240results about How to "Real-time identification" patented technology

The invention discloses a man-machine interactive system and a real-time gesture tracking processing method for the same. The method comprises the following steps of: obtaining image information of user side, executing gesture detection by a gesture detection unit to finish separation of a gesture and a background, and automatically determining a smaller rectangular frame which surrounds the hand as a region of interest in the image information through a vision algorithm; calculating the hand profile state of each frame in a video sequence by a gesture tracking unit; executing validity check on hand actions according to the calculated hand profile state, and determining the gesture action finished by the user; and generating a corresponding gesture action control instruction according to the determined gesture action, and making corresponding feedbacks by a three-dimensional user interface according to the gesture action control instruction. In the system and the method, all or part of gesture actions of the user are sensed so as to finish accurate tracking on the gesture, thus, a real-time robust solution is provided for an effective gesture man-machine interface port based on a common vision sensor.

The invention discloses a human body identification and tracking method applied to a security system. The method comprises the steps that a video input device acquires the image information of each frame of a video; a processing unit detects a human body target according to the information of each frame of the video; the processing unit carries out face detection according to the human body targetdetection result; the processing unit carries out feature extraction on a detected face area to acquire a face feature vector; the processing unit matches face feature data and a face feature in a database, and assigns an ID to the detected human body target according to the matching result, and at the same time assigns the ID to the human body target corresponding to the face; and the processingunit compares the ID of the detected human body target in each frame and connects the positions of human body targets with the same ID in different frames to form the movement trajectory of the humanbody target corresponding to the ID. According to the method provided by the invention, real-time face detection and identification can be realized, and target lost or failed tracking can be prevented.

The invention discloses a coal-rock interface identifying method and system based on an image. The method comprises the following steps of: acquiring multiple color images of coal and rock on a coal mining working face; extracting a vector based on an image characteristic serving as a sample characteristic vector specific to each color image to obtain a known sample set of coal and rock; and establishing a coal-rock classifier model by adopting a Fisher linear judging method and taking the known sample set of the coal and rock as a training sample set. In the working process of a coal mining machine, a color image of the coal and rock which is cut by using a drum is acquired in real time, and the extracted characteristic vector is input into the coal-rock classifier model to identify a coal-rock type. The system consists of a light source module, an imaging module, a processing module and an anti-explosion shell. The coal-rock interface identifying method and the system provided by the invention have the characteristics of simple structure, easiness for distributing, high suitability and the like, the coal-rock type cut by using the drum can be automatically identified in real time, and reliable coal-rock interface information is provided for automatic heightening of the drum of the coal mining machine.

The invention relates to a hierarchical classification method for internet flow, comprising the following steps: firstly, coarse grained classification is conducted as follows: a flow statistical property method based on machine learning is adopted for quickly classifying network flow, the network flow is divided into applied categories with different characteristics, if the flow distinguished by the coarse grained classification needs to be used, the flow can be output directly, and if the flow needs further classification, next step is conducted; and secondly, fine grained classification is conducted as follows: classification is further conducted on the applied categories distinguished by the coarse grained classification, the network flow can be intelligently, accurately and effectively identified in real time, good integrity and expandability are realized, and the requirements on the internet flow classification with different application targets and classification grain size can be met.

The invention relates to a real-time recognition method for baby cry. The method comprises the steps of (1) recording baby cry signals and establishing a voice bank for preprocessing; (2) when the cry starts, sampling to obtain real-time digital signals after the baby cry is acquired, and extracting a real-time characteristic vector sequence after complete voice sections in the real-time digital signals are conducted for framing and windowing in sequence in a fixed time window; (3) matching the real-time characteristic vector sequence with a GMM (Giant Magnetostrictive Material) template to obtain a category with maximum likelihood, taking the maximum category as a recognition result, and voting on the current recognition result; and (4) after the cry completes, judging the categories of the baby cry according to the voting results, finishing the recognition, and carrying out baby status processing. The real-time recognition method for the baby cry can quickly judge the type namely the hunger type, the sleepy type, the wetting type and the pain type the baby cry belongs to in real time, so that a nurse can be aided to react in time, and recognition performance is high.

