30results about How to "Identify fit" patented technology

The invention provides a multipoint identification method and a multipoint identification system for an infrared touch screen. The method comprises the following steps of: acquiring light path data; generating an image / data structure which is in a preset proportion to the size of the infrared touch screen; establishing the attribute that pixels / data elements are correlated with light paths on the infrared touch screen; and extracting touch point information according to the attribute that the pixels / data elements are correlated with the light paths on the infrared touch screen and the light path data. The number of the light paths correlated with the image pixels / data elements is changed through the light path data; and when the number of the light paths correlated with a certain image pixel / data element is changed into zero, the image pixel / data element is marked as a target point, the target points are clustered according to the density of the marked target points, and the target points clustered into the same type form one touch point. By determining the target points according to the orders of magnitude of the image pixels or the data elements, ghost points and noise points can be effectively removed, and high identification precision can be achieved.

The invention discloses a robot barrier identification method based on a gradient histogram and a support vector machine. The method comprises two parts of a characteristic extraction stage and a target identification stage, for the characteristic extraction stage, a characteristic extraction algorithm of a power transmission line barrier of a principal component gradient histogram is proposed, the characteristic that typical barriers have different structures and space layouts is utilized, statistics characteristics of common online barriers are calculated, characteristic extraction is carried out by utilizing an HOG algorithm, characteristic points irrelevant to illumination and scale change can be acquired, interference can be effectively removed, moreover, dimension reduction operation for acquired characteristic vectors can be realized by utilizing main component analysis to acquire the principal component gradient histogram, irrelevant characteristics can be effectively reduced, operand is reduced, least characteristics are utilized to establish a characteristic set of the corresponding barriers, and excellent support is provided for next target identification; for the target identification stage, the linearity support vector machine is utilized for identification, and the excellent identification effect is acquired.

The invention relates to a method for distinguishing speakers based on a protective kernel Fisher distinguishing method. The method comprises steps as follows: (1) pretreating voice signals; (2) extracting characteristic parameters: after framing and end point detection of voice signals, extracting Mel frequency cepstrum coefficients as characteristic vectors of speakers; (3) creating a speaker distinguishing model; (4) calculating model optimal projection vector: by using optimal solution of LWFD method, calculating to obtain an optimal projection vector group; (5) distinguishing speakers: projecting original data xi to yi belonging to R<r>( r is more than or equal to 1 and less than or equal to d) according to optimal projection classification vector phi, wherein r is cut dimensionality;the optimal projection classification dimensionality of original c type data space is c-1, then solving a central value of data of each type after injection and normalizing; after projecting data tobe classified to a sub space and normalizing, calculating Euclidean distance from the normalized protecting data to the central point of each type of data in the sub space, and judging the nearest tobe a distinguishing result. The invention has high distinguishing rate, simple model construction and favorable rapidity.

Provided are a motion pedestrian identification method and a device. A video camera shooting direction is taken as a z axis, a pedestrian height direction is taken as a y axis, a plane perpendicular to the ground is taken as an xy axis plane, and then an xyz rectangular coordinate system is established. The motion pedestrian identification method comprises obtaining a candidate pedestrian contour of a time sequence form a depth map of the time sequence, obtaining at least one of x-direction distance distribution, y-direction distance distribution and z-direction distance distribution of the candidate pedestrian contour of the time sequence on the basis of the candidate pedestrian contour of the time sequence, and judging whether a candidate pedestrian contour is belonged to a pedestrian on the basis of at least one of all directions of the candidate pedestrian contour of the time sequence. The distance distribution of a pedestrian contour in all the directions is basically only related to body characteristics and walking characteristic of a person, so the pedestrian contour is unlikely to be influenced by external environment and is suitable for pedestrian identification, the pedestrian can be identified rapidly, and the real-time requirement of many applications of, for example, a vehicle driving assistance system is fitted.

The invention relates to an online identification and visualization method for power grid disturbances based on multi-resolution wavelet analysis, which belongs to the technical field of wide-area dynamic monitoring and visualization of power systems. The method comprises the following steps: carrying out offline simulation analysis on the features of typical dynamic disturbance accidents in power grid, and selecting a wavelet function Psi and the decomposition layer-number I corresponding to the obtained maximum wavelet coefficient energy as online wavelet multi-resolution analysis tools; carrying out regionalization according to the PMU (power management unit) configuration situation of a target power grid, so that each region has one and only one PMU substation for measurement, and transmitting frequency signals measured by each PMU substation to a WAMS (wide area measurement system) master station; acquiring frequency signals of each PMU substation from the WAMS master station, and carrying out I-layer online wavelet multi-resolution analysis on the frequency signals by using the determined wavelet function so as to obtain wavelet coefficients of an I(th) layer; and discriminating disturbance occurrence time, disturbance occurrence potion and disturbance type according to the wavelet coefficients; and carrying out visualized display through taking a map as the background. The method disclosed by the invention can improve the running reliability of an electrical power grid.

