317results about How to "Comprehensive description" patented technology

The invention discloses a target following method in a complex interference scene, comprising the steps of: detecting a moving target by confirming the initial position of every moving target which appears in video, detecting with self-adapting Gaussian background models and separating a moving target area with a morphologic treating method, modeling for target feature by acquiring a plurality ofmatching feature messages for each detected moving target area, fusing a plurality of feature messages and building a target feature template, searching the matching area by searching the area which is matched with the target feature template in a specific area in current video as a target following location of current video, updating the target feature template by updating the feature of the target feature template of current video to finish the self-adapting update for the target feature template after locating the target according to the feature in the current video target area and the feature of the target feature template in the former three videos.

The invention discloses a method for predicting service life of a product by an accelerated degradation testing based on degradation distribution non-stationary time series analysis. The method comprises the following steps of: 1, acquisition and preprocessing of test data; 2, parameter estimation of degradation distribution; 3, time series modeling of degradation distribution parameters; 4, accelerated degradation modeling based on the degradation distribution; and 5, service life prediction based on degradation distribution. The method can macroscopically describe degradation statistical regulation of all samples under accelerated stress, overall analyze the time series of the degradation distribution in an accelerated degradation process, can extrapolate the degradation distribution under the accelerated stress to normal stress to obtain product reliability reflecting the product accelerated degradation random process volatility regulation and service life relationship prediction, improve the credibility on estimation result of service life prediction and reliability, and save more time and is more efficient compared with performance degradation prediction under normal stress level.

The invention discloses a method for evaluating the reliability of a stepping stress quickened degradation experiment based on a time sequence. The method utilizes a correlation coefficient stationary sequence analytical method to describe non-equidistant stochastic information in stepping stress quickened degradation experiment data, establishes a regressive-non-equidistant autoregression degradation model on the basis of obtaining the respective advantages of a deterministic regression model and a stochastic correlation coefficient stationary sequence autoregression model, combines a grey theory to predict a product degradation trend and forms a function and a curve for predicting the life and evaluating the reliability of the stepping stress quickened degradation experiment based on the time sequence. The method obviously decreases the sample quantity of a product and shortens the time of the degradation experiment, thereby saving a large amount of expenses and resources, enhancing the fitting accuracy of the degradation model and enhancing the reliability of a reliability evaluating result.

The invention provides an automobile automatic driving test case generation method based on scenes and tasks, and the method comprises the following steps: 1), selecting typical traffic behaviors in acomplex and multi-element road traffic scene, and constructing a test task library; 2) defining terms of a traffic scene and related elements thereof, describing concepts, attributes and hierarchicalrelationships of scene elements, and constructing an ontology model of the road traffic scene; 3) generating a traffic scene based on the traffic scene model and the traffic behavior task library, the traffic scene being generally composed of a plurality of traffic participants, a specific time-space road environment and a traffic scene, and generally including a plurality of driving tasks; and 4) further describing a traffic scene in detail by defining a parameter range and correlation of variables in the state space, and selecting representative discrete values for combination by applying boundary value analysis, equivalence class division and other methods in software testing to generate a plurality of test cases.

The invention discloses a prediction method of multi-functional parameter accelerated degradation testing product life based on multivariate hybrid time sequence analysis, belonging to the technical field of accelerated test evaluation, comprising the steps of test data acquisition and pretreatment; multi-functional parameter time sequence modeling; multi-functional parameter degradation failure prediction; multi-functional parameter degradation accelerated modeling and multi-functional parameter product life prediction; the problem that the prediction is inaccurate since the mutual influence among a plurality of property parameter degradation processes of the product is unconsidered in the single property parameter degradation prediction of the product is solved; a multivariate hybrid time sequence model is put forward in considering monotonicity degradation trend, randomness and cyclical fluctuation in the product multi-functional parameter degradation process and the relativity of the characteristics among the property parameter; a failure criterion of the product multi-functional parameter degradation is put forward; the description to the multi-functional parameter degradation process is more comprehensive and the accuracy of the prediction result of multi-functional parameter accelerated degradation testing life is improved.