The invention discloses a human body action recognition method and a mobile intelligent terminal. The human body action recognition method comprises the steps that human body action data are acquired for training so that feature extraction parameters and template data sequences are obtained, and the data requiring performance of human body action recognition are acquired in one time of human body action recognition so that original data sequences are obtained; feature extraction is performed on the original data sequences by utilizing the feature extraction parameters, and the data dimension of the original data sequences is reduced so that test data sequences after dimension reduction are obtained; and the test data sequences and the template data sequences are matched, and generation of human body actions corresponding to the template data sequences to which the test data sequences are correlated is confirmed when the successfully matched test data sequences exist. Dimension reduction is performed on the test data sequences so that the requirements for the human body action attitudes are reduced, and noise is removed. Then the data after dimension reduction are matched with the templates so that calculation complexity is reduced, accurate human body action recognition is realized and user experience is enhanced.

The invention discloses an online detection method for a circular plate casting based on machine vision, and the method comprises the steps: obtaining the front and back images of a circular plate casting sample through collecting the image of the circular plate casting sample, and performing the preprocessing of the images; detecting the size of the sample according to the collected sample image,performing the preliminary discrimination of defects, and performing the next detection; performing the amplification of a gap through the image processing, extracting the coordinates, and achievingthe detection marking of the sample gap; performing the detection of the rag defects of the sample according to the comparison between the degree of deviation of the edge from the reference and a setthreshold value; and finally completing the discrimination of whether the sample is qualified or not. The method can achieve the real-time quick recognition of the gap and rags of the circular plate casting, and is suitable for a high-speed online detection system.

The invention provides a cigarette smoke detection method based on video monitoring. The method is divided into a classifier training phase and a cigarette smoke detection phase. In the classifier training phase, geometrical characteristics and Hog characteristics of all prepared positive and negative sample sets are extracted, a support vector machine is used for training the extracted characteristic vectors, and a classifier is generated; and in the cigarette smoke detection phase, firstly, a Gaussian mixture model is utilized to obtain a foreground motion region from a video image obtained by a monitoring camera, then noisy points and cavities are removed by means of morphological filtering, projection histograms in X and Y directions are counted according to the characteristics of cigarette smoke, a precise interested region is obtained, finally characteristics of the interested region are extracted and combined into characteristic vectors, the characteristic vectors are input into the classifier, and whether the interested region has cigarette smoke is judged. The method provided by the invention has the advantages that the interference resistance is high, the detection sensitivity is high, the false alarm rate is low, the cigarette smoke can be recognized in realized, the position of the cigarette smoke is determined, and the cigarette control work of indoor public places is facilitated.

The invention discloses a driver road rage real-time identification and warning method based on facial features. The method comprises the following steps: first, using Kinect to track the face of a driver from real-time video data, and extracting a RGB image and a depth image of the driver face; and then, carrying out graying, image size normalization, median filtering for noise removing and other preprocessing operations on the two types of image information, inputting the image information into a trained CNN model to extract features, inputting the extracted features in a Softmax classifier for classification, fusing the classification results of 30 continuous frames of the RGB image and the depth image of the face, calculating the probability values of two channel confidences, comparing the two confidences, taking the result of highest confidence as a driver facial expression final identification result, and warning the driver according to the identification result. According to the invention, RGB-D driver face feature information is fused, and angry feelings of drivers can still be identified in real time and with high accuracy and warning can be carried out in a night or bad driving environment.

The present invention provides a fake plate vehicle identification and inspection system and method based on a vehicle-borne RFID tag and a dual-frequency information collection base station. The RF / video collection base station in a road in the system collects the RFID electronic tag number, the vehicle plate number, the vehicle color, the vehicle photo and other information of a running vehicle in the road and sends the information to a background database system in real time through a network, then the database compares whether the RFID electronic tag number, the vehicle plate number and the vehicle color are consistent with a database record or not, if not, the vehicle is identified as a vehicle with a fake plate, and the information of the vehicle plate number, the vehicle color, a vehicle model, the vehicle photo and the like are sent to nearby traffic police on duty to stop the vehicle. According to the system and the method, an RFID electronic tag is collected through the radio frequency technology, the vehicle plate number and the vehicle color are identified through the radio frequency identification technology, the real-time background comparison and warning information sending are carried out, and the real-time and accurate inspection of a vehicle with a fake plate can be realized.

