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

36results about How to "Reduce the number of multiplications" patented technology

Convolution acceleration method, convolution calculation processing method, devices, electronic apparatus and storage medium

ActiveCN108229645ATake full advantage of reconfigurableTake full advantage of parallel computingProgram controlNeural architecturesParallel computingData store
The embodiments of the invention disclose a convolution acceleration method, a convolution calculation processing method, a convolution acceleration device, a convolution calculation processing device, an electronic apparatus and a storage medium. The acceleration method includes the following steps that: the to-be-processed data of a preset size in a to-be-processed task are sequentially read from the off-chip memory of an accelerator through an FIFO (first-input-first-output) port, and are stored in the input cache regions of a first cache region in the on-chip memory of the accelerator; to-be-processed input window data are sequentially read from the input cache regions in response to a condition that the to-be-processed data are stored into the input cache regions, and convolution calculation is performed on the to-be-processed input window data through convolution kernel data, so that output window data are obtained and are stored in a third cache region; and output window data inthe third cache region are sequentially stored to the off-chip memory through the FIFO port. With the methods, devices, electronic apparatus and storage medium provided by the embodiments of the invention adopted, a condition that the on-chip memory and bandwidth of the accelerator are insufficient and a condition that processor resources are limited can be avoided, and the efficiency of convolution calculation processing is improved. The methods and devices can be applied to hardware platforms such as an FPGA and an ASIC.
Owner:BEIJING SENSETIME TECH DEV CO LTD

Pedestrian real-time detection method based on neural network

The invention discloses a pedestrian real-time detection method based on a neural network, and the method comprises the steps: 1) collecting data, and constructing an original training data set; 2) calibrating pedestrian positions corresponding to the acquired images for the images in the original training data set, and constructing real position output of the training data set; 3) constructing aneural network model; 4) in order to increase the network training data volume and enhance the applicability of the method, performing data enhancement on the original input image by using a data enhancement method; 5) setting training parameters for the designed neural network model for training, and storing the trained neural network model parameters; and 6) acquiring image data of pedestrian detection to be performed by using an image acquisition device, and then inputting the image of the pedestrian to be detected into the stored neural network model to obtain the pedestrian position of the image of pedestrian detection to be performed. The method can reduce a large amount of detection time on the premise of meeting the detection accuracy of pedestrian detection.
Owner:SOUTH CHINA UNIV OF TECH

Spherical multi-probe antenna test data processing method

The invention discloses a spherical multi-probe antenna test data processing method in order to solve the technical problem that the existing method is of poor practicability. According to the technical scheme, a field established in a space by a to-be-tested antenna is expanded into the sum of spherical wave functions, a weighted function in the expansion contains the information of a far-field directional pattern, the weighted function can be worked out according to near-field sampling data and the far-field directional pattern is thus worked out, multi-probe spherical near-field scanning is adopted in near-field sampling, the to-be-tested antenna azimuthally rotates around the antenna center, probes carry out sampling at a distance of a few wavelengths from the antenna aperture, and therefore, the far-field coverage angle provided by spherical near-field measurement is within the range of + / -180 degrees for azimuth angle and + / -90 degrees for pitch angle, and all radiation information of the to-be-tested antenna can be obtained. Thus, the method is applicable to measurement of all types of beam antennas. The number of times of program looping is reduced, the speed of program running is improved, and the time for antenna test data processing is shortened.
Owner:NORTHWESTERN POLYTECHNICAL UNIV

Ultra-wideband beam former based on differential operator and achieving method thereof

The invention discloses an ultra-wideband beam former based on a differential operator and an achieving method thereof. The ultra-wideband beam former comprises a beam summation module and a plurality of array element channels. Each array element channel comprises a weight coefficient and delay quantity storage module, a narrow-band phase compensation module and an accurate delay compensation module. The achieving method of the ultra-wideband beam former comprises the steps of storage of weight coefficient and delay quantity, narrow-band phase compensation, accurate delay compensation and beam summation. A differential operator delay compensator is adopted to perform delay compensation, multiplication times is reduced, and calculation amount and multiplier resource expenses are reduced. The delay compensator is an all-pass filter, accurate delay compensation in full bandwidth can be achieved, the beam former based on the differential operator delay compensator can achieve wide-angle continuous beam scanning by adopting a programmable logic device field programmable gate array (FPGA), and has high phase shift / delay accuracy.
Owner:XIDIAN UNIV

