Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

295results about How to "Cannot restrict disclosure" patented technology

Human body tracking algorithm based on attention mechanism and double-flow multi-domain convolutional neural network

ActiveCN112651998ASolve the problem of poor tracking effectCannot restrict disclosureImage enhancementImage analysisHuman bodyVideo sequence
The invention discloses a human body tracking algorithm based on an attention mechanism and a double-flow multi-domain convolutional neural network. The double-flow multi-domain convolutional neural network tracks a human body target in a double-flow input and multi-flow output mode. The apparent characteristics of the human body target are extracted through the original video sequence, the motion characteristics of the human body target are extracted through the frame difference sequence, and the proportion of the motion amplitude of the human body target in the image is taken as an adaptive fusion parameter to realize the adaptive fusion of the apparent characteristics and the motion characteristics of the target; the algorithm effectively solves the problem that the tracking effect is poor when the shape, the scale, the texture and the like of the human body target to be tracked are changed, and feature optimization and improvement are performed on the tracking target in real time by adopting a mode of offline training shared layer network parameters and online learning independent layer network parameters. It can be known through experimental results that the network can achieve accurate human body target tracking on the premise that real-time tracking is guaranteed.
Owner:SHENYANG AEROSPACE UNIVERSITY

Lens surface defect detection method and system based on machine vision, product and terminal

The invention discloses a lens surface defect detection method and system based on machine vision, a product and a terminal, and relates to the technical field of digital image processing. Two area array light sources are placed at a certain angle, the light sources irradiate the surface of the lens, residual light of the light sources is used for defect detection, so that the defects such as pocks, fingerprints and scratches are highlighted, and the defects are located at the light and shade junction of the light sources; the lens is conveyed on the transmission structure, and when the lens reaches the position below the camera, the camera starts to acquire images and transmits image information to the industrial personal computer in real time, so that rapid online detection of lens defects is realized; and algorithm processing is performed on the acquired image, the characteristics of the defects are highlighted, the defects are classified by utilizing the characteristics of the defects, and finally a detection result is obtained. The mode that the two area array light sources emit light to the two sides is adopted, residual light of the light sources is used for detecting the defects of the lens, the showing force on the defects on the surface of the lens can be enhanced, and a high-quality defect image is obtained.
Owner:菲特(天津)检测技术有限公司

Unmanned aerial vehicle virtual air combat entertainment system

The invention discloses an unmanned aerial vehicle virtual air combat entertainment system. According to the entertainment system, a targeting gun and a camera are carried on an electric holder of anunmanned aerial vehicle; a player can control the electric holder to move relative to the unmanned aerial vehicle through a wireless remote controller to perform aiming adjustment of the targeting gun; an object shot by the targeting gun is the virtual unmanned aerial vehicle projected onto a projection curtain through a laser holographic projector; during game playing, the camera sends an acquired image to an aiming display, the player determines an aiming state by observing image information displayed in the aiming display and controls the targeting gun to perform shooting through the wireless remote controller; and after bullets are shot onto the projection curtain, a pressure sensor in the projection curtain detects corresponding pressure values and sends the pressure values to a maincontroller, and the main controller controls a loudspeaker box to play corresponding music according to the received pressure values. The unmanned aerial vehicle virtual air combat entertainment system has the advantages of being simple in structure, reasonable in design, convenient to operate, good in entertainment and the like.
Owner:淮安鱼鹰航空科技有限公司 +1

Airplane surface ice accretion evolution state identification method and system

The invention belongs to the technical field of airplane surface ice accretion detection, and discloses an airplane surface ice accretion evolution state identification method and system. The airplane surface ice accretion evolution state identification method comprises the following steps: collecting historical airplane surface historical ice accretion meteorological factor data and ice accretion evolution microscopic information, constructing Hankle form state space expressions of airplane key part ice accretion in the past and future for airplane surface ice accretion, obtaining an observable matrix and judging generalized observability of the observable matrix; carrying out oblique projection on the state space expressions, eliminating external non-meteorological variables and future ice accretion microscopic information noise influences, and obtaining a state estimation sequence; in combination with the state estimation sequence and the observable matrix, obtaining a system matrix of a spatial model through least square, obtaining an evolution state identification model of airplane surface ice accretion, and predicting airplane ice accretion evolution state information. The invention has the advantages of being fast in response, being able to change along with the environment, easy to implement and the like, and has good application prospects.
Owner:CIVIL AVIATION UNIV OF CHINA

