32results about How to "Meet identification requirements" patented technology

The invention discloses a natural environment recognition method based on time-frequency domain statistical feature extraction. The method comprises the steps of (1) collecting various types of natural environmental sounds such as the sound of an engine, a car horn and wind noise, and establishing a sound sample library, (2) carrying out window framing processing on a sound sample signal, (3) extracting the statistical features of all frame signal in a time domain, (4) marking the type of a sound source to which feature vectors belong, and establishing a sample feature library, (5) training a feature vector by using a support vector machine, and establishing a training model, (6) extracting the feature vectors of a target sound, (7) carrying out matching classification on the feature vectors of the target sound by using the support vector machine, and (8) providing an recognition result. According to the method, the disadvantages of traditional sound LPCC and MFCC feature extraction methods in the aspect of time and frequency combination are made up, and the type to which various types of sounds belong can be judged.

Disclosed is an iris recognition method for a mobile terminal. A face camera and an iris camera are provided in the mobile terminal. The iris recognition method comprises: controlling the face camera to shoot a face image of a target; detecting whether the face image shot by the face camera meets a preset condition; and when it is detected that the face image shot by the face camera meets the preset condition, controlling the iris camera to shoot an iris image of the target. The present invention solves the problem in the related art that the shooting position cannot easily meet the iris recognition requirements during iris recognition, and achieves an effect that the iris recognition requirements can be easily satisfied during shooting of the iris image.

The invention relates to a face recognition system. The face recognition system comprises a face detection and positioning module, a standardization module, a feature extraction module and a face recognition module in sequence. The face recognition system has the advantages that the recognition accuracy can reach above 90%, and recognition requirements are basically met. The face recognition system is good in real-time performance, convenient to carry, and capable of being popularized to the fields of dynamic image tracking, motion detection and the like through modification of programs.

The invention discloses an identity card recognition method in a complex scene based on an OCR technology. The method comprises the steps: S1, carrying out the recognition of the feature points of anidentity card based on an SIFT algorithm, and determining the position coordinates of the identity card in a real scene; s2, performing positioning detection on a target area of the identity card in the real scene based on the obtained position coordinates of the identity card, and determining a to-be-identified character area; s3, extracting text information in the to-be-identified text area; andS4, correcting the extracted text information to obtain accurate identity card information, and completing identity card recognition. The method solves the problems of positioning and recognition ofthe identity card in complex environments such as dark light, overexposure, fuzzy shooting and the like; the character minimum distance correction algorithm is added to the recognition result, so thatthe recognition accuracy is remarkably improved, a certain result error correction function is achieved, and the recognition result is more reliable.

The invention relates to a character cutting method, a device, a computer readable storage medium and an electronic device. The method comprises steps of acquiring a picture including N characters and calculating position information of each character in the picture; based on a selection frame with a preset parameter, determining M target characters in the N characters in the picture; and according to the position information corresponding to the M characters, cutting the picture to obtain single characters corresponding to the M target characters. According to the invention, efficiency and accuracy of character identification can be improved.

The invention discloses an engineering machinery recognition method and a recognition device based on improved MFCC (Mel Frequency Cepstrum Coefficient) sound features. The method comprises a step of building a sound sample library, a step of acquiring sample sounds, a step of dividing the sample sounds in regions, and a step of extracting sound sample features. According to the step of extracting the sound sample features, corresponding sample features are extracted from each about 1.7s sound folder respectively, and the step comprises substeps: pre-treatment is carried out, a Hamming window is added, FFT transformation is carried out, through a Mel triangular filter bank, a mean and a variance are solved, transpose is solved, FFT transformation is carried out, logarithmic operation is carried out through a filter bank, and DCT transformation is carried out. The method also comprises a step of building a sample feature model and a step of recognizing a target sound. According to the engineering machinery recognition method and the recognition device based on the improved MFCC (Mel Frequency Cepstrum Coefficient) sound features, false alarms can be reduced, and the recognition performance can be improved.

