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

34results about How to "Meet actual application needs" patented technology

Artificial synthesizing method of training samples and verification code recognizing method based on samples

The invention discloses an artificial synthesizing method of training samples and a verification code recognizing method based on the samples. The artificial synthesizing method of the training samples comprises the following steps: S1, generating a verification code lexicon; S2, generating a background photo gallery; S3, screening rectangular background blocks; S4, collecting typeface sets; S5, randomly matching words and expressions with the background blocks; and S6, taking the background blocks which have been written with the words and expressions as samples. The invention further discloses a verification code recognizing method based on the training samples. The verification code recognizing method based on the training samples comprises the following steps: S1, feature extraction; S2, sequence calibration; and S3, correction of recognition results. The training samples are acquired by a mode of synthesizing samples artificially, only recognized character types need to be set, but samples in the corresponding types can be generated, good effect can be achieved by adding a small amount of actual website samples, and a large number of labors and financial resources for sample collection are reduced while the samples are diversified.
Owner:厦门美亚商鼎信息科技有限公司

TCM complexion automatic sorting method applying superficial layer neural network

A TCM (Traditional Chinese Medicine) complexion automatic sorting method applying a superficial layer neural network belongs to the field of computer vision. The designed superficial layer neural network has five layers and adopts three different layer structures including an input layer, a characteristic extraction layer and an output layer. The input layer is composed of a convolution layer and a rectified linear unit (ReLU). The characteristic extraction layer is composed of three layers of network, wherein each of the first two layers is composed of a convolution layer and a ReLU activation function; one of the convolution layer and the ReLU is subjected to batch normalization and a pooling layer is added behind the second ReLU of the characteristic extraction layer; the third layer of the characteristic extraction layer is a whole connection layer connected with a ReLU in a followed manner. The output layer is composed of the whole connection layer and is provided with a followed softmax sorter. The method provided by the invention is distinctively advantaged in sorting precision and has invariance property in distortion such as zooming, translation, rotation and the like, is high in robustness and capable of improving sorting precision effectively. Therefore, the depth learning theory is applied to TCM face diagnose objectivity study.
Owner:BEIJING UNIV OF TECH

Multi-range capacitance measurement circuit, device and method

The invention relates to a multi-range capacitance measurement circuit, device and method. The multi-range capacitance measurement circuit comprises a square wave generation circuit, a waveform shaping circuit, a singlechip microcomputer minimum system and a display, wherein the square wave generation circuit comprises an NE555 chip, a resistor network and a reference capacitor; the resistor network comprises a resistor matrix and a first multi-way switch connected with the resistor matrix; the resistor matrix comprises multiple resistors with different resistance values; the first multi-way switch comprises multiple first switches connected with the resistors one by one and a first control chip for controlling the multiple first switches; the multiple resistors are connected in parallel. According to the multi-range capacitance measurement circuit, a singlechip microcomputer controls the first control chip to turn on different first switches, to change connected resistors of the resistor matrix and adjust the range of the multi-range capacitance measurement circuit, so as to measure to-be-measured capacitors with different capacitance values; the multi-range capacitance measurement circuit is simple, easy and convenient to operate and adjustable in measuring range, and can meet an actual application requirement.
Owner:HUBEI ENG UNIV

Giant dielectric constant strontium titanate dielectric ceramic and preparation method thereof

The invention belongs to the technical field of ceramic materials, and discloses a giant dielectric constant strontium titanate dielectric ceramic and a preparation method thereof. The chemical formula of the ceramic is Sr1-xEuxTiO3, wherein x is more than or equal to 0.01 and less than or equal to 0.015. The preparation method comprises the following steps: proportioning SrCO3, Eu2O3 and TiO2, mixing, and carrying out ball milling; drying the ball-milled raw materials, sieving, pre-sintering at 1000 DEG C for 3 hours, and mixing and ball-milling the pre-sintered powder; drying and sieving the ball-milled raw materials, adding a binder into the obtained powder, sieving, granulating, pressing into a green body, and discharging glue; and sintering the blank body after complete glue removal at 1400 DEG C in N2 atmosphere, and preserving heat for 6 hours to obtain the final sample. According to the invention, the Sr1-xEuxTiO3 dielectric ceramic is prepared through doping modification by adopting a traditional solid-phase method, the dielectric constant is improved, the lower dielectric loss is kept, and the dielectric property of the ceramic material is further improved, so that the actual application requirements are better met.
Owner:TIANJIN UNIV

Motor experience distortion index test method, electronic equipment and storage medium