Method for predicting self-interaction effect of protein

ActiveCN107609352ASelf-interacting effectiveThe prediction of self-interactions worksSpecial data processing applicationsData setFourier descriptor
The invention discloses a method for predicting self-interaction effect of protein. The method comprises the steps of selection and establishment of a data set, generation of a PSSM matrix, extractionof a feature value by a Fourier descriptor, construction of a training set and a testing set and construction of a classifier model. According to the method, the feature value of a sample set is extracted by the Fourier descriptor, the number of times of multiplication required for discrete Fourier transform of a computing data set of a computer is greatly reduced, and the computing quantity is reduced. A model can be constructed by a random projection method, the prediction precision is greatly improved, and a good prediction effect can be achieved. The method is low in computing cost and small in power consumption; the self-interaction effect of the protein can be predicted effectively; and the prediction effect can reach 93% or above.
Owner:XINJIANG TECHN INST OF PHYSICS & CHEM CHINESE ACAD OF SCI

Data processing method and device

The invention provides a data processing method and device.. The data processing method includes: determining the non-zero element in the vector to be calculated; Obtaining a data processing instruction, the data processing instruction carrying a first base address and a column number, the first base address being a first address stored in a memory by a sparse matrix multiplied by a vector to be calculated, and the column number being a position of a target column in the sparse matrix; Decoding the data processing instruction, and performing multiplication on the non-zero element in the vectorto be calculated and the target element in the target column according to the data processing instruction; Wherein the value of the target element of each column in the sparse matrix and the positionof the target element in the column to which the target element belongs are stored in the memory, and the target element comprises a non-zero element existing in the column and a zero element meetinga preset condition; And constructing a scheme of the multiplication result of the vector to be calculated and the sparse matrix according to the multiplication result of each non-zero element in thevector to be calculated, and reducing the number of times of multiplying the zero element in the sparse matrix by another vector to improve the utilization rate of operation resources and memory resources.
Owner:LOONGSON TECH CORP

Self-adaptive Cartesian grid generation method for three-dimensional streaming problem of any shape

ActiveCN113505443AEfficient and robust generationReal-time display of feature structureGeometric CADDesign optimisation/simulationComputational scienceAlgorithm
The invention discloses a self-adaptive Cartesian grid generation method for a three-dimensional streaming problem of any shape, and the method comprises the steps: generating an isotropic self-adaptive Cartesian grid suitable for an immersed boundary method based on geometric information in the three-dimensional streaming problem, carrying out the calculation of a flow field, and encrypting an area containing key flow features according to the calculation result of the flow field. In order to solve the problem of computational fluid mechanics numerical simulation with complex three-dimensional streaming, a surface set composed of triangles is adopted as input, a grid intersection judgment method based on a separation axis theory and a grid inside and outside judgment method based on an improved ray algorithm are adopted for grid classification, a grid subdivision method based on a unit is adopted for encrypting and coarsening grid units, and a self-adaptive Cartesian grid meeting the requirements of an immersed boundary method and flow field calculation resolution can be efficiently and robustly generated; and a region containing a feature structure is selectively encrypted according to flow field parameters obtained subsequently, and the flow field feature structure in the current flow field area is displayed in real time.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

Inverse discrete cosine transform lookup fast algorithm with variable complexity

The invention discloses an inverse discrete cosine transform (IDCT) lookup fast algorithm with variable complexity, which mainly aims at reducing the redundant operation in the original acceleration IDCT algorithm and accelerating the computation speed of an IDCT module. The algorithm provided by the invention comprises the steps of decomposing an inverse quantized DCT (discrete cosine transform) coefficient matrix F into polynomial matrixes to add; performing inverse discrete cosine transform for each decomposed matrix; and finally, adding the inverse discrete cosine transformed matrixes to obtain an original matrix f. The algorithm disclosed by the invention reduces the redundant operation in the original algorithm and effectively accelerates the computation speed of the inverse discrete cosine transform.
Owner:LEADCORE TECH