The invention discloses a method for real-time monitoring of the face image quality of a customer collection terminal, comprising the following steps: Step 1, face image collection; Automatically detect and segment the exact position of the face in the face image; step 3 comprehensively evaluates the face quality, and performs intelligent prompts and real-time intelligent monitoring of the image quality through the face image quality detection system for the acquisition process of the client's face image. The invention can effectively improve the accuracy of subsequent image comparison and reduce the false alarm rate (False Alarm Rate), thereby ensuring the reliability of identification information and making the intelligent monitoring system based on face recognition truly practical.

The invention provides a Chinese electronic case text analysis method and system. The method includes following steps: obtaining a case data set by employing case texts; separating value variables andtext information through regularization processing, determining value information meaning according to context description, and searching and structurally saving time information of different categories by employing a regular expression; performing word segmentation and part-of-speech tagging on the texts by employing natural language processing, performing further screening with the combinationof medical entity identification, and determining positions and types of medical key vocabularies in the texts; analyzing and screening the medical key vocabularies and information; simulating cases by employing the above related results; converting content of the texts to value vectors; adding similarity tags to the case texts; learning marks of the case texts; and screening the cases with the similarity on new case samples from the case data set according to the marks and training results. According to the method and system, for different evaluation standards, different similar case text clusters can be found from existed case texts for each new case text.

The invention discloses a rotor bearing system accident comprehensive information analysis method and a device thereof, wherein one section of a rotor is radial and vertically mounted with two sensors X, Y, the axle end of the rotor is axially mounted with a sensor Z, the radial and axial information of the rotor motion transmitted by the sensors are fused to analyze the motion state of the rotor bearing system, the axis center line position and the axis center track of the rotor in the X-Y, Y-Z and Z-X planes of a dimension space are described and displayed visibly to completely and accurately describe the spatial motion information of the rotor, and the information contains the unique diagnosis information as dimension space eccentricity, an attitude angle, a precession direction and vibration vector direction and the like, thereby improving the ability for visually recognizing the accidents of a rotary machine.

The invention discloses a multi-sensor-based vehicle-mounted environment recognition system. The system comprises an omni-directional vision module, an IMU module, a GPS module, a radar module and a data fusion module, wherein the omni-directional vision module is used for acquiring environment three-dimensional information in a 360-degree range around a vehicle, the IMU module is used for acquiring accelerated speed and angular speed information of the vehicle, the GPS module and the IMU module jointly work to acquire position and posture information of the vehicle, the radar module is used for acquiring front target position information of the vehicle and has the characteristics of all weather and all time, and the data fusion module is used for fusing environment information acquired by each sensor to more accurately recognize a vehicle driving environment. The invention further discloses an omni-directional vision module, which utilizes each stepping motor to control a corresponding binocular camera to perform a pitching motion in order to acquire omni-directional scene images of an observed environment; and an image processing module is used for in real time collecting and processing scene images acquired by each group of the binocular cameras.

Gate-driving circuitry of a thin film transistor array panel is formed on the same plane as a display area of the transistor array panel. The gate-driving circuitry includes driving circuitry and signal lines having apertures. Thus, a sufficient amount of light, even though illuminated from the thin film transistor array panel side, can reach a photosetting sealant overlapping at least in part the gate-driving circuitry. The thin film transistor array panel and the counter panel are put together air-tight and moisture-tight. Consequently, the gate-driving circuitry can avoid corrosion by moisture introduced from outside. Gate-driving circuitry malfunctions can also be reduced.

