidate scenes are subjected to subsequent video accurate classification, thus ensuring the accurate requirement of video classification with the classification accurate rate of no less than 95%.

The invention provides a method for extracting coal bed gas by negative pressure on the ground through gas-coal simultaneous extraction aiming at a waterless coal bed in an ore field. The extraction of the coal bed gas sequentially comprising three stages, namely pre-extraction before coal mining, extraction during coal mining, and extraction after coal mining; and in the method, in the well drilling process, nonmetal sleeves are descended into an old top section and a coal bed section of a vertical well, steel sleeves are descended in other layers, the nonmetal sleeves arranged in the coal bed section avoid hidden danger of sparkles caused by collision of a coal cutter and metal sleeves in subsequent coal mining, and the nonmetal sleeves arranged in the old top section reserve channels for extraction in a worked out section after the rock and shaft below the old top section are collapsed. In the method for extracting the coal bed gas, one well has three purposes, namely gas is pre-extracted in the coal bed section before coal mining, air return and exhaust are performed in the coal mining process and gas in the worked out section is extracted by negative pressure after coal mining, and the utilization rate of a coal bed gas well is improved; and the gas-coal simultaneous extraction is realized, the safety production of a coal mine is ensured, and multi-win effects of efficiently extracting the coal bed gas, utilizing new energy, protecting environment and the like are achieved.

The invention provides a V2X-based roadside unit system and an information distribution method thereof. The system mainly comprises a communication network management module, a message proxy module, an information security component module, a log management component module, a message bus interface module and the like. An RSU (roadside unit) message proxy module is responsible for parsing, priority scheduling and distribution of messages, sent by an application layer through a message bus interface, to a message bus. The RSU communication network management module is responsible for shieldingprotocol difference of a platform and RSU auxiliary communication equipment, and a unified communication calling interface is provided for middleware message proxy. According to the system and the method in the invention, heterogeneous network multi-source information from a road sensor, a vehicle, a cloud platform, a differential positioning base station and the like is managed through the V2X RSU in a unified way, a unified communication calling interface is provided for an upper-layer application, and the application development efficiency and the function expandability are improved; the messages are scheduled on the basis of a priority policy so as to ensure that the information with relatively high information tightness is preferentially sent.

The invention provides a blood vessel extraction method. The blood vessel extraction method comprises steps that S1, a blood vessel center line in an original blood vessel image is extracted, a to-be-segmented blood vessel area is acquired according to the center line, the center line is set as an initial segmentation result, and boundary values of the blood vessel at the original image and at least one characteristic image are initialized and taken as blood vessel model conditions; S2, the present segmentation result is taken as a seed point, whether original image values and at least one characteristic image value of neighborhood points adjacent to the seed point satisfy the present blood vessel model condition is determined, if yes, the neighborhood points are taken as candidate blood vessel segmentation results. Through the method, blood vessel extraction can be more accurately extracted.

The invention discloses an antibacterial cotton textile fiber preparation method, which is characterized by comprising: immersing 5 g of a fabric in a caustic soda solution with a mass fraction of 5-30% for 15-60 min at a temperature of 20-35 DEG C, taking out the fabric, placing into 100 g of a chloroacetic acid aqueous solution with a mass concentration of 10%, taking out after 30-45 min, sealing the fabric in a polyethylene plastic bag, placing in a 30-80 DEG C oven, carrying out a reaction for 1-3 h, washing 3-5 times with 50 ml of an ethanol and water mixing solution with a volume ratio of 80:20, drying, dispersing 0.6 g of sodium alginate, 0.7 g of chitosan and 2.2 g of a coptis chinensis extract in 10 g of distilled water, carrying out mixing stirring for 15 min at a temperature of 30 DEG C to obtain the chitosan solution of microcapsule, immersing the modified cotton textile fiber in the chitosan solution of the microcapsule for 30-45 min, taking out, immersing 6 g of a modified starch cross-linking agent, carrying out a reaction for 0.5-1.5 h, adding 0.3 g of a material C, carrying out a reaction for 1 h at a temperature of 90 DEG C, drying for 5-15 min at a temperature of 150 DEG C, carrying out water washing 3-5 times, and drying so as to obtain the antibacterial textile. The prepared textile of the present invention has the lasting antibacterial effect.