RUMSWF based low-complexity reduced rank balancing method in MIMO system

The invention provides a RUMSWF based low-complexity reduced rank balancing method in an MIMO system, which is improved from an MSWF based self-adapting reduced rank linear balancing method and is a reduced rank self-adapting MIMO linear balancing method which realizes a multilevel Weiner filter based on a rectangular block matrix and by a related subtraction structure. By improving the block matrix of the unitary multilevel Weiner filter, a rectangular matrix block of the square block matrix is selected as the block matrix, and the number of dimension of a received signal is reduced step by step in the forward recursive decomposition of the unitary multilevel Weiner filter, thereby reducing the iteration complexity of the self-adapting balancing and simultaneously increasing the convergence rate. The theoretical analysis and simulation result show that the low-complexity quick reduced rank self-adapting balancing method has the advantages of low complexity and quick convergence rate. In a V-BLAST system provided with 4 transmitting antennas and 8 receiving antennas and adopting the BPSK modulation, by only utilizing one half of complexity of the multilevel Weiner filter based balancing method, the error code performance which is only 0.78 dB lower than that of the multilevel Weiner filter based balancing method can be achieved.
Owner:XI AN JIAOTONG UNIV

Data processing method, device and equipment and storage medium

The invention discloses a data processing method, device and equipment, and a storage medium; the method is applied to a first electronic device, and the first electronic device and at least one second electronic device jointly participate in secure multi-party computing, and the method comprises the steps of obtaining to-be-processed service data; generating a first random number matrix and a second random number matrix, and generating a third random number matrix according to a preset encoding mode, the first random number matrix and the second random number matrix, the number of elements ofthe random number matrix encoded by the preset encoding mode being smaller than the number of elements of the random number matrix before encoding; and encrypting the service data in a secure multi-party computing process according to the first random number matrix, the second random number matrix and the third random number matrix. According to the invention, the generation efficiency of the matrix multiplication triad can be improved, and the encryption efficiency of service data in secure multi-party calculation is further improved.
Owner:WEBANK (CHINA)

Fast Fourier analysis algorithm

The invention discloses a fast Fourier analysis algorithm, and relates to the technical field. The algorithm comprises the steps of obtaining a sequential sorting sequence x (n) of input data of N sampling points; calculating inverted sequence y(n) of input data by y(n)=x (G(m)+2Lmax-1*l); taking 8-point, 10-point or 12-point data from the inverted sequence y (n) for the first time for Fourier transform; amplifying the number of analysis points by two times each time to perform Fourier transform; and carrying out Fourier transform with mutually independent N points. According to the algorithm,the inverted sequence y (n) of input sequences by y(n)=x (G(m)+2Lmax-1*l) is calculated according to the difference between the total number of points participating in the sequence and the number ofpoints participating in the Fourier transform for the first time, and then the first-time 8-point, 10-point or 12-point data Fourier transform is rapidly completed through a set of formulas; the number of analysis points is amplified by two times each time to perform Fourier transform; according to the invention, the time complexity during calculation is reduced, the calculation rate is increased, and electric signal spectrum analysis is facilitated.
Owner:安徽慕京信息技术有限公司

Interpolation data processing method, device and equipment and computer readable storage medium

The embodiment of the invention provides an interpolation data processing method, device and equipment and a computer readable storage medium. The method comprises the following steps: acquiring information of two adjacent key frames of a spline curve, wherein the information of the two adjacent key frames comprises first time and second time of two-dimensional coordinate points respectively corresponding to the two adjacent key frames on a time axis; determining interpolation data calculation parameters according to the information of the two adjacent key frames; and when the third time is between the first time and the second time, determining target interpolation data corresponding to the third time according to the interpolation data calculation parameters. According to the method, interpolation data calculation is carried out through the spline curve, so that the calculation amount of the interpolation data is reduced, and the processing efficiency of the interpolation data is improved.
Owner:XIAMEN YAJI SOFTWARE

Portable multi-frequency electrical impedance tomography front-end data acquisition and processing method