The invention discloses a high-resolution intelligent network camera array system with global synchronization function. The system adopts global synchronization signals to realize the synchronous acquisition of a camera array, comprises at least two intelligent network cameras used for image acquisition, wherein the working mode of one camera is configured as a main mode, while the other cameras are configured to be in a slave mode; the main-mode camera produces and provides global horizontal synchronization signals and vertical synchronization signals for the slave-mode cameras; and every intelligent network camera is connected with a central computer through a network and is used for receiving control commands in real time and transmitting image data so as to realize distributed computing function. The system has the advantages of satisfying the real-time high-performance processing for acquired data, realizing the global synchronous shooting of the camera array, supporting large-scale array expansion, realizing the control of the central computer over a plurality of working modes of the intelligent cameras and realizing high-resolution video-image data shooting.

The invention discloses a kind of array sensor for gas and the preparing method. The sensor uses the aurum, platinum or argentum as electrode; the RuO2, MoSi2, W or Pt is as heating membrane material; the random one or several kinds of the ZnO, SnO2 or Fe2O, or ZnO, SnO2 or Fe2O3 and Pt0.1wt%-3wt%, Pd0.1wt%-3wt%, Ni1wt%-30wt%, TiO21wt%-30wt%, WO31wt%-40wt%, Co2O3 1wt%-15wt%, V2O51wt%-30wt%, Al2O31wt%-20wt%, MnO21wt%-30wt% is as the sensing layer material. It gets the gas sensing array on the alumina, aluminium nitride or silicon base sheet with anti high temperature, electric insulating and heat conducting by the screen printing technology. The gas sensing array is fixed on the basal seat by the axis to form the gas sensing array sensor by ultrasonic hot press welding electrode leg wire. The sensor has the low production cost, low power cost, good mechanism stability. Each unit has the good gas sensing character and the process is proper for the big batch production.

The invention discloses a method and device for identifying abnormal flow, and relates to the technical field of computers, wherein the method comprises the steps of: counting the log report data volume in a current monitoring period, and judging whether the log report data volume accords with a pre-set abnormal judgement condition or not; and, after the fact that the log report data volume accords with the pre-set abnormal judgement condition is determined, inputting the log report data in the current monitoring period into a pre-constructed decision-making tree, and performing abnormal identification on the log report data in the current monitoring period according to the decision-making tree. Due to the above steps, the abnormal flow can be identified effectively in real time; and thus,occupation of the abnormal flow on storage resources and computing resources of a server side can be reduced.

The present invention relates to a remote repair system for an elevator in which a repairer uses a mobile application. The purpose of the present invention is to provide the remote repair system for the elevator in which the repairer uses the mobile application, capable of allowing the repairer of the elevator to search an operating state of the elevator and various related information by connecting the remote repair system by a medium of a mobile phone. [Reference numerals] (100) Internet;(20) Main server;(30) WAS server;(40) Mobile G / W server.

The invention provides a continuous character gesture track recognizing method based on a structural dynamic time domain wrapping algorithm (STDW) for solving the problems in the prior art. Through the automatic cutting problem of continuous gesture tracks, extraction of angular point new characteristics, the recognition of the STDW and the processing between character gesture tracks, natural, real-time and accurate recognition on continuous character gesture tracks is achieved. Compared with the prior art, the interaction mode is more natural, the constraint of users is smaller, and the requirement of man-machine interaction is met. Compared with the traditional situation that multiple gesture actions are more diversified in type, under the condition that accuracy and real-time performance are considered at the same time, the recognition effect of an original algorithm is improved by means of the STDW formed by combining a character gesture track structure and a dynamic time wrapping algorithm, the recognition rate is increased, and the method is high in accuracy and good in man-machine interaction experience and can meet the requirement for real-time performance.

The invention discloses a credit sale network lottery drawing system and an implementation method thereof. According to the system, credit accounts of customers are bound to lottery drawing eligibilities, so that the powerful combination of credit rewards and a lottery drawing system is realized; meanwhile, the credit accounts of the customers, basic information of the customers and a reward item list are updated in real time, and the customer lottery drawing eligibility is identified in real time, so that the fairness of lottery drawing results is guaranteed, and the possibility of more rewards and chaos rewards are reduced.