The invention relates to a system and method for intelligently identifying and shooting a multi-person scene. The shooting system comprises a shooting module and a feature recognition module. The camera module can shoot a main picture image and a shooting area image and is set to shoot a specific shooting area; and the feature recognition module is used for executing a character feature recognition algorithm on the shot image, and comparing features obtained after processing of the character feature recognition algorithm with a feature library to obtain a feature label. The method has extremely high adaptability to a multi-person scene, quickly extracts character features, accurately realizes character feature recognition, meets the intelligent analysis shooting requirements of multiple scenes such as remote teaching, conference sign-in, teaching attendance and behavior analysis, and greatly improves the information presentation capability and the analysis capability of a main picture.

The invention discloses a manifold learning-based data dimensionality-reduction method and a device. The method comprises the steps of firstly, dividing a to-be-detected human face image into sub-images according to an equal-strip rule and converting the image in a corresponding sub-mode; secondly, subjecting the divided to-be-detected human face image to the data dimensionality-reduction treatment; thirdly, classifying the low-dimensional vectors of the dimensionality-reduced image according to K sub-modes in a training set to obtain K recognition results; fourthly, calculating the K recognition results according to the weighting method to obtain a final recognition result of the to-be-detected human face image. Namely, the to-be-detected human face image can be recognized as one human face image in the training set.

The invention relates to the technical field of speech recognition and discloses a method for training a speaker recognition model. The method comprises steps of obtaining a to-be-classified corpus; inputting the to-be-classified corpus into a preset classification model, and determining whether the to-be-classified corpus belongs to the vertical field or not; and under the condition that the to-be-classified corpus belongs to the vertical field, training a first speaker recognition model by using the to-be-classified corpus. According to the method, the obtained to-be-classified corpora are inputted into the classification model for classification, and the to-be-classified corpora belonging to the vertical field are utilized to train the speaker recognition model, so accuracy of the speaker recognition model related to the vertical field is improved, and speaker recognition requirements for the vertical field can be met. The invention further discloses a device for training the speaker recognition model, electronic equipment and a readable storage medium.

The invention discloses a fingerprint identification method and device, a storage medium and a terminal. The method comprises the steps of collecting a to-be-identified fingerprint image of a to-be-identified fingerprint; extracting prior data from the to-be-identified fingerprint image, wherein the prior data comprises an outer contour of the to-be-identified fingerprint; comparing whether the priori data is consistent with standard data or not, wherein the standard data comprises an outer contour of a standard fingerprint; and when the comparison result is consistent, identifying the to-be-identified fingerprint at least based on the to-be-identified fingerprint image. Through the scheme provided by the invention, the fingerprint identification efficiency can be improved, the calculationamount is reduced, and computer resources are saved.

The invention provides COB mounting-based optical module circuit board and marking point design. A finished circuit board and marking points are included; bonding pads can be placed on the finished circuit board; the finished circuit board is provided with a six-layer board, including a top layer, a second layer, a third layer, a fourth layer, a fifth layer and a bottom layer from top to bottom in sequence; the marking points can recognize the position of the finished circuit board; the number of the marking points is five; in the optical module design of the COB scheme, a bare chip is directly mounted on the circuit board, and the circuit board replaces a conventional transitional block, so that mounting equipment is required for chip mounting, and routing equipment is required for routing a gold wire; and therefore, compared with a conventional circuit board, the circuit board used for the COB scheme needs to satisfy recognition requirement of the mounting equipment on the marking points, and the binding pulling force requirement of the wiring equipment on bonding pad routing.