The invention discloses a portable multi-frequency electrical impedance tomography front-end data acquisition and processing method which comprises the following steps: converting a sinusoidal signal generated by a DAC into a current signal by using a voltage-controlled constant current source, and performing current excitation under the action of a first analog switch; obtaining a reference input signal through a sampling resistor, gating different groups of electrodes through a second analog switch, and performing program control gain by using an inverted T-shaped program control gain amplifier to obtain a measurement input signal; synchronously acquiring the reference input signal and the measurement input signal, performing multiple single-point Fourier demodulation according to the periodicity and symmetry of the digital signal and the symmetry of the trigonometric function, and obtaining the phase information of the signal. The existing demodulation method is improved, the required multiplication frequency is greatly reduced, and the demodulation speed is improved.
Owner:GUILIN UNIV OF ELECTRONIC TECH

Method, system and equipment for accelerating image processing and storage medium

The invention provides a method, a system and equipment for accelerating image processing and a storage medium. The method comprises the following steps: dividing obtained picture data into data blocks, and executing the following steps during convolution operation of each data block: carrying out picture data conversion and convolution kernel data conversion on the data blocks to respectively obtain a picture data conversion item and a convolution kernel data conversion item; and accelerating the convolution operation of the image data conversion item and the convolution kernel data conversion item in the data block through a GPU acceleration unit, and performing inverse transformation on the result of the convolution operation to obtain a calculation result. The invention has the advantages that the image convolution operation speed can be greatly increased, and the performance requirement during reasoning is met. In addition, the invention also provides an optimal data partitioning structure for executing the method disclosed by the invention, so that further implementation of the method disclosed by the invention is facilitated.
Owner:INSPUR SUZHOU INTELLIGENT TECH CO LTD

Method for computing low-entropy universal high-order masks based on composite field

The invention provides a method for computing low-entropy universal high-order masks based on a composite field. In the prior composite field mask plans, each mask order is a first-order mask, so thatthe high-order mask is not realized; and in other words, the extensibility of the algorithm is not high. The complexity of the algorithm is high; and relatively larger resources are occupied to realize the algorithm. The algorithm is not flexible; and the plans only designed according to the AES (Advanced Encryption Standard) cannot be applied to other algorithms. The method comprises the steps of: (1) inputting a masked value and a mask value; (2) mapping an inputting value on a finite field; (3) converting an inversion operation on the finite field GF(22n) into addition, multiplication andinversion operations on a finite field that is as shown in the description; and (4) evaluating a mask protective value of d that is as shown in the description. The invention is used for the method for computing low-entropy universal high-order masks based on the composite field.
Owner:HARBIN UNIV OF SCI & TECH

A fast ray tracing method and system

The invention discloses a rapid ray tracing method and system, and the method comprises the steps: two-stage refusal detection and one-stage line-plane intersection calculation. The method specifically comprises the steps that In order to search a surface element intersected with a ray, the refusal detection at the first stage is used for locking a preliminary range, wherein the basic idea comprises the steps: taking the maximum side length Lm of a triangular surface element K as the radius, taking any top point Q0 of the triangular surface element as the sphere center, forming a sphere Q through rotation, and determining that the ray cannot be intersected with the triangular surface element if the ray is located outside the sphere Q formed by the triangular surface element; the refusal detection at the second stage is used for locking a more precise range, wherein the basic idea comprises the steps: determining that the surface element intersected with the ray is definitely intersected with two planes which pass through the ray and are not parallel; finally the line-plane interaction calculation is employed, wherein the basic idea comprises the steps: converting the line-plane intersection calculation of a three-dimensional space into the point-plane calculation of a two-dimensional space through parallel projection, and jumping there-dimensional inversion matrix solving. The method achieves the acceleration of ray tracing under the condition that the precision is not affected.
Owner:HUAZHONG UNIV OF SCI & TECH