The invention discloses a method and equipment for identifying a mining water source. The method comprises the following steps of: establishing a water source sample warehouse according to the chemical characteristic of a known water source; and determining identification indexes, and judging the type of an unknown water source. Correspondingly, the invention also discloses equipment for identifying the water source. Through the method and the equipment, chemical components of underground water in mine water-bearing strata are measured simply and quickly, and the unknown water burst source can be judged accurately in real time so as to prevent water hazards in coal mine from occurring, avoid economic loss and casualties which are caused by water burst accidents and guarantee the safety production of the coal mine.

The invention provides an identification method for judging home invasion in real time, belonging to the security field of video surveillance. The method comprises the steps of: collecting a real-time surveillance video or image and preprocessing the video or the image; and identifying the invasion time, number, body figures and threatening degree of the invaders in the surveillance region by zoning and segmentation processing. According to the method, the problems that the current video or image surveillance system is only used for static forensics and cannot identify a warning condition or give an alarm without the assistance of other sensors are solved; and specific information of the warning condition such as the number, body height, body width and threatening degree of the invaders can be analyzed by the method according to the surveillance video or image, and the surveillance effect and the alarming effect for the surveillance region can be greatly improved.

The invention discloses a coal and rock identification method based on image discrete multi-wavelet transform. The method includes obtaining a group of coal sample images and rock sample images under the same image-forming conditions according to kinds of coal and rocks in a coal mine to be identified; extracting m samples from the images respectively and capturing sub-images f1, f2,..., fm and g1, g2,..., gm which have the same size and contain no backgrounds; performing a first-level Geronimo Hardin Massopust (GHM) multi-wavelet transform respectively on f1, f2,..., fm and g1, g2,..., gm and calculating multi-scale texture energy distribution vectors based on transform domain data for all the sub-images to obtain average multi-scale texture energy distribution vectors Vcoal and Vrock of the coal samples and the rock samples; during a coal and rock classification and identification stage, collecting to-be-identified coal and rock images under the same image-forming conditions, capturing a sub-image fx with the same size with the samples, performing the first-level GHM multi-wavelet transform on fx, and calculating the multi-scale texture energy distribution vector Vx; and determining coal and rock object types according to relations among the Vx, Vcoal and Vrock. According to the method, the coal and rock object types are identified through images, devices for collecting and processing images are convenient to arrange, the reliability and the identification rate are high, and upgrade maintenance of software and hardware is convenient.

The invention discloses a velocity estimation and friction identification method of a four-wheel turning / driving vehicle. The velocity estimation and friction identification method comprises the steps that firstly, a vehicle body coordinate system O-XY is established; secondly, a vehicle kinetic equation is established; thirdly, a wheel kinetic equation is established; fourthly, a wheel velocity equation is established; fifthly, a wheel slip rate equation is established; sixthly, the longitudinal force and lateral force of a wheel are obtained; seventhly, a friction model is parameterized; eighthly, a kinematical equation describing the relation of the vehicle velocity, the yaw velocity and the acceleration is established; ninthly, an observer is designed under the situation that detectable signals z comprise the longitudinal acceleration ax, the lateral acceleration ay, the yaw velocity r, the wheel angular velocity wj and the steering wheel rotation angle delta i; tenthly, persistent excitation conditions are defined. According to the velocity estimation and friction identification method, the nonlinearity self-adaptation observer for estimating the vehicle longitudinal velocity and the lateral velocity is established, the tire road surface friction parameters are established in real time according to the change of road surface conditions, and the velocity is estimated more accurately.

The invention discloses an underwater robot state and parameter joint estimation method based on self-adaption unscented Kalman filtering (UKF). The method comprises building an expanding reference model of an underwater robot, and enabling the reference model to have a dynamical model of the underwater robot and a fault model of a thruster; adopting a main filter of the self-adaption UKF to deliver and update expanding states composed of poses and speed of the underwater robot state and faults of the thruster according to pose information detected by a position sensor, and timely estimating speed information of the underwater robot and fault message of the thruster; and simultaneously adopting an accessory filter of the self-adaption UKF to timely estimate noise information of a system according to innovation information of the main filter. The underwater robot state and parameter joint estimation method has good instantaneity, can conduct joint estimation on states and parameters of the system, and can achieve high estimation accuracy when priori information of process noise and measurement noise is unknown.