Ice accumulation detection method and system based on temperature sequence similarity measurement, terminal and application

The invention discloses an ice accretion detection method and system based on temperature sequence similarity measurement, a terminal and application, and relates to the technical field of wing ice accretion detection. The method comprises the steps: collecting a wing surface temperature change sequence with the ice accumulation phenomenon, and obtaining a wing temperature change rate sequence; calculating the distance D1 between the wing surface temperature change sequence and the standard temperature change sequence when the wing surface is iced under the current meteorological condition and the distance D2 between the wing temperature change rate sequence and the standard temperature change rate sequence when the wing surface is iced through a dynamic time warping method; processing the distance D1 and the distance D2 according to an icing model to obtain an icing detection result, and giving an alarm in time when icing occurs. The ice accretion detection method provided by the invention has the advantages of fast response, strong anti-interference capability, easy implementation and the like; in combination with a non-contact infrared temperature sensor, non-contact ice accumulation detection can be achieved, the aerodynamic configuration of the wing is prevented from being affected, and the application prospect is good.
Owner:CIVIL AVIATION UNIV OF CHINA

Expansion shell type yielding anchor cable and construction method thereof

The invention belongs to the technical field of anchoring and supporting of underground engineering such as tunnels, and discloses an expansion shell type yielding anchor cable and a construction method thereof. A variable-section yielding anchor cable is fixedly mounted at the upper end of an expansion shell anchoring part, and an anchorage device is fixedly mounted at the lower end of the expansion shell anchoring part; a hollow metal pipe is arranged in the middle of an anchor cable body, and the hollow metal pipe is wrapped with a high-strength conical pipe; and the outer side of the anchor cable body is wrapped with a variable-section threaded sleeve through a binder. The expansion shell type yielding anchor cable is mainly applied to anchoring and supporting of water-rich tunnels, especially high-stress large-deformation water-rich underground engineering, and the variable-section yielding anchor cable, the expansion shell anchoring part, the anchorage device and a base plate form an anchoring technology integrating rapid anchoring, yielding supporting and grouting protection; and surrounding rock is rapidly supported through the biting force formed by the expanded expansion shell anchoring part and the hole wall surrounding rock, the deformation capacity of the anchor cable is improved through the yielding function of the variable-section yielding anchor cable, and post-grouting protection is achieved through the hollow metal pipe arranged in the variable-section anchor cable.
Owner:ERCHU CO LTD OF CHINA RAILWAY TUNNEL GRP +3

Self-weight fitness auxiliary coach system, method and terminal based on human body posture recognition

PendingCN113762133AConvenient fitness guidanceConvenient professional fitness guidanceCharacter and pattern recognitionNeural architecturesHuman bodyInformation processing
The invention discloses a self-weight fitness auxiliary coach system, method and terminal based on human body posture recognition, and relates to the technical field of fitness training information processing. The method comprises the steps: employing a camera for capturing and recording actions of the body builder; during fitness training, detecting the actions of a body builder from the exercise frame sequence; identifying the motion category of the body builder by extracting the motion amplitude and joint angle features of the body building motion; and carrying out labeling and data processing on the exercise standard of the exercise category of the body builder. The body builder is assisted in training so as to get rid of the state that household self-weight body building only depends on experience, convenient and professional body building guidance is brought to the body builder, unnecessary exercise injuries are avoided, and the body building efficiency is improved. According to the invention, a training method based on visual observation in traditional fitness is taken into account, and an analysis method based on human body motion characteristics is realized. The invention of the self-weight fitness auxiliary coach system has important scientific research value and potential huge commercial value.
Owner:HARBIN INST OF TECH AT WEIHAI

Audio playing system, external extension device, and audio playing method and apparatus