An Adaptive Cartesian Mesh Generation Method for 3D Flow Around Arbitrary Shapes

ActiveCN113505443BEfficient and robust generationReal-time display of feature structureGeometric CADDesign optimisation/simulationAlgorithmImage resolution
The invention discloses a method for generating an adaptive Cartesian grid for a three-dimensional surrounding flow problem with arbitrary shapes. Based on the geometric information existing in the three-dimensional surrounding flow problem, an isotropic adaptive Cartesian grid suitable for the immersion boundary method is generated, and the Perform flow field calculations, and encrypt areas containing key flow characteristics based on the flow field calculation results. Aiming at the computational fluid dynamics numerical simulation problem with complex three-dimensional flow around, the present invention adopts the surface set composed of triangles as input, and adopts the grid intersection determination method based on the separation axis theory and the grid internal and external determination method based on the improved ray algorithm to carry out grid meshing. Classification, using the element-based meshing method to refine and coarsen the grid cells, can efficiently and robustly generate an adaptive Cartesian grid that meets the requirements of the immersion boundary method and the resolution of the flow field calculation, and obtains according to the subsequent The flow field parameters of the device selectively encrypt the area containing the characteristic structure, and display the flow field characteristic structure in the current flow field area in real time.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

Self-adaptive filtering method and related device

An input signal is filtered for creating an output signal using an adaptive filter. An error signal is derived from the output signal. The adaptive filter has coefficient whose value can be modified. A value of a coefficient is modified using a derived updating amount. The updating amount is obtained from the product of a value of the input signal, a value of the polarity of the error signal, and a step gain. The step gain has the form 2 with K being an integer and being dependent on a magnitude of the value of the error signal and on a step gain parameter. The updating amount is dependent on both the magnitude and the polarity of the error signal, therefore allowing a precise update of the coefficient. The specific form of the step gain allows a fast derivation of the product.
Owner:KONINKLIJKE PHILIPS ELECTRONICS NV

Dynamic electromagnetic spectrum posture method and system based on tensor and neural network

The invention discloses a tensor and neural network-based dynamic electromagnetic spectrum posture method and system. The method comprises the steps of obtaining spectrum data of an area surrounded by current unmanned aerial vehicle flight; performing analog-to-digital conversion on the frequency spectrum information, then reading the frequency spectrum information after performing fast Fourier transform, and checking whether the frequency spectrum data displayed for one time is full or not; then an original matrix is constructed, and the constructed original matrix is visualized; performing data completion on the visualized original matrix to obtain an optimized final completion matrix; and inputting the optimized final completion matrix result value into a BP neural network for fuzzification processing to obtain a final completion matrix value, drawing a completed spectrum posture two-dimensional graph by using the final completion matrix value, drawing a three-dimensional posture spatial resolution graph, and displaying a spectrum posture signal of the dynamic signal according to time and frequency changes. According to the invention, by monitoring the radio signal and the electromagnetic spectrum posture, the purposes of eliminating electromagnetic interference and improving the spectrum utilization rate are achieved.
Owner:NORTHWESTERN POLYTECHNICAL UNIV

A Real-time Pedestrian Detection Method Based on Neural Network

The invention discloses a real-time pedestrian detection method based on a neural network, comprising the steps of: 1) collecting data and constructing an original training data set; 2) for the images in the original training data set, marking the pedestrian position corresponding to the collected image, and constructing the training data 3) Construct a neural network model; 4) In order to increase the amount of network training data and the applicability of the enhancement method, it is necessary to use a data enhancement method to perform data enhancement on the original input image; 5) The designed neural network model Set the training parameters to train, and save the training neural network model parameters; 6) use the image acquisition device to obtain the image data to be detected pedestrians, and then input the images of the pedestrians to be detected into the preserved neural network model, to obtain the pedestrians to be detected Pedestrian locations of detected images. The invention can reduce a lot of detection time under the premise of satisfying the detection accuracy for pedestrian detection.
Owner:SOUTH CHINA UNIV OF TECH

Audio frequency equalizing method of decoder based on sub-band filter frame

The invention discloses an audio frequency equalizing method of a decoder based on sub-band filter frame, which realizes relatively low complexity and relatively small consumed resource. The technical scheme is as follows: the method comprises the following steps: (1) obtaining time sampling points by an inverse quantization step in a decoder based on the sub-band filter frame; and limiting the bandwidth of the signal constituted by the time sampling points in the frequency range corresponding to each decoder sub-band; and (2) performing plus adjustment on the time sampling point in the frequency band which is obtained by the inverse quantization step and is adjusted by an audio frequency equalizer before the sub-band of the decoder composites filtering.
Owner:SPREADTRUM COMM (SHANGHAI) CO LTD

Self-adapting medicine sequence image values interpolation method based on interested region