The invention relates to a multi-point identification method and system of an infrared touch screen. The multi-point identification method comprises the following steps of: firstly, generating a data structure with a two-dimensional matrix arrangement property corresponding to a light path in a primary identification process of the infrared touch screen; and then, extracting touch point information according to the data structure. Preferably, the method comprises the following steps of: in the primary identification process, judging whether touch points exist on the infrared touch screen according to light path data; if not, finishing the identification process. Preferably, the method comprises the following steps of: if the touch points exist, continually judging regions of the touch points according to the light path data and generating the data structure corresponding to the light path in the regions relative to the regions of the touch points; and then extracting the touch point information according to the data structure. The invention provides the novel multi-point identification method and system of the infrared touch screen, which are applicable to the identification of a plurality of the touch points and have the advantages of fast touch response speed, high identification accuracy and no ghost images or fewer images.

The invention discloses a method for recognizing a speaker based on a multivariate core logistic regression model, comprising the following steps: (A) extracting voice features of the speaker: collecting voice signals of the speaker to be recognized to pre-process, and then extracting mel cepstrum parameters; (B) constructing a speaker model: using a multivariate core logistic regression model asa speaker recognition model; (C) training the speaker recognition model: using the feature vectors extracted from the step A as input training samples, through a minimal sequence optimization algorithm, carrying out an iterative training to optimize the model parameters; (D) recognizing the speaker: extracting the feature vectors of the voice signals of the speaker to be recognized and inputting to the recognition model of the trained speaker, and giving out a posterior probability of each speaker by the multivariate core logistic regression model, wherein the highest probability value is a recognition result. The invention has high rate of recognition, simple model construction and good rapidity.

The invention relates to an identification method of channel catfish species. The identification method comprises following steps: genome DNA of channel catfish muscular tissues is extracted; a PCR primer is designed in mitochondrial DNA D-loop region; the genome DNA is subjected to PCR amplification with the presence of the PCR primer so as to obtain mitochondrial DNA D-loop sequences; the mitochondrial DNA D-loop sequences are subjected to enzyme digestion for 2 to 3h at a temperature of 37 DEG C by using restriction enzyme EcoR I; and then the mitochondrial DNA D-loop sequences are subjected to agarose gel electrophoresis after enzyme digestion, and the channel catfish species are identified according to the results of electrophoresis. Enzyme sites are searched based on sequencing of D-loop, and enzyme digestion identification is performed, so that sequencing is not necessary for identification of channel catfish species. The identification method is capable of identifying channel catfish species at a molecular level; operation is simple and easy; consumed time is short; and identification method is suitable for species identification of channel catfish in import and export trade.

The invention provides a Pleurotus nebrodensis volatile flavor component extract and an extraction method and an identification method of the extract. The extract contains fatty acids, esters and pyrimidines, wherein the weight of fatty acids accounts for 97.96% of the total weight of the extract, the weight of esters accounts for 1.46% of the total weight of the extract, and the weight of pyrimidines accounts for 0.58% of the total weight of the extract. The invention adopts supercritical CO2 extraction technology to extract Pleurotus nebrodensis volatile flavor components, studies the optimum technological parameters systematically and analyzes and identifies the volatile flavor components by adopting optimized chromatography-mass spectrometry analytical method, so that the invention provides a basis and a method for full understanding of composition of Pleurotus nebrodensis and further development and application of Pleurotus nebrodensis.

The invention relates to an online identification and visualization method for power grid disturbances based on multi-resolution wavelet analysis, which belongs to the technical field of wide-area dynamic monitoring and visualization of power systems. The method comprises the following steps: carrying out offline simulation analysis on the features of typical dynamic disturbance accidents in power grid, and selecting a wavelet function Psi and the decomposition layer-number I corresponding to the obtained maximum wavelet coefficient energy as online wavelet multi-resolution analysis tools; carrying out regionalization according to the PMU (power management unit) configuration situation of a target power grid, so that each region has one and only one PMU substation for measurement, and transmitting frequency signals measured by each PMU substation to a WAMS (wide area measurement system) master station; acquiring frequency signals of each PMU substation from the WAMS master station, and carrying out I-layer online wavelet multi-resolution analysis on the frequency signals by using the determined wavelet function so as to obtain wavelet coefficients of an I(th) layer; and discriminating disturbance occurrence time, disturbance occurrence potion and disturbance type according to the wavelet coefficients; and carrying out visualized display through taking a map as the background. The method disclosed by the invention can improve the running reliability of an electrical power grid.