The invention discloses a method for testing a motor experience distortion index. The method for testing the motor experience distortion index comprises the following steps: acquiring a motor parameter, wherein the motor parameter is a linear parameter or a nonlinear parameter; selecting a specified frequency of the distortion test and generating a single-frequency voltage signal with the specified frequency; predicting the predicted displacement of the motor vibrator under the excitation of the single-frequency voltage signal according to the linear parameter or the nonlinear parameter of themotor; adjusting the amplitude of the single-frequency voltage signal according to the target displacement value of the motor oscillator and the maximum value of the predicted displacement so as to enable the maximum value of the predicted displacement to approach the target displacement value; acquiring vibration data of the motor vibrator under the excitation of the single-frequency voltage signal after amplitude adjustment; and calculating the motor distortion from the vibration data. Compared with the prior art, the method for testing the experience distortion index of the motor can reflect the distortion condition of the motor in practical application.
Owner:AAC TECH PTE LTD

End-side-cloud cooperative data transmission method for AGV scene movement of intelligent factory

ActiveCN112153595AReduce computing and storage pressureReduce task response timeParticular environment based servicesTransmissionSmart factoryData transmission time
The invention discloses an end-side cloud cooperative data transmission method for AGV scene movement of an intelligent factory, and relates to the technical field of AGVs of intelligent factories. The method comprises the following steps: establishing a mobile terminal side cloud collaborative architecture for a delay tolerant mobile sensor network DTMSN composed of a wireless sensor, an AGV andan AGV scheduling system AGVS in an intelligent factory; fusing AGV, end-side cloud collaboration and DTMSN technologies; considering the influence of the movement of an end (wireless sensor) and an edge (AGV) on an end-edge cloud architecture network structure and a data transmission effect; improving an existing DTMSN algorithm, and using an idea that simulated quenching accepts inferior solutions according to probabilities for replacing a staged method to select intermediate nodes; combining data transmission task characteristics, node density, node speed, node type and other factors into autility function. Therefore, the data transmission quality is improved, the data transmission time delay and energy consumption are reduced, the feasibility and effectiveness of mobile terminal sidecloud cooperation are ensured, the computing resources of the AGV are fully utilized, and the task response time and the computing storage pressure of the cloud are reduced.
Owner:SHANGHAI JIAO TONG UNIV

Sum-difference beam sparse array synthesis method for broadband wide-angle beam scanning

The invention provides a sum-difference beam sparse array synthesis method for broadband wide-angle beam scanning, and relates to array antenna pattern synthesis and antenna array element position optimization. The method comprises the following steps: step 1, introducing a concept of'design frequency ', and converting a sum-difference beam array synthesis problem of broadband wide-angle beam scanning into a sum-difference beam array synthesis problem based on a single frequency point; step 2, carrying out initial layout on the array elements by adopting a power series increasing (RPS) analysis method; and step 3, performing multi-step array element position perturbation optimization at the'design frequency 'by adopting an iterative convex optimization method, and setting a plurality of constraint conditions in the optimization process, such as a sidelobe level constraint of a sum beam and a difference beam, a difference beam slope constraint in a target direction, an array element position perturbation quantity amplitude constraint, an array element minimum spacing constraint and the like. According to the method, the sum beam and the difference beam with the broadband wide-angle beam scanning performance can be generated at the same time by optimizing the array element layout, and a plurality of constraint conditions are set to meet actual application requirements.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Traditional Chinese medicine tongue image tooth mark automatic detection method based on convolutional neural network multi-scale feature fusion

The invention discloses a traditional Chinese medicine tongue image tooth mark automatic detection method based on convolutional neural network multi-scale feature fusion. The method comprises steps: using convolutional neural network VGG16 as a basic network to extract features; providing a multi-scale feature fusion module and a feature enhancement module which are respectively used for fusing different scale features of the convolutional neural network and enhancing the fused features to form four detection layers; and finally, respectively carrying out tooth mark detection on the four detection layers by adopting an SSD method, and fusing all detection results by adopting a non-maximum suppression method to obtain a final tooth mark detection result. The method can improve the feature expression capability of each detection layer, achieves the automatic and accurate detection of the tooth marks in the traditional Chinese medicine tongue image, has obvious advantages in the detection precision compared with a traditional method, and can meet the actual application demands. According to the invention, the complexity of a network model is greatly reduced. The method has obvious advantages in detection precision, and can meet actual application requirements.
Owner:BEIJING UNIV OF TECH

Method and device for determining actual position of object in fisheye camera image