The invention belongs to the technical field of signal processing, and particularly relates to a space-time block code blind identification method and system based on a space-time correlation matrix, and the method comprises the steps: receiving a signal through synchronous sampling of a transmitting end and a receiving end in a communication system, carrying out the preprocessing of the received signal, and obtaining a plurality of groups of received signal samples; setting a value range of the delay quantity, and calculating a space-time correlation matrix corresponding to each group of received signal samples under each delay quantity in the value range of the delay quantity; performing redundancy-eliminating whitening processing on elements in the space-time correlation matrix under each delay amount in the received signal sample, and constructing a space-time block code feature vector by calculating a Frobenis norm square value of the space-time correlation matrix after the whitening processing; and taking the space-time block code feature vectors as model input, and performing classification identification on space-time block code features by using the trained BP neural network. The method can meet the identification requirement of the STBC type in MIMO communication, can be directly applied to a non-cooperative MIMO system and the like, and has good applicability.

The invention discloses a method and device for constructing an interactive interface, and electronic equipment. The method comprises the steps of acquiring a service configuration protocol; interpreting the service configuration protocol, and generating a service configuration protocol model; saving the service configuration protocol model to the current operating environment, and generating a data factory; in response to an interface initialization event, reading initialization parameters, and converting the initialization parameters into a memory data factory of the data factory; generatinginteractive interface data according to the interface initialization event; constructing the interactive interface according to the data factory and the interactive interface data.

The invention discloses an efficient identity recognition device, which includes a control box, a device bracket, four identity recognition terminals and a top box body; a controller, a WiFi module and a memory are arranged in the control box; the identity recognition terminal includes a square shell, Fingerprint recognition module, camera, temperature sensor and finger warming mechanism. The high-efficiency identification device adopts four identification terminals installed in four directions respectively, so that four-way identification can be carried out at the same time, which meets the identification requirements of a large number of people during rush hour; the temperature sensor can be used for identification The temperature measurement is carried out at the same time, so as to record the body temperature of each person correspondingly; the panel wiping mechanism can be used to wipe the finger pressing panel of the fingerprint recognition module to meet the cleaning requirements of the finger pressing panel and prevent the spread of bacteria.

The invention discloses a service processing method, device and system for an application. The method is applied to an identification gateway. The method may comprise: the identification gateway parses an initial service message sent to a user terminal by an application server, and acquires a feature code at a pre-set position of the initial service message; the identification gateway acquires a control strategy corresponding to the feature code in accordance with a pre-set identification rule according to the feature code, and the identification rule comprises a correlation between the feature code and the control strategy; and the identification gateway performs corresponding service control on the initial service message and a subsequent service message between the application server and the user terminal according to the control strategy.

The invention discloses a behavior recognition method and device and a computer storage medium. The method comprises the following steps of: acquiring a to-be-identified image, and inputting the to-be-identified image into a pre-trained neural network; obtaining outputs of the neural network, wherein the outputs comprise a first output representing the probability that a character behavior is normal in the to-be-recognized image, a second output representing the probability that the character is performing a first behavior, and a third output representing the probability that the character isperforming a second behavior; and determining the behavior of the person according to the outputs. Visibly, according to the embodiment of the invention, the behavior of the person in the to-be-identified image can be determined by using the pre-trained neural network, and specifically, the neural network can extract rich visual features and can pay attention to the specific behavior of the personin the to-be-identified image, so that the accuracy of behavior identification is higher. Moreover, the behavior recognition method provided by the embodiment of the invention can meet the real-timerequirement, and can perform real-time operation, thereby meeting the recognition requirements of various different application fields.

The present disclosure relates to a character cutting method, device, computer readable storage medium and electronic equipment. The method includes: obtaining a picture containing N characters, calculating the position information of each of the characters in the picture; determining M of the N characters in the picture based on a marquee with preset parameters target characters: according to the position information corresponding to the M target characters, cutting the picture to obtain individual characters corresponding to the M target characters respectively. The disclosure can improve the efficiency and accuracy of character recognition.