The invention discloses a robot positioning system and a reflection device identification method thereof. The robot positioning system comprises a mobile robot and at least two reflection devices. The robot is provided with a laser emitting device and a laser receiving device. The reflection devices can reflect received laser, each reflection device is provided with reflection members, and the reflection devices have different numbers of reflection members. When laser emitted by the laser emitting device scans any reflection device, the laser receiving device outputs a signal every time the laser receiving device receives reflected laser of the reflection device, and the robot can directly determine which reflection device is currently scanned according to the number of received signals. According to the technical scheme provided by the invention, the robot can rapidly identify the reflection devices in the movement process so that the coordination position of the robot can be determined by utilizing the reflection devices, the operation load of a processing system is mitigated, and the positioning error of the robot is reduced.

The present invention discloses a speaker rapid identification method and a system based on the growing and clustering algorithm of models. The method comprises the processes of model training and model identification. The model training process comprises the steps of acquiring voiceprint signals from multiple persons including a speaker, pre-treating all the voice-print signals and extracting voiceprint characteristic parameters to form a plurality of models, and conducting the adaptive classification for all models based on the growing and clustering algorithm of models. The model identification process comprises the steps of acquiring voice signals from a speaker, pre-treating the voice signals, extracting voiceprint characteristic parameters, calculating the characteristic parameters of to-be-identified voice signals and the likelihoods of all model types, selecting a model type for the to-be-identified voice signal based on the maximum likelihood principle, calculating the likelihood scores of all models in the above selected model type, and adopting a model of the highest likelihood score as an identification result. According to the technical scheme of the invention, the operation of matching the to-be-identified voice characteristics with all models is not required, so that the method is short in matching period and good in real-time performance. The method can be well adapted to large-scale model bases.

The invention relates to a monitoring, tracing and early warning regulation and control system and method for virus transmission in a building. Indoor and outdoor environment parameters of a buildingcan be monitored in real time, physiological indexes of a human body are obtained in real time, the health state of the human body is judged, and the moving track and the droplet propagation track ofeach person are obtained, so that suspected infected persons and people in close contact with the suspected infected persons are timely and accurately recognized, and meanwhile response measures are taken for the suspected infected persons and people in close contact with the suspected infected persons. Besides, the propagation risk of the current indoor environment can be evaluated, the indoor environment is ventilated and purified according to the propagation risk evaluation result, the suspected infected person and the local area of the close contact person are effectively ventilated and purified, and the propagation risk is reduced. Building indoor environment parameters, human body health characteristics and virus transmission risk assessment are combined, the building indoor environment state can be accurately judged, suspected infected persons and people in close contact with the suspected infected persons can be recognized in real time, and overall ventilation purification andlocal purification treatment on the indoor environment are achieved.

The invention relates to the field of intelligent food delivery, and provides a take-away system and method of a robot. The method comprises the steps of building communication connection with an equipment end in response to a connection request of the equipment end, and navigating and positioning to a designated position by the robot according to a service instruction sent from the equipment end;selecting an optimal path and moving to the designated position by the robot according to path planning; building communication connection between the robot and a user terminal; sending information and a box opening link instruction to the user terminal; and continuing to achieve delivery of a next target by the robot after a user takes out the meal until the delivery task is ended and the robotreturns to a dining room. The take-away system has the capability of real-time positioning, obstacle identification and navigation by comprehensive information, and the path can be planned by the robot when the line is deviated during the travelling process; and by the take-away method, the labor can be saved, the generation rate of delivery mistake, delivery delay and the like also can be reduced, and the intelligent level of delivery is greatly improved.

The invention belongs to the field of the non-destructive testing technology for a nuclear power station and in particular relates to an eddy-signal-recognition-technology-based full-automatic data acquisition method. The eddy-signal-recognition-technology-based full-automatic data acquisition method comprises the following steps: 1. setting the detection boundary and the data recording boundary and building characteristic signal models of eddy digital signals at the boundary positions; 2. controlling a probe to enter a to-be-detected pipeline through a pipe opening and move inwards at designated speed; 3. acquiring an eddy digital signal in real time by an eddy current instrument, and when the acquired signal is matched with the characteristic signal for detecting the boundary, stopping advancing; 4. starting to acquire data automatically, enabling the probe to move back from the to-be-detected pipeline outwards after the probe reaches the detection boundary, simultaneously, starting to acquire the signal automatically until the probe reaches the recording boundary; and 5. stopping automatic data acquisition, after the probe reaches the recording boundary, stopping data acquisition, and enabling the probe to move back out of the to-be-detected pipeline. Compared with the method that a sensor is additionally arranged at the pipe end of a push-pull device, the signal-recognition-based positioning method can save the cost and has high reliability.