The invention discloses a method and device for determining the actual position of an object in a fisheye camera image. The method comprises the following steps: determining an included angle between the positive direction of the fisheye camera image and the positive direction of a world coordinate system; determining a fitting polynomial of the installation height of the fisheye camera and the coordinate of the object in the fisheye camera image with respect to the actual distance between the coordinate of the object in the fisheye camera image and the imaging center of the fisheye camera image; and calculating the actual position of the object in the fisheye camera image under the world coordinate system. According to the invention, the actual position of the object in the image in the real world coordinate system is efficiently and accurately estimated. The mounting plane of the fisheye camera is parallel to the ground below the mounting position of the fisheye camera; meanwhile, the included angle between the positive direction of the image of the fisheye camera and the positive direction of the world coordinate system and the installation height of the fisheye camera are determined by means of the calibration plate. The efficiency is high.
Owner:上海迈外迪网络科技有限公司

An automatic classification method of complexion in traditional Chinese medicine using shallow neural network

ActiveCN107516312BRealization of complexion classificationAvoid manual feature selection processImage enhancementImage analysisSkin complexionActivation function
A method for automatically classifying complexion in traditional Chinese medicine by applying a shallow neural network belongs to the field of computer vision. The designed shallow network has 5 layers in total, using three different layer structures, which are input layer, feature extraction layer, and output layer. The input layer consists of a convolutional layer and a rectified linear unit; the feature extraction layer consists of a 3-layer network, each of the first two layers consists of a convolutional layer and a ReLU activation function, between the convolutional layer and the ReLU A batch normalization, and add a pooling layer after the second ReLU of the feature extraction layer, the third layer of the feature extraction layer is a fully connected layer, followed by a corrected linear unit ReLU; the output layer consists of a fully connected layer , followed by a softmax classifier. The invention has obvious advantages in classification accuracy, is invariant to distortions such as zooming, translation, and rotation, has strong robustness, can effectively improve classification accuracy, and applies the theory of deep learning to the objectivity research of face-to-face diagnosis in traditional Chinese medicine. .
Owner:BEIJING UNIV OF TECH

A device-edge-cloud collaborative data transmission method for smart factory AGV scene movement

ActiveCN112153595BReduce computing and storage pressureReduce task response timeParticular environment based servicesTransmissionSmart factoryMobile end
The invention discloses a device-edge-cloud collaborative data transmission method for intelligent factory AGV scene movement, and relates to the technical field of intelligent factory AGV. For the delay-tolerant mobile sensor network DTMSN composed of wireless sensors, AGV, and AGV dispatching system AGVS in smart factories, establish a mobile terminal-edge-cloud collaboration architecture; integrate AGV, terminal-edge-cloud collaboration, and DTMSN technology; consider terminal (wireless sensor) and edge ( The impact of AGV) movement on the network structure of the device-edge-cloud architecture and the effect of data transmission; improve the existing DTMSN algorithm, and use the idea of ​​analog quenching to accept inferior solutions by probability to replace the staged method for the selection of intermediate nodes; the data transmission task Factors such as its own characteristics, node density, node speed, and node type are combined into the utility function; thereby improving data transmission quality, reducing data transmission delay and energy consumption, ensuring the feasibility and effectiveness of mobile terminal edge-cloud collaboration, and making full use of AGV's computing resources reduce task response time and cloud computing and storage pressure.
Owner:SHANGHAI JIAOTONG UNIV

Gel electrolyte and preparation method thereof, and sunlight-driven self-charging supercapacitor and self-charging method thereof

The invention provides a gel electrolyte, a preparation method of the gel electrolyte, a sunlight-driven self-charging supercapacitor and a self-charging method of the sunlight-driven self-charging supercapacitor, and relates to the technical field of gel electrolytes. The gel electrolyte provided by the invention has relatively high cation migration characteristic, low thermal conductivity and low specific heat capacity. The invention further provides a sunlight-driven self-charging supercapacitor which comprises a first photo-thermal conversion film electrode, a second photo-thermal conversion film electrode and the gel electrolyte loaded between the first photo-thermal conversion film electrode and the second photo-thermal conversion film electrode, wherein the first photo-thermal conversion film electrode and the second photo-thermal conversion film electrode are oppositely arranged in parallel; the exposed part of the gel electrolyte is packaged with a packaging material; the first photo-thermal conversion film electrode and the second photo-thermal conversion film electrode are respectively bonded with a metal lead and are used for being externally connected with a load. The super capacitor provided by the invention can complete self-charging only by means of sunlight without the help of an external auxiliary charging system, so that direct and efficient conversion from solar thermal energy to electric energy and storage and utilization are realized.
Owner:LANZHOU INST OF CHEM PHYSICS CHINESE ACAD OF 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