The embodiment of the present invention discloses a voice interaction method and device. The method includes: acquiring voice input information when it is detected that the voice input function is awakened; and separating the voice input information according to the pre-stored first noise spectrum information Operation, generating vocal spectrum information and second noise spectrum information; determining the first volume amplitude according to the second noise spectrum information, and determining the average vocal amplitude according to the vocal spectrum information; according to the first volume The amplitude relationship between the amplitude and the average amplitude of the human voice provides matching voice reminder information. According to the technical solution of the embodiment of the present invention, through the obtained noise information of the surrounding environment, the voice information input by the user is separated, and the spectrum information of the human voice is generated, so as to realize the volume guidance of the voice input by the user and ensure that the voice of the user conforms to the The recognition requirements of the voice input function have improved the accuracy of voice recognition.

The invention discloses a high-speed photographic apparatus component and a vehicle exemption label self-service issuing machine. The high-speed photographic apparatus component comprises a high-speedphotographic apparatus side face, and a high-speed photographic apparatus supporting plate is fixed on a top face of a high-speed photographic apparatus base; a lens for shooting is fixed on the high-speed photographic apparatus supporting plate; a light shielding plate is further installed on the inner side of the high-speed photographic apparatus side face, and a fill-in light is installed on the inner side of the light shielding plate; light shielding mounting plates are respectively arranged on two ends of the light shielding plate, the light shielding mounting plates are respectively assembled and fixed with one end of a light shielding rotating shaft, the other end of the light shielding rotating shaft is rotationally assembled with a high-speed photographic apparatus connecting plate, the light shielding plate is assembled and fixed with one end of a light shielding spring, and the other end of the light shielding spring is assembled and fixed with the inner side of a high-speed photographic apparatus side plate; the light shielding plate is further assembled and fixed with one end of a light shielding stay cord, the other end of the light shielding stay cord leans againsta light shielding guide wheel after penetrating through the high-speed photographic apparatus side plate and is connected and fixed with a light shielding winding wheel; and the light shielding winding wheel is rotationally assembled with a light shielding winding shaft, the light shielding winding shaft is assembled and fixed with a second fixing plate, and the second fixing plate is assembled and fixed with the high-speed photographic apparatus side plate.

The invention discloses a reservoir dam body seepage safety online automatic monitoring system. A water level measurement apparatus comprises a water level wheel, a measurement cable, a float, a hoisting wheel, a gear set, a worm, a gear wheel, a print wheel set and a constant torque motor, wherein the circumferential surface of each print wheel of the print wheel set refers to two-dimensional code identifications corresponding to values from 0 to 9, which are respectively arranged on ten arc segments from 0 to 9. The water level monitoring apparatus drives print wheels through the water level wheel and replaces digits on the print wheels with two-dimensional codes such that the water level data visual recognition rate is greatly improved; and the water level and the temperature of each piezometer tube of a reservoir dam body, the water level and the temperature of a measurement point are uploaded to a host computer through GPRS, the host computer, according to the obtained water level of each piezometer tube of each segment of the dam body, quantitatively analyzes influence factors of the water levels of the piezometers, then through combination with external temperature and climate conditions, a safety warning about a dam is given, and thus a dam accident is effectively prevented.

The invention provides a method, device, electronic equipment and sorting system for arranging identification equipment of a dry sorter. The method is applied to a dry sorter including a vibrating feeder and an identification device. The particles to be sorted are fed from the vibrating feeder. After the machine is discharged, it falls in a free fall, and the identification equipment is used to identify the particles in the free fall; when the identification equipment is arranged, according to the particle size range and the preset rotation angle of the particles to be sorted by the dry separator interval, determine the first height range and the second height range corresponding to the particle size interval, and then determine at least one target height range. When the rotation angle of the particle belongs to the rotation angle range, the angle between the maximum surface of the particle and the vertical plane of the ray irradiation direction of the identification device is smaller than the preset angle, so that the identification device arranged within the target height range can ensure that the particle is on the vertical plane. The size of the projection surface above meets the requirements of particle identification, so the accuracy of particle identification is improved.