The invention discloses a framework crack monitoring method and monitoring devices thereof, mainly used for realizing the fault diagnosis seriously endangering safety, such as framework cracks and the like, through detecting the generalized resonance of a framework which is aroused by random impact and crack starting and closing impact caused by the generalized resonance during the running process of a vehicle and identifying the corresponding relation between the generalized resonance and the crack starting and closing impact. The method comprises the following steps of: performing modal analysis on the framework; further selecting the monitoring point of a vibration impact compound sensor which is used for monitoring the framework; using the generalized resonance of the framework modal, which is aroused by external random impact obtained at the monitoring point, and the generalized resonance impact aroused by the generalized resonance to judge a relative product used for crack identification; carrying out narrow-band filtering, relative product quantifying and the confidence calculation and the like on the modal frequency of the generalized resonance obtained at the monitoring point to judge crack early warning. The invention also provides monitoring devices applying any one of the framework crack monitoring methods and the monitoring devices include a compound sensor, a vibration impact processor and a crack analysis diagnosis computer system.

The invention discloses a method and a device for detecting network performance abnormity, equipment and a storage medium, and relates to the field of computers and communication. The method comprises the following steps: determining a confidence interval of a KPI by using historical data of a key performance index KPI for representing network performance; monitoring the KPI to obtain real-time data of the KPI; and determining whether the KPI is normal or not by using the confidence interval and the real-time data. According to the embodiment of the invention, operation and maintenance personnel can observe the current network state anytime and anywhere, obtain abnormal information immediately and adjust the network in time, so that the network always keeps the optimal state.

The invention relates to a coin identifying or discriminating device and method, especially relating to the discrimination of the true coins from the false coins. The method vertically slides the tested coin along a guiding groove down onto a substrate, collects a sound signal emitted by the coin as striking the substrate through sound sensor, directly sends the sound signal or sends the sound signal amplified by a power amplifier to a frequency spectrum analyzing device for making frequency spectrum analysis so as to obtain an inherent frequency of the tested coin, and compares the inherent frequency with a standard data in a frequency spectrum standard memory for making frequency spectrum discrimination; then uses a pressure sensor to measure the weight of the tested coin and sends the generated force signal amplified by a power amplifier into a weight analyzing device and calculates to obtain the weight of the tested coin and compares it with a standard data in a weight standard memory for making weight discrimination; then by an operational policy maker, synthetically analyzes and obtains the final discrimination result. The discriminating device has simple structure, lower manufacturing cost, and high identifying accuracy, applied to identification of various coins and having a wide application prospect.

The invention relates to a self-adaptive cruise method based on pavement adhesion coefficient recognition. The method comprises the following steps that 1), a camera shoots a road picture in front ofthe vehicle, and the pavement adhesion coefficient mu of the current running road surface is obtained according to a prediction neural network model; 2), a vehicle-mounted radar is used for measuringthe distance S between the vehicle and the front vehicle and the vehicle speed v1 of the front vehicle, and the theoretical braking distances X1 and X2 of the two vehicles are calculated according tothe vehicle speed v2, S and v1; and 3), the minimum distance for preventing rear-end collision is assumed as the minimum safety distance X during emergency braking of the current vehicle, X is calculated according to X1, X2 and S0, wherein S0 is a distance margin, comparing X and S, when S is larger than X, the vehicle is in a safe state, the vehicle is accelerated to a set vehicle speed or keepsa set vehicle speed, and when S is less than or equal to X, the vehicle is in a dangerous state, and the vehicle is reduced to the state of S is larger than X. The self-adaptive cruise method can be applied to auxiliary driving and can automatically adjust the distance between vehicles so as to protect the vehicle to be in the safe state.