The invention discloses a method for extracting key phrases based on a lexical chain, relating to a method for extracting the key phrases and aiming at solving the problems that the traditional method for extracting key words cannot exactly reflect subject information described by articles and the traditional method for extracting the key phrases based on the lexical chain has lower accuracy for extracting the key phrases and lower coverage for subject information of documents. The method comprises the following steps of: firstly, acquiring word meanings; secondly, eliminating the ambiguity of words by using a dictionary 'Hownet'; thirdly, constructing the words eliminating the ambiguity into the lexical chain and acquiring a plurality of strong chains; fourthly, selecting center words from the strong chains to form a center word set; and fifthly, calculating the concurrence rate among different center words in the center word set and selecting the center words the concurrent rate of which is greater than the extraction threshold set by a user as key phrases. By using the invention, the subject information of the documents can be effectively reflected, the accuracy for extracting the key phrases can be improved and the subject information of the documents can be effectively covered by fewer key phrases. The invention is applied to the filed of key word extraction.

The invention provides an unbalance excitation force compensation method of an LMS adaptive filtering bearingless motor. The method comprises the following steps: firstly, measuring the rotating mechanical angular velocity Omegar and the real-time radial displacement x(k) of a motor rotor, and constructing dual reference signals of an LMS filter according to the Omegar and the x(k); inputting the x(k) and the dual reference signals into the LMS filter, adjusting the weight of the filter by use of a weight adjustment algorithm containing a simple step length factor adjustment function, and extracting the optimal estimation value of an unbalance vibration displacement signal of the rotor; and performing rotor synchronous revolution transformation, closed-loop feedback and inverse rotor synchronous revolution transformation on the optimal estimation value in sequence, thereby obtaining an unbalance vibration compensation control force signal. The adaptive variable-step length factor adjustment function is capable of adjusting a critical tracking error absolute value between the adaptive fast adjustment stage and the slow adjustment stage of the step length factor by use of a parameter c. The unbalance excitation force compensation method is small in calculation amount, and capable of overcoming unbalance centrifugal excitation force generated due to the mass eccentricity of the rotor; as a result, the suspension control accuracy of the rotor and the unbalance displacement tracking extraction and convergence rates are increased.

The invention discloses an astragalus polysaccharide extraction method, comprising the following steps: adding ethanol to soak astragalus root powder; adding water to carry out thermal extraction; carrying out the enzymolysis of filter cakes and enzyme destruction; concentrating extract, depositing the extract with ethanol, drying the extract and the like. Compared with the traditional method, by using ethanol solution as strippant, taking water as extraction solvent, and adopting desorbing-internal boiling method, the method provided by the invention has significant improvement in extraction speed and extraction ratio, keeps the activity of the astragalus polysaccharide, and has the advantages of short time consumption, high extraction rate, less impurity and low energy consumption in the extraction process, therefore, the method is a green, safe and efficient . The method is also suitable to the extraction and the preparation of other plant-derived medicine polysaccharide.

The invention relates to a coal bed gas exploiting technique, in particular to a process for pumping and exploiting coal bed gas by together using ground well drilling and underground hole drilling, solving the problem of low efficiency of the traditional method for pumping and exploiting the coal bed gas. The process comprises the following steps of: (1) carrying out well drilling on the ground; (2) carrying out long hole drilling in underground coal beds; (3) making drilled holes and ground drilled wells crosslink and communicate in the appointed points by utilizing an orientation technique of a long hole drilling machine; (4) effectively closing the underground long drilled holes; (5) fracturing the ground drilled wells and the underground long drilled holes by utilizing fracturing equipment on the ground; (6) after fracturing, closing the ground drilled wells; and (7) opening the closed ends of the underground long drilled holes and then installing relevant pumping and discharging pipelines connected with underground pumping and discharging systems. The process improves the pumping and exploiting speed and efficiency of the coal bed gas, shortens the construction period of coal bed gas pumping and exploiting engineering and saves the construction investment of the coal bed gas pumping and exploiting engineering.