The invention provides an identification equipment arrangement method and device of a dry separator, electronic equipment and a separation system.The method is applied to the dry separator comprising a vibrating feeder and identification equipment, particles to be separated fall in a free falling mode after being discharged from the vibrating feeder, and the identification equipment is used for identifying the particles falling in the free falling mode; when identification equipment is arranged, a first height range and a second height range corresponding to a particle size interval are determined according to the particle size interval of particles to be sorted by a dry separator and a preset rotation angle interval, and then at least one target height range is determined. When the rotation angle of the particle belongs to the rotation angle interval, the included angle between the maximum surface of the particle and the vertical plane of the ray irradiation direction of the identification equipment is smaller than the preset angle, so that the identification equipment arranged in the target height range can ensure that the projection plane size of the particle on the vertical plane meets the particle identification requirement; therefore, the particle identification precision is improved.

The embodiment of the present invention discloses a method, device and system for application business processing. The method is applied to an identification gateway, and the method may include: the identification gateway parses the initial service message sent by the application server to the user terminal, and executes the Acquiring the feature code at the preset position of the initial service message; the identification gateway acquires the control strategy corresponding to the feature code according to the preset identification rule according to the feature code; wherein, the identification rule includes the feature code and the control strategy The identification gateway performs corresponding service control on the initial service message and subsequent service messages between the application server and the user terminal according to the control policy.

The invention discloses a front door sheet assembly and a vehicle inspection-free tag self-service issuing machine with the same. The front door sheet assembly disclosed by the invention comprises a front door sheet; a front door sheet groove and a tag outlet are formed on the front door sheet; the tag outlet is formed on a front door supporting shell, and a shooting entrance and exit port is formed at a position on the front door supporting shell, which corresponds to a high-speed photographic apparatus assembly; a front door lock and a front lock rod guide plate are also mounted on the frontdoor sheet; a front lock cylinder part and a front lock driving pillar are respectively arranged inside the front door lock; the front lock driving pillar is assembled and fixed with a front lock connection plate; both ends of the front lock connection plate are respectively hinged with one ends of front lock connection rods; the other ends of the front lock connection rods are respectively hinged with one ends of front lock rods; after respectively passing through the front lock rod guide plate, the other ends of the front lock rods are clamped and assembled with front matching lock plates;and the front matching lock plates are respectively fixed on first supporting steels. According to the invention, automatic printing and getting of a six-year inspection-free tag can be implemented, so that labor intensity of window service personnel is reduced, work efficiency is improved and the front door sheet assembly and the vehicle inspection-free tag self-service issuing machine are also convenient for office clerks to handle businesses at any time.

The invention discloses a natural environment recognition method based on time-frequency domain statistical feature extraction. The method comprises the steps of (1) collecting various types of natural environmental sounds such as the sound of an engine, a car horn and wind noise, and establishing a sound sample library, (2) carrying out window framing processing on a sound sample signal, (3) extracting the statistical features of all frame signal in a time domain, (4) marking the type of a sound source to which feature vectors belong, and establishing a sample feature library, (5) training a feature vector by using a support vector machine, and establishing a training model, (6) extracting the feature vectors of a target sound, (7) carrying out matching classification on the feature vectors of the target sound by using the support vector machine, and (8) providing an recognition result. According to the method, the disadvantages of traditional sound LPCC and MFCC feature extraction methods in the aspect of time and frequency combination are made up, and the type to which various types of sounds belong can be judged.