An audio playing system, an external extension device, and an audio playing method and apparatus, relating to the field of audio processing. The audio playing method comprises: obtaining a user soundsignal acquired by a microphone device (401); receiving, by means of an integrated physical interface connected to a television device, an accompaniment audio signal transmitted by the television device (402); mixing the user sound signal with the accompaniment audio signal (403); and transmitting a mixed signal to the power amplifier circuit of the television device by means of the integrated physical interface, the television device being used for playing, by means of a built-in loudspeaker, the mixed signal received by the power amplifier circuit (404). The present invention solves the problems of the high cost of providing a karaoke processing chip in a television device, and of being inconvenient for upgrading. A karaoke function is integrated into an external extension device, a mixed signal is generated by using the external extension device and transmitted to the television device, and the built-in loudspeaker of the television device plays the mixed signal, so that the manufacturing costs are reduced, and it is convenient to upgrade the karaoke function.
Owner:BEIJING XIAOMI MOBILE SOFTWARE CO LTD

Intelligent household moxibustion robot

The invention discloses an intelligent household moxibustion robot. The intelligent household moxibustion robot is mainly composed of a base, an intelligent mechanical arm, a moxibustion operation executing mechanism, a moxibustion auxiliary box feeding mechanism, a moxibustion auxiliary box, a controller and a smoke exhaust system, wherein the base is used for overall supporting, the moxibustionoperation executing mechanism is connected with the intelligent mechanical arm and used for clamping, rotating and other operations of a moxibustion stick; the moxibustion auxiliary box feeding mechanism is connected with the intelligent mechanical arm and the moxibustion auxiliary box; under the driving of the moxibustion auxiliary box feeding mechanism, the moxibustion auxiliary box can achieveposition adjustment and is matched with the moxibustion operation executing mechanism to complete moxibustion, ash removal and other operations; the controller collects signals of the intelligent mechanical arm, a distance sensor and a visual sensor and controls the intelligent mechanical arm to be matched with the moxibustion auxiliary box so as to keep the consistency with a traditional handheldmoxibustion technology; the characteristics of traditional Chinese medicine moxibustion stick moxibustion are reproduced; and the problem that an existing robot scheme cannot retain the characteristics of traditional Chinese medicine moxibustion stick moxibustion is well solved.
Owner:王宝成

Equipment for molten steel continuous casting and continuous casting method

InactiveCN111136256AQuality improvementSolve the problem of high labor intensityMelt-holding vesselsSlagSteel ball
The invention discloses equipment for molten steel continuous casting, and a continuous casting method. The equipment for molten steel continuous casting replaces a long nozzle of a previous steel ladle through a non-oxidation immersion cover, a molten steel injection flow from the steel ladle to a tundish can be protected against secondary oxidation due to contact with air, the non-oxidation immersion cover is inserted into the tundish, argon is blown into the non-oxidation immersion cover, a slag-free oxygen-free area full of the argon is formed in the non-oxidation immersion cover, a condition is provided for preventing molten steel secondary oxidation, the quality of steel balls can be improved, meanwhile, the non-oxidation immersion cover is arranged, the structure of the long nozzleis omitted, and the problems that the long nozzle needs to be continuously cleaned, the pouring takt is affected, and labor intensity of a worker is affected can be effectively solved. The continuouscasting method uses the equipment, the equipment for molten steel continuous casting has the advantages of being simple in structure, reasonable in design, convenient to use, low in secondary oxidation degree of molten steel pouring, high in made steel ball quality and the like, cleaning of the long nozzle is not needed, and labor intensity of the worker is low.
Owner:BENGANG STEEL PLATES

Surface-modified porous polyether-ether-ketone artificial bone and preparation method thereof

The invention discloses a preparation method of a surface-modified porous polyether-ether-ketone artificial bone, which comprises the following steps: establishing discrete points of a three-dimensional topological structure by adopting a three-period minimal curved surface method, carrying out reverse modeling by utilizing the discrete points to generate an artificial bone model, and preparing the porous polyether-ether-ketone artificial bone by adopting an additive manufacturing method according to the artificial bone model; converting carbonyl in a polyether-ether-ketone molecular structure on the surface of the artificial bone into hydroxyl by utilizing a chemical reaction, and respectively reacting high-activity isocyanate serving as an intermediate with the hydroxyl on the surface of polyether-ether-ketone and an amino-containing compound to obtain the surface-modified polyether-ether-ketone artificial bone. The surface-modified porous polyether-ether-ketone artificial bone prepared by the technical scheme of the invention not only has a gradient continuous change pore structure, but also can promote division and growth of cells by introducing enough bioactive substances on the surface, and improves the adhesion ability of cells on the surface of the bone.
Owner:GUANGXI UNIV FOR NATITIES