The invention provides a ginger detergent processed under ultrahigh pressure and a preparation method thereof and belongs to the daily chemical application field. The ginger detergent processed under ultrahigh pressure is used for solving the problem of skin irritation caused by an antiseptic substance added in the existing detergent. The ginger detergent mainly comprises the following raw materials: sodium fatty alcohol polyoxyethylene ether sulfate, sodium dodecyl benzene sulfonate, coco fatty acid diethanol amide, a chelating agent, a solubilizer and deionized water containing a ginger extract component. Due to the adoption of deionized water containing the ginger extract component, the ginger detergent has a good effect of removing fishlike smell, and as the ginger has a natural fragrance, no any perfume needs to be added in the raw materials; ultrahigh processing technology is carried out in the step of preparing the deionized water containing the ginger extract component and the step before bottling; as a result, the yield and extraction speed of the effective component can be increased obviously, and the utilization efficiency of ginger is improved. The ginger detergent is processed at ultrahigh pressure, so that the sterilization effect can be achieved and the use of the antiseptic substance is avoided.

The invention relates to a human face feature extraction and classification method. The method includes the following steps that feature dimension reduction is performed on human face images through a 2D-PCA method, and a high-dimensional image matrix is converted into a low-dimensional image matrix; the low-dimensional image matrix is converted into one-dimensional column vectors; according to the one-dimensional vectors of images of a training set, an intra-class divergence matrix S<W> and an inter-class divergence matrix S in the training set are solved, and eigenvalue decomposition is performed on the S<W> and the S respectively, wherein the eigenvalue decomposition formulae are as follows: and ; D<alpha> is used for estimating , D<beta> is used for estimating , and the formulae and can be obtained; the column space W<1> of and the column space W<2> of are respectively solved, and the optimum projection space W=[W<1>,W<2>] of a feature extraction algorithm with two stages of LDA based on 2D-PCA is obtained; the low-dimensional image matrix in the first step is projected into the optimum projection space W, and then feature vectors of the images are obtained; classifier training is performed on the feature vectors obtained in the sixth step through an SVM+NDA model, and a final human face classifier is obtained.

The invention discloses a superjunction IGBT with a shield gate and a manufacturing method thereof. The superjunction IGBT comprises a semiconductor substrate, a cell region and a terminal protectionarea. The semiconductor substrate comprises a second conductivity type collector region, a first conductivity type field termination layer and at least one first conductivity type epitaxial layer. Thecell region comprises a plurality of cells connected in parallel with each other, which includes a plurality of cell trenches and a gate conductive polysilicon and a shield gate filled in the cell trenches, A seventh oxide layer is arranged on both sides and sidewalls of the notch of the cell trench opposite to the gate conductive polysilicon, a fifth oxide layer is arranged between the gate conductive polysilicon and the shield gate, and a fourth oxide layer is arranged on the bottom and sidewalls of the cell trench opposite to the shield gate. A P-pillar is also provided in the first conductivity type epitaxial layer, one end of the P-pillar is connected to the second conductivity type well layer, and the other end extends toward the first conductivity type field termination layer. Theinvention adopts the shielding gate structure, reduces the Miller capacitance, thereby reducing the switching loss.

A fingerprint information processing method and apparatus in which the method includes: obtaining a fingerprint image; calculating an average value of shading values of pixels in a specific region based on a pixel with respect to each pixel of the fingerprint image, performing a first processing of calculating a sum of average values of shading values of pixels included in an expanded region while gradually expanding the specific region, and generating a first processing image for the fingerprint image using a first processing-performed value for each pixel; and forming a window including a predetermined region in the first processing image, and selecting feature points among pixels in a window region while moving the window.

The invention discloses a traditional Chinese medicine extracting and filtering device which comprises a reaction kettle and an inner container. The inner container is arranged in an inner cavity of the reaction kettle, a heating cavity is formed between the reaction kettle and the inner container, the top end of one side face of the heating cavity is penetratingly connected with an air incoming pipe, an air shift box is arranged at the top wall of an inner cavity of the inner container, a stirring rod is arranged in the inner cavity of the inner container, the top end of the stirring rod penetrates the top walls of the air shift box and the reaction kettle and is fixedly connected on a motor shaft of a motor, a plurality of stirring pipes are uniformly arranged on the stirring rod, the bottom end of the inner container is penetratingly connected with an agent discharging pipe, the other end of the agent discharging pipe penetrates the bottom wall of the reaction kettle and is penetratingly connected with a filter, and a plurality of supporting legs are arranged at the bottom end of the reaction kettle uniformly. By arranging the heating cavity to wrap the whole inner container to enable the outside of the same to be heated uniformly, lowering of quality of medicine liquid due to charring of traditional Chinese medicine caused by local overheating can be prevented effectively; the medicine liquid is heated from the inside of the same through the stirring rod and the stirring pipes, so that heating effect of the medicine liquid is further improved, and extraction speed and quality of traditional Chinese medicine are improved substantially.