The invention discloses an online automatic monitoring system for reservoir security. The system comprises a dam seepage monitoring device, a water level monitoring device, a dam sinking and translating monitoring device, and a wireless communication network and background management system. The dam seepage monitoring device, the water level monitoring device and the dam sinking and translating monitoring device are incorporated to and transmit the onsite monitored data to the wireless communication network and background management system respectively; and the background management system makes early warnings and determination according to the onsite monitored data; and if the monitored data reach the preset value, early warning information comes out. According to the invention, through the connection of onsite monitored data from the dam seepage monitoring device, the water level monitoring device and the dam sinking and translating monitoring device to the wireless communication network and background management system, the background management system is capable of making early warnings and determination according to the onsite monitored data; if the monitored data reach the preset value, early warning information comes out. In this way, comprehensive monitoring becomes available to the security of the reservoir and practical basis can be provided for future reservoir construction.

The present invention provides a double-door refrigerator with an imaging device, which comprises: a box body, which defines a storage compartment; an opening and closing assembly, which has two oppositely arranged a vertical beam on the door body, and the opening and closing assembly is configured to open or close the storage compartment; and at least one camera device, installed on the vertical beam, configured to monitor the items in the storage compartment shoot. Because the camera device is installed on the vertical beam, especially inside the vertical beam, the double-door refrigerator can make the installation position of the camera device reasonable, can protect the camera, and has excellent shooting effects, and can also improve the product quality of the double-door refrigerator. aesthetics.

The invention relates to a system and method for intelligently recognizing and shooting a multi-person scene. Wherein the shooting system includes: a camera module and a feature recognition module; the camera module can shoot the main picture image and the shooting area image, and is set to shoot a certain specific shooting area; the feature recognition module is used for capturing the obtained The image executes a character feature recognition algorithm, and compares the features obtained after being processed by the character feature recognition algorithm with a feature library to obtain a feature label. This method has high adaptability to multi-person scenes, quickly extracts character features, accurately realizes character feature recognition, and meets the needs of intelligent analysis and shooting in various scenes such as distance teaching, meeting sign-in, teaching attendance, behavior analysis, etc. The information presentation ability and resolution of the main screen.

The invention relates to the field of direct water dispensers, in particular to an AI interactive direct water dispenser based on the Internet of Things, which comprises a water accumulation structure, a direct water dispenser body, an interactive screen, a storage structure, a processing structure, a storage structure, an information acquisition structure and a sealing structure, a sealing structure used for sealing is arranged at the top end of the direct water dispenser body, a storage structure used for storing drinking water is clamped with the sealing structure, and therefore the drinking water can be conveniently stored through the storage structure, the direct water dispenser can be suitable for different environments, and meanwhile the drinking water can be stored and taken more conveniently and rapidly. The bottom end of the direct water dispenser body used for purifying a water source is provided with the water accumulation structure used for accommodating water flow, and the bottom end of the direct water dispenser body is provided with the accommodating structure which is communicated with the water accumulation structure and used for accommodating sweeps, so that thetea water sweeps can be conveniently accommodated through the accommodating structure, and water drinking is more convenient and faster.

The invention discloses a rear door plate component and a vehicle inspection-free label self-issuing machine thereof. The rear door plate component comprises a rear door plate, wherein one side of therear door plate is hinged with a side plate through a rear door hinge, the other side thereof is provided with a rear door lock, one end, close to the rear door plate, of a lower back plate is provided with a rear lower lock plate, one end, close to the rear door plate, of an upper back plate is provided with a rear upper lock plate, the rear lower lock plate and the rear upper lock plate are separately assembled with one end of a rear lock rod, so as to lock the rear door plate, the other end of the rear lock rod is hinged with one end of a rear lock connection rod, the other end of the rearlock connection rod is assembled and connected with one end of a rear lock drive rod, the other end of the rear lock drive rod is assembled in the rear door lock, one end, driving the rear lock driverod and the rear lock connection rod to assemble, is provided with a rear lock assembling groove, the rear lock assembling groove and the rear lock connection block are assembled and fixed, and the rear lock connection block is arranged on the rear lock connection rod. Through adoption of the rear door plate component and the vehicle inspection-free label self-issuing machine thereof, six-year inspection-free labels can be printed and issued automatically, so that labor intensity of workers at the window is lowered, and the working efficiency is improved.