Post-healing rehabilitation auxiliary detection system and method based on three-dimensional human skeleton model and medium

PendingCN114049683AAid in recoveryReal-time human gesture and behavior recognitionHealth-index calculationCharacter and pattern recognitionRehabilitationPinhole camera model
The invention discloses a post-healing rehabilitation auxiliary detection system and method based on a three-dimensional human skeleton model and a medium, and relates to the technical field of computer vision. The method comprises the following steps: utilizing and deploying a classic OpenPose neural network in Raspberry Pi, and obtaining two-dimensional coordinate points of a plurality of joint points in real time; converting two-dimensional coordinates into three-dimensional coordinates by using a pinhole camera model, and establishing a three-dimensional human skeleton model; calculating skeleton spacing and skeleton angles in the three-dimensional human skeleton model, and inputting the skeleton spacing and the skeleton angles into a trained random forest model; recognizing real-time postures of a user in real time by using a socket; and displaying the real-time postures in an APP, and detecting and judging health conditions of the user. According to the invention, eight-element data including seven skeleton angles and one posture label is used as a training set to train a random forest classification model, and the recognition accuracy rate reaches 88.8%.
Owner:HARBIN INST OF TECH AT WEIHAI

Three-shaft six-direction adjustable tool for automobile power assembly

The invention discloses a three-shaft six-direction adjustable tool for an automobile power assembly. Two support frames are used as a base body of the tool, the two support frames are connected through length adjusting devices so as to realize distance adjustment between the two support frames and realize two-way adjustment in the X-axis direction, and first support adjusting devices, second support adjusting devices and third support adjusting devices are slidably mounted on first longitudinal beams and second longitudinal beams of the two support frames correspondingly for being directly connected with the power assembly in a supporting mode. The three kinds of support adjusting devices can slide along the respective longitudinal beam, two-way adjustment in the Y-axis direction is achieved, the first support adjusting device and the second support adjusting device can both achieve height adjustment, and therefore two-way adjustment in the Z-axis direction is achieved. Since the three kinds of support adjusting devices for supporting the power assembly are adjustable in position, and therefore the three-shaft six-direction adjustable tool can be adjusted according to the practical type of the power assembly so as to be applicable to power assemblies with different types, and has the advantages of being wide in application range and the like.
Owner:LIAONING JUNFENG MACHINERY MFG

Chinese character OCR method and system based on neural network, medium and application

The invention discloses a Chinese character OCR method and system based on a neural network, a medium and application, and relates to the technical field of image recognition. The method includes: performing text detection on the input image; performing text mask information extraction on the detected text area image, positioning the position coordinates of each character in the text area, and intercepting each character mask from the text mask according to the character position coordinates; performing feature extraction on the character mask, and mapping the feature to a multi-dimensional vector space to obtain a character vector; enabling the character vectors to be grouped into a text vector matrix according to a character order, and utilizing the text vector matrix for learning context dependence information to complete text recognition. According to the method, the problems of Chinese text detection of the input image, image text area interception and text area character recognition are solved. The Chinese character recognition accuracy is improved, the recognition accuracy of characters with similar shapes is improved, and the direction of the character is insensitive.
Owner:深圳市网联安瑞网络科技有限公司

Intelligent multi-channel digital signal data acquisition and storage system and method

The invention discloses an intelligent multi-channel digital signal data acquisition and storage system and method. The data acquisition and storage system has a sleep function and a wake function; when a detector is stuck in a pipeline, after an operation state recognition module performs recognition, the operation state recognition module automatically adjusts the data acquisition and storage system to be a sleep state; when the detector operates normally again, the operation state recognition module automatically awakens the data acquisition and storage system to perform normal data acquisition and storage after performing recognition; a time trigger mode and a mileage trigger mode are set for the data storage system, so that a user can select the modes according to needs, and therefore, the applicability of the data acquisition and storage system can be greatly improved, after data are acquired, a data analysis unit needs to perform data analysis so as to judge whether the data aredefective data; and only when the data are defective data, the data are stored normally, and therefore, a storage quantity can be decreased. The system has the advantages of reasonable design, convenient use, small power consumption, high applicability and the like.
Owner:SHENYANG ACAD OF INSTR SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products