The embodiment of the invention provides a human face detection method and a human face detection device. The method comprises the following steps of extracting a plurality of face features of different hierarchical networks from a to-be-detected face image through a pre-trained convolutional neural network model so as to obtain a plurality of face feature vectors corresponding to different hierarchical networks; fusing the plurality of face feature vectors into a face feature vector; conducting dimensionality reduction treatment on the face feature vector after fusion treatment, and obtainingtwo face feature vectors in the same dimension; subjecting one face feature vector of the two face feature vectors to face detection treatment to obtain a face detection result; and subjecting the other face feature vector of the two face feature vectors to gesture estimation treatment to obtain a gesture estimation result. According to the invention, an image can be described more abundantly based on face detection features and gesture estimation features, and the accuracy is higher. The error rate of subsequent human face detection is reduced, and multiple related tasks can be executed at the same time. The performance of a single task is improved and the processing efficiency of the task is improved.

The invention discloses a complex mechanism dynamic cascading reliability modeling method based on a colored stochastic Petri network. The method comprises the following five steps that firstly, a system is layered from top to bottom according to the structure, functions and operational profile of the system, the upper layer of the system is obtained by means of functional layer division, and the lower layer of the system is obtained by means of structural layer division; secondly, the fault modes and the influences of all the layers of the system are analyzed through an FMEA method, a logic relation between correlated faults is given, fault criteria influencing the functions are determined, and the hierarchy and transmissibility of the faults of the system are clarified; thirdly, a CSPN reliability submodel of each layer is established according to the logic relation between the faults and the functions; fourthly, the CSPN submodels of all the layers are connected organically by means of transition substitution to establish an overall reliability model of the system; fifthly, a color function, a node function, an arc expression function and a defense function of each layer are determined, and the reliability of the system is analyzed and evaluated by means of Monte Carlo simulation.

The invention relates to an apparatus and method for measuring elastic turbulence characteristics of a viscoelastic polymer in a porous medium, wherein a side connector of a fluorescent microscope of the apparatus is connected to a CCD (couple charged device) camera, a saline container, a polymer solution container and a synthetic oil container that are connected with an oil/water container are parallelly connected between an inlet of a micro metering pump and an inlet of a two-dimensional mesh microscopic simulation model, the two-dimensional mesh microscopic simulation model is provided with an inlet pressure sensor and an outlet pressure sensor, and a light guide arm of a micro field particle image speed-measuring system is positioned right above the two-dimensional mesh microscopic simulation model; all the micro field particle image speed-measuring system, the fluorescent microscope, the inlet pressure sensor and the outlet pressure sensor are connected to a data graphic data station. The apparatus and method are capable of effectively recognizing elastic turbulence of viscoelastic polymer solution in porous medium seepage, and providing qualitative and quantitative description for elastic turbulence characteristics.

The invention discloses a method for diagnosing mechanical faults of a distribution switch based on a two-dimensional feature vector cluster of vibration signals. The method comprises the following steps that an acceleration sensor is arranged near a switch operating mechanism and the vibration signals of the distribution switch on different states in the processes of gate switch-on and gate switch-off are obtained; two-dimensional information feature vectors of vibration signal energy and gravity center frequency are extracted through HHT band-pass filtering and conducted on the obtained vibration signals, the normalized two-dimensional feature vectors stand for effective feature quantity of the mechanical state of the distribution switch, and different kinds of mechanical fault diagnosis of the distribution switch can be achieved through a vague K mean value cluster (FKM).

The invention discloses a time domain identification method of random dynamic loads. The time domain identification method is characterized by comprising the following steps that (1) a strain sensor and an acceleration sensor are arranged on an engineering structure, and dynamic strain samples and acceleration response samples of the structure are obtained through measurement; (2) an accurate structure finite element model is established for the engineering structure, model condensation is conducted, and a condensed finite element model matched with the freedom degree of a dynamic strain response signal obtained through actual measurement is obtained; (3) on the basis of a dynamic strain sample set obtained through actual measurement, the deterministic component of structure random dynamic response is obtained through the KL decomposition technology; (4) deterministic components of the random dynamic loads are identified by the adoption of the direct inversion method and the Tikhonov regularization method; (5) the statistic characteristics, namely, the average time history and the variance, of the random dynamic loads are obtained.