The invention discloses an adaptive inter-sectional interpolation method of medical sequence images based on regions of interest. The method makes full use of the characteristics of tissue correlation and voxel correlation, and first judges whether the point to be interpolated belongs to the region of interest, and then Choose an appropriate interpolation method. The invention not only considers the change of gray level, but also takes into account the change of anatomical structure outline, thereby better solving the deficiency of the traditional interpolation method, and making the newly interpolated picture closer to the actual situation. Experiments prove that the visual effect and image quality of the present invention have been greatly improved, and the interpolated pictures have clear outlines, less noise, no double outlines, improved interpolation quality, and more importantly, greatly reduced calculation time. Applying the present invention to a three-dimensional reconstruction and display system makes the border of the drawn image clear, reduces the terrace effect to a minimum, and is more in line with actual needs, thereby improving the accuracy of doctor's judgment and work efficiency in a large range.
Owner:SUZHOU SOUKE INFORMATION TECH

A Dynamically Reconfigurable Convolutional Neural Network Accelerator Architecture for the Internet of Things

The present invention is a dynamic reconfigurable convolutional neural network accelerator architecture for the Internet of Things field, including a cache architecture, etc. The cache architecture is used to store data from external storage or data generated during the calculation process, and organize and arrange them , is transmitted to the processing unit array in a data structure for calculation; the processing unit array is used to receive data from the cache architecture, and is stored in the cache architecture after convolution operation processing; the calculation module is used to receive data from the processing unit array, select Perform three operations of pooling, normalization, or activation functions, and store the output data in the cache architecture; the controller is used to send commands to the cache architecture, processing unit array, and computing module, and is designed with an external interface for communicating with the outside system to communicate. The present invention improves the performance of the convolutional neural network accelerator and reduces power consumption by designing a processing unit array with high parallelism and high utilization rate and a cache architecture that can increase the data multiplexing rate.
Owner:XI AN JIAOTONG UNIV

Fast timing synchronizing method of full-duplex communication system

The invention relates to a fast timing synchronizing method of a full-duplex communication system, which belongs to the technical field of the timing synchronization of wireless communication. The method is characterized by comprising the steps: adopting two identical PN code sequences as pilot signals through a transmitting-terminal data frame, receiving the signal through a receiving terminal, delaying the signal for a PN code period, aligning a second PN code with a first PN code of the delayed received signal, multiplying the first PN code and the second PN code by taking the first PN code and the second PN code as corresponding sampling points, counting the quantity of the multiplied values which are greater than a set threshold value in the length of the PN code period, carrying out the multiplication operation of the corresponding sampling points on the delayed receiving signal and the local PN code sequence if the statistic quantity is greater than a critical value of the quantity, then counting the quantity of the multiplied values which are greater than the set threshold value in the length of one PN code period, and if the statistic quantity still exceeds the critical value of the set quantity, determining the head part of a delaying unit as a timing synchronizing point. By adopting the method, the timing synchronization is realized by ingeniously utilizing the self-correlation characteristic of the PN code and the random characteristic of the noise, the multiplication times can be reduced, the operand is reduced, the accumulation operation is not needed by adopting the threshold value comparison way, and the timing synchronizing point can be rapidly and accurately found.
Owner:SHANDONG UNIV

Data processing method and equipment

The invention provides a data processing method and device.. The data processing method includes: determining the non-zero element in the vector to be calculated; Obtaining a data processing instruction, the data processing instruction carrying a first base address and a column number, the first base address being a first address stored in a memory by a sparse matrix multiplied by a vector to be calculated, and the column number being a position of a target column in the sparse matrix; Decoding the data processing instruction, and performing multiplication on the non-zero element in the vectorto be calculated and the target element in the target column according to the data processing instruction; Wherein the value of the target element of each column in the sparse matrix and the positionof the target element in the column to which the target element belongs are stored in the memory, and the target element comprises a non-zero element existing in the column and a zero element meetinga preset condition; And constructing a scheme of the multiplication result of the vector to be calculated and the sparse matrix according to the multiplication result of each non-zero element in thevector to be calculated, and reducing the number of times of multiplying the zero element in the sparse matrix by another vector to improve the utilization rate of operation resources and memory resources.
Owner:LOONGSON TECH CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products