The invention discloses an anti-light face recognition method based on transform domain robust watermarks under big data. The anti-light face recognition method mainly includes a watermark embedding process and a watermark extraction process, wherein face recognition is realized in the watermark extraction process. The anti-light face recognition method mainly comprises the following steps that in the watermark embedding process, firstly, global DCT is conducted on all original faces so that feature vectors can be worked out, and secondly, the watermark of each face is associated with the feature vector of the corresponding face through a hash function in cryptology; in the watermark extraction process, thirdly, the feature vector of the face to be tested is obtained, the correlation coefficient maximum value of the feature vector of the face to be tested and the original face feature vector is worked out, face recognition is completed according to a serial number corresponding to the correlation coefficient maximum value, and the corresponding embedded watermark is obtained, and fourthly, the feature vector of the face to be tested is used for watermark extraction, and the correlation coefficient of the watermark is calculated. The anti-light face recognition method has the advantages of being applicable to the big data because sampling training is not needed, and has high resistance to light, occlusion and other attacks.

The invention discloses a method for identifying the authenticity of subprostrate sophora. The authenticity of a sample of the subprostrate sophora medicinal material is judged by measuring the content of phenylbenzofuran in the sample of the subprostrate sophora medicinal material. The detection method provided by the invention is accurate and reliable, the interference of other impurities in thesample can be effectively eliminated, a satisfactory separation detection effect can be obtained, and the method is simple and convenient to operate, feasible, low in cost and high in practicability,thereby being very suitable for product quality detection.

The invention discloses a big-data DWT (Discrete Wavelet Transform) robust watermarking-based anti-illumination attack face recognition method, which mainly comprises a watermarking embedding part and a watermarking extraction part, in the watermarking extraction part, the face recognition is realized simultaneously; the face recognition method mainly comprises the following steps: watermarking embedding: I) firstly, carrying out wavelet transform to all original faces, then carrying out DFT (Discrete Fourier Transform) on approximation coefficient to obtain a characteristic vector; II) enabling the watermarking of each face to be correlated with the characteristic vector of the face through a cryptography hash function; watermarking extraction: III) acquiring the characteristic vector of a to-be-measured face, thereby obtaining the maximum value of the correlation coefficient of the characteristic vector of the to-be-measured face and the original face, and according to the value, completing the face recognition and acquiring the corresponding embedded watermarking; and IV) carrying out watermarking extraction by utilizing the characteristic vectors of the to-be-measured face, and calculating the correlation coefficient of the watermarking. The face recognition method does not need sample training, and is suitable for big data, as well as can better resist attack from illumination, shielding and the like.

The invention relates to an identification method of channel catfish species. The identification method comprises following steps: genome DNA of channel catfish muscular tissues is extracted; a PCR primer is designed in mitochondrial DNA D-loop region; the genome DNA is subjected to PCR amplification with the presence of the PCR primer so as to obtain mitochondrial DNA D-loop sequences; the mitochondrial DNA D-loop sequences are subjected to enzyme digestion for 2 to 3h at a temperature of 37 DEG C by using restriction enzyme EcoR I; and then the mitochondrial DNA D-loop sequences are subjected to agarose gel electrophoresis after enzyme digestion, and the channel catfish species are identified according to the results of electrophoresis. Enzyme sites are searched based on sequencing of D-loop, and enzyme digestion identification is performed, so that sequencing is not necessary for identification of channel catfish species. The identification method is capable of identifying channel catfish species at a molecular level; operation is simple and easy; consumed time is short; and identification method is suitable for species identification of channel catfish in import and export trade.

The invention discloses a chlorate sensitivity-based nitrogen-efficient barley screening method, which comprises the following steps: revealing the relationship between chlorate and each index of nitrogen-efficient barley through research, determining a chlorate screening system and evaluation indexes suitable for nitrogen-efficient barley, and utilizing seedling height and overground part dry weight as screening indexes of nitrogen-efficient barley. The barley variety of which the seedling height and the overground part are sensitive to chlorate at the same time is defined as nitrogen-efficient barley, the nitrogen absorption efficiency can be rapidly distinguished in the early stage of barley, the identification time is saved, and the identification efficiency is improved; the method can be used for large-scale screening and identification, is obvious in phenotype and higher in accuracy, avoids inaccuracy of a single index, is good in repeatability, and is suitable for large-scale material identification.