The invention discloses a malicious domain name detection method based on a knowledge graph, which comprises the following steps: S1, extracting key features for constructing a domain name reputation knowledge graph according to domain name information features; S2, integrating the information contained in the domain name by using the obtained existing data, and providing a data source for the creation of the knowledge graph; amplifying domain name knowledge by the data through a knowledge graph construction engine; enabling the knowledge graph construction engine to construct a domain name reputation knowledge graph by using the obtained domain name data; and S3, for a given specific domain name, giving a prediction result by using the multi-source information and a plurality of method judgment models. The method has the advantages that 1, domain name information is richer in dimension, and contained information is more complete; 2, outbreak growth of domain name data can be better coped with, and mass data can be processed more quickly through construction of a knowledge graph; and 3, a relational network is formed by the domain name information and the behavior pattern, and the information in the relational network is deeply mined by utilizing the sensitivity of the knowledge graph to the relational network.

This invention is based on the time and events sequence to analysis attacking detection methods and devices involved in exchange for the function of the network. It is a methods and devices to prevent removing data from the data transmission channel without permission .The method provides a complex text-based attack description language to make the user can amend the built-related characteristics, and add new features of related events. The present invention include: basic rules of the incident, the rules of the association, first level detection engine, correlation analysis engine. The whole process of attacking are described more comprehensive and reasonable taking into account the time factor and the order of events Such description and testing which distinct the basic incidents more carefully are more in line with the requirements of detecting attacks. The invention also describes the relation between alarm incident and not alarm incident.

The invention relates to an adaptive generation system for image semantic description, and belongs to the technical field of image semantic description. The system comprises an image reader, an encoder, a decoder and a semantic description display. The image reader is used for acquiring an image to be described semantically. The encoder comprises a global feature extraction unit, a local feature extraction unit and an image feature combination unit. The decoder comprises a single-layer or multi-layer neural network. The neural network comprises an LSTM part, an Attention part and an MLP part,and a semantic description model is generated. The Attention part adopts a self-adaptive attention mechanism. The decoder generates words and sentences of image semantic description by utilizing the semantic description model according to the combination information output by the encoder. The semantic description display is used for outputting and displaying words and phrases of the image semanticdescription. The focus key point of the image is determined, higher-level semantic information is mined, and the detail information problem describing words or sentences is perfected.

The invention relates to an image semantic description generation method, and belongs to the technical field of image semantic description. The method comprises the following steps: S1, extracting global features and local features of an image; S2, combining the global features and the local features to obtain combined information; and S3, focusing image features and focusing texts according to the combination information through an adaptive attention mechanism to generate a semantic description model, and generating words and phrases of image semantic description by utilizing the semantic description model. According to the method, the problems that the focus key point of the image is determined, higher-level semantic information is mined, and detail information describing words or sentences is perfected in the existing image semantic description technology are solved.

The invention discloses an identity authentication method and device, and belongs to the field of image processing and mode identification. The method comprises the steps of: obtaining a face image sample pair, wherein one face image is an identity card photograph, and the other face image is an image collected on site; utilizing a face image pyramid algorithm to process the obtained face image sample pair, and obtaining a face image pyramid structure; calculating gray scale gradient directions of the face images of all scales in the face image pyramid structure of the obtained face image sample pair, and obtaining gradient direction pyramid GOP characteristics; and according to the GOP characteristics of the obtained face image sample pair, calculating and judging whether the face image sample pair belongs to the same person. By adopting the identity authentication method and device, the interference of factors such as illumination, age and expression is effectively avoided, and the accuracy of identity authentication is substantially improved.