The invention discloses a metal identification method based on information entropy and belongs to the image identification technology field. The method is characterized by firstly, carrying out multistage wavelet transformation on a sample image; then, using a K mean value algorithm to carry out identification for one time; for a sample which is not identified, analyzing a transformed multistage wavelet transformation entropy value; and using a decision tree algorithm to carry out the identification for the second time. In the method, through multilevel identification, an identification speed is increased and simultaneously algorithmic accuracy is guaranteed. An algorithm speed is fast, an accuracy degree is high, identification efficiency is high too and a real-time performance requirement can be satisfied. The method supports parallel processing. Equipment needed during identification is simple and environmental-protection. The method has wide applicability. The method is suitable for identification of copper, aluminum alloy and stainless. Under the condition that there are sufficient samples, because hardness and colors of different metal and alloy are different to some extent, the method can be used for distinguishing various kinds of non-ferrous metal.

The invention discloses an enterprise internal auditing system based on a block chain technology. The enterprise information processing system comprises an upper application layer, a development interface layer, a block chain platform layer, a data layer, an infrastructure layer and a terminal department, the upper application layer and the development interface layer are in bidirectional connection, and the open interface layer and the block chain platform layer are in bidirectional connection. The invention discloses an enterprise internal auditing system based on a block chain technology. Aplurality of auditing modes can be set. The actual requirements of enterprise departments are met. Actual services of different enterprises can be met. The standardization and the unification of audit services in an enterprise can also be completed. According to the method, the original data information can be obtained accurately, the efficiency and quality of obtaining audit evidences by internal audit of enterprises are improved, and the reliability of obtaining the evidences by the internal audit is effectively guaranteed, so that the efficiency and effect of executing audit work in the enterprises are improved, and the storage mode of the networked internal audit data is improved.

The invention relates to a microblog spread prediction method based on user influences and contents. The method includes the following steps that firstly, by means of the scrapy technology, personal information of two users, the forward relationship between the two users and forwarded microblogs are crawled; secondly, the user influences are extracted through the RankPage influence analysis technology to form a user authority predictive factor; thirdly, the percentage of fans forwarded microblogs in all published microblogs in unit time is used, and a fan forward activeness predictive factor is extracted; fourthly, the microblog contents are subjected to importance analysis through the TF-IDF word weight technology to extract a microblog importance predictive factor; fifthly, the extracted forward relationship is divided into 10 microblog forward training sets and microblog ignoring training sets through the snowball sampling method; sixthly, the training sets are trained by means of a monitored Bayesian network till classifier parameters are converged. By means of the method, the prediction accuracy of Sina microblogs forwarded by special fans can be improved.

The invention relates to a novel ICF (initial confinement fusion) pellet detection device and a positioning method. The device comprises a laser, a beam expander, a first bean splitter, a second beamsplitter, a third beam splitter, a plane mirror, a to-be-detected pellet, a lens, a CCD image sensor, a computer, a high-power LED and a diaphragm. The pellet is moved at the internal of 10 mu m, a backlight projection image in each position is collected, image preprocessing work such as filtering and aperture extraction is performed, the acutance of an area near a bright ring is calculated, and the position with the largest acutance meets the object-image conjugate relation. The pellet is moved in the interval of 10 mu m at the interval of 1 mu m continuously after being moved to the position. The steps of collection, image preprocessing and acutance calculation are repeated, the pellet is moved to the new position with the largest acutance, and positioning is finished. The novel pellet detection device and the positioning method are applicable to conventional cylindrical vacuum space and can realize automatic positioning quickly and accurately.

The invention relates to a preparation method for a solid-phase micro-extraction fiber bundle, and relates to preparation of the fiber bundle. The preparation method comprises the steps of (1) preparation of single SPME (solid-phase micro-extraction) fibers based on an integral material: mixing a monomer, a crosslinking agent, a pore-forming agent and an initiator to prepare a uniform solution, then injecting the uniform solution into a capillary which is 5-20cm long and of which the inner diameter is 0.2-0.8mm and the wall thickness is 0.05-0.2mm, sealing two ends of the capillary, performing polymerization, removing a section, which is 1-3cm long, of the capillary, putting the integral material fiber into a soxhlet extractor, performing extraction through a solvent, or putting the fiber into the solvent, and soaking the fiber until no impurity is detected from the liquid so as to obtain the single SPME fibers based on the integral material; (2) preparation of the solid-phase micro-extraction fiber bundle: combining 2-6 of the single SPME fibers based on the integral material on the capillary to obtain the solid-phase micro-extraction fiber bundle. The preparation method is simple, high in extraction capacity, high in extraction speed, low in cost and easy to operate.