The invention belongs to the field of network security, in particular to an automatic malicious code analysis method and an automatic malicious code analysis system. The automatic malicious code analysis system comprises a client browser, a control center, a sample receiving and registering module, a sample processing module, a report generation module and a client information database. The automatic malicious code analysis method comprises the following steps of: calculating a HASH value of a sample file; comparing the HASH value of the sample file with that of an originally analyzed sample, judging whether the sample file is analyzed before, and if the sample file is analyzed before, directly returning the old analysis result; if the sample file is not analyzed, calling a virus killing engine to scan viruses, judging whether a malicious code is a known malicious code, and if the malicious code is the known malicious code, acquiring information of the malicious code, such as the name, the type and the hazard grade; if the sample file has the unknown malicious code, comprehensively and dynamically analyzing the sample file, wherein the malicious code can be classified as a computer malicious code and a mobile phone malicious code according to different platforms on which the malicious code runs. By the automatic malicious code analysis method and the automatic malicious code analysis system, known malicious codes in a computer and an intelligent mobile phone can be quickly and effectively identified; and various malicious operation behaviors of unknown malicious codes during running can be accurately analyzed.

The invention relates to a judgment method of power grid power-saving level. The judgment method comprises establishment of judgment index set of 'power-saving power grid' : all kinds of energy consumption of each link in the process of power grid operation are analyzed from power transmission and distribution, and all kinds of energy consumption are classified into for categories according to different properties, namely energy consumption of power grid structure, energy consumption of power grid operation, energy consumption of power grid equipment, and energy consumption of power grid management. Corresponding judgment index and formulas of the index are given to each kind of energy consumption, and energy consumption index system of assessment of power grid energy-saving features is formed based on all above. Power grid energy-saving level is judged by the formulas of the index to obtain energy consumption level of each link. Energy consumption structure of each link and energy-saving potential of each link is disclosed. Low efficiency link and reasons of low efficiency are found through identification and comparison. Construction of an evaluation index system of energy-saving power grid provides direct basis for the analysis and evaluation of energy-saving level of power grid.

The invention discloses a three-dimensional face recognition method based on a vertical and horizontal local binary pattern on the mesh. The method comprises: pretreatment including face region cutting, smoothing processing, and pose normalization is carried out on a three-dimensional face model and all faces are placed under a pose coordinate system; seven profile lines of a face is extracted from a semi-rigid region of the three-dimensional face model and uniform resampling is carried out on each profile line to obtain 28 key points to represent a facial curved surface; surrounding adjacent regions of the key points are extracted, adjacent regions of the key points are represented by using a descriptor of a vertical local binary pattern on the mesh and a descriptor of a horizontal local binary pattern on the mesh, feature fusion is carried out to obtain a descriptor of a vertical and horizontal local binary pattern on the mesh; and then a 3D face is identified by using an LC-KSVD2 dictionary learning classification algorithm. According to the invention, local shape changes of a uniform face mesh surface due to factors like an expression can be described comprehensively; and the identification performance is excellent.

The invention discloses a night visual monitoring method based on information fusion to aim at the characteristics of low contrast, low Signal-to-Noise and low information load of night visual monitoring. At first, image enhancing is carried out on image sequences collected at night; then moving target detection is carried out on the image sequence which is enhanced; then the illumination area of the night image sequence is extracted and finally image fusion is carried out on an interested target (night moving target, the night illumination area) and a background image under the same scene in the day. The night visual monitoring method based on information fusion has very important meanings to improve the capacity of the intelligent monitoring system of next generation to pick out the night monitoring scene. The night visual monitoring method based on information fusion is used for intelligent visual monitoring at night and provides a practical scheme for the night monitoring, thus leading the monitoring system to be capable of really understanding what is happening under a scene; the night visual monitoring method based on information fusion is used for detecting a night moving target to improve the detection quality of the moving target and is used for human-computer interaction to help the human beings to have more comprehensive and clearer description on the monitoring scene.