The embodiment of the invention provides a video data processing method and video data read-write equipment. The video data processing method comprises the following steps: receiving a data read request message, wherein the data read request message comprises initial time of video data to be read; according to the initial time, determining position information of an initial I frame of the video data to be read in a video data storage region; and according to the position information, reading the video data to be read from the video data storage region. According to the embodiment of the invention, by receiving the data read request message comprising the initial time of the video data to be read, the accurate position information of the initial I frame corresponding to the video data to be read in the video data storage region is rapidly determined and the video data to be read is read from the video data storage region according to the position information so as to improve the extraction speed and the working efficiency of the video data.

The invention provides a method for extracting regular noise from single record web pages. The method comprises the steps of firstly converting multiple record web pages into document object model (DOM) trees, and classifying the DOM trees according to structures; then aligning and integrating the DOM trees belonging to the same type to obtain site section style trees; and positioning approximate positions of web page text headline nodes and approximate positions of web page text main body nodes in the site section style trees, and finally extracting the regular noise in front of texts, in the texts and after the texts according to the web page text headline nodes and the web page text main body nodes. By means of the method, space resources required by construction of the site section style trees is decreased, possible extraction leakage situations are decreased, and extracting speed is accelerated. In addition, an extracting result has high accuracy, good effect is obtained, and the reliability is high.

The invention belongs to the technical field of additive manufacturing on-line measurement, and discloses an on-line detection method of the contour accuracy of a molding layer in the laser selectivemelting process. The on-line detection method includes the following steps that S1, a model of a to-be-machined part is sliced, and an auxiliary image is generated; S2, powder spreading is conduced, and then powder is melted and molded in a laser selective mode; S3, an image of a basic molding area is collected, and a contour of a segmented image is extracted; S4, the image contour is subjected tothree-dimensional reconstruction to obtain an image actual contour; S5, the image actual contour Cr is compared with a contour of a corresponding slicing layer to analyze the accuracy, if the requirements are met, the step S6 is executed, and otherwise, machining is completed; and S6, whether machining of the to-be-machined part is completed or not is detected. According to the on-line detectionmethod of the contour accuracy of the molding layer in the laser selective melting process, the range of image contour extraction can be effectively and accurately delimited, and then the number of iteration times in the calculation process is greatly decreased; and thus the extraction accuracy of the image contour reaches the sub-pixel level, and accurate and rapid contour detection is realized.

The invention relates to a liquid sampling apparatus including: a sampling pipeline, a liquid discharge pipeline and a plunger pump which is connected to the sampling pipeline and liquid discharge pipeline, so that a liquid is sucked via the sampling pipeline and is discharged via the liquid discharge pipeline. The apparatus also includes: a peristaltic pump which is arranged on the sampling pipeline for pumping the liquid into the plunger pump; a controller for control coordinated action of the peristaltic pump and the plunger pump; a filter disposed between the peristaltic pump and the plunger pump; and a second pressure sensor and a first pressure sensor that are arranged at the front and the back of the filter. The controller controls the operation of the sampling apparatus according to pressure signals. In the apparatus, the liquid is sucked in a double-power manner, so that it is guaranteed that the interior of the plunger pump is under micro-positive pressure always. The structure not only increases suction speed of the liquid and reduces suction time, but also can avoid bubbles in the plunger pump and increases sampling precision.

The invention provides a video playback method and device. The method comprises the steps that a video playback command is received, a corresponding video storage region is searched out, the corresponding video storage region comprises at least one group, and each group comprises n I-framed data; according to the video playback command and the reverse order of the groups, all the groups are subjected to the following operations that the n I-framed data in the groups in the corresponding video storage region are extracted in the forward order to be stored in a playback buffer memory space, and frames are taken out reversely in the playback buffer memory space and sent to a video play device for playing. Every time when a group of I-framed data is extracted, the frequency of rotation of a video storage device hard disk is reduced, the data reading speed is improved, pressure and damage of the hard disk are relieved, and the playback performance is improved.