The invention discloses a rotation speed steady-state value fitting method and device for a ball-wire mill, and the method comprises the steps: controlling a set rotation speed value of the ball-wiremill to rise or fall to a target rotation speed value in a process that the ball-wire mill works according to the set rotation speed value, and enabling the actual rotation speed value of the ball-wire mill to present a step response process; acquiring an actual rotation speed value of the ball-wire mill in a stable stage; fitting the actual rotation speed value of the ball-wire mill in the stablestage into a straight line in a coordinate system which takes time as horizontal coordinates and takes the rotation speed as vertical coordinates; determining an longitudinal axis intercept value ofthe straight line as an actual rotation speed steady-state value of the ball-wire mill in the steady stage. The accuracy of transfer function parameter identification can be improved, so that a transfer function which conforms to the actual rotation speed of the ball-wire mill better is obtained, and the ball-wire mill is controlled to work, and thus, the situation that bars and wires on the millare accumulated is effectively prevented, and the bar and wire production efficiency is improved.

The invention discloses a method for rapidly identifying Sitophilus oryzae, Sitophilus zeamais and Sitophilus granarius. The method is characterized in that on the basis of acquiring COI (Cytochrome C Oxidase Subunit I) gene sequences of the Sitophilus oryzae, the Sitophilus zeamais and the Sitophilus granarius, a non-homologous region among the COI gene sequences is found; three pairs of specific primers are designed according to the sequences in the non-homologous region; the Sitophilus oryzae, the Sitophilus zeamais and the Sitophilus granaries are distinguished according to the absence and presence of finally-obtained electrophoretic bands or the differences among the sizes of the electrophoretic bands after amplification of these specific primers. By adopting the method, the conventional classification and identification steps are simplified, and the problems of high sequencing cost and the like are solved; a provided PCR (Polymerase Chain Reaction) primer method can increase the identification efficiency and accuracy of the Sitophilus oryzae, the Sitophilus zeamais and the Sitophilus granarius, is easy, convenient and rapid to operate, is suitable for identifying the Sitophilus oryzae, the Sitophilus zeamais and the Sitophilus granarius, and has an important theoretical meaning and an application value.

The invention provides a method for non-destructive and rapid identification of all silty and waxy corn germplasm by using an optical illuminometer, and belongs to the technical field of corn germplasm identification methods. The method comprises the steps of: placing a seed endosperm upward on an optical measuring point of the illuminometer, irradiating the seed endosperm with a miniature fixed white light source from top leaning against the pericarp, recording light intensity reading on the illuminometer, judging whether the seed is an all silty seed according to the light intensity of the light source passing through the seed endosperm, if the light intensity of the light source passing through the seed endosperm is less than 5 lux, indicating that the seed is an all silty waxy corn, and if the light intensity of the light source passing through the seed endosperm is greater than 100 lux, indicating that the seed is waxy corn. According to the method, the all silty and waxy corn germplasm is identified according to the optical properties of different structural starches, the cultivation and verification of all silty and waxy corn selfing lines can be assisted by the method, as well as the breeding of all silty and waxy corn new species, so that the application prospect is very good.

A moving pedestrian recognition method and device are provided, in which the camera shooting direction is the z-axis, the pedestrian height direction is the y-axis, and the plane perpendicular to the ground is the xy-axis plane, and an xyz rectangular coordinate system is established. The moving pedestrian recognition method can The method includes: obtaining the candidate pedestrian outline of the time series from the depth map of the time series; based on the candidate pedestrian outline of the time series, obtaining at least one of the x-direction distance distribution, y-direction distance distribution, and z-direction distance distribution of the time-series candidate pedestrian outline. ; and based on at least one of the various directions of the time series candidate pedestrian outline, determine whether the candidate pedestrian outline is a pedestrian. The distance distribution in each direction of the pedestrian outline is basically only related to the individual's own physical characteristics and walking characteristics. Therefore, it is not easily affected by the external environment. It is suitable for pedestrian recognition and can quickly identify pedestrians. It is suitable for many applications such as vehicle driving assistance systems. real-time requirements.

The invention discloses a method for constructing an inverter dead-zone voltage compensation model and its application, including: S1, fitting the compensation voltage of each phase and its phase current, and the power circuit of the inverter respectively by means of e-exponential function fitting Among them, the nonlinear relationship between the hardware parameters corresponding to the phases and the maximum compensation voltage of the inverter is obtained, and the compensation voltage equation of each phase is obtained; S2, based on the linear transformation relationship between the dead zone compensation voltage and the compensation voltage of each phase, combined with Compensation voltage equation of each phase, get the nonlinear relationship between the compensation voltage in the dead zone and the current of each phase, the hardware parameters corresponding to each phase in the power loop of the inverter, and the maximum compensation voltage of the inverter, which is recorded as the dead zone of the inverter Area voltage compensation model. The present invention does not need to use the electrical parameters of the inverter, and also avoids the compensation error of the traditional compensation algorithm and the cumbersome theoretical model calculation, can obtain better compensation effect, and can real-time and self-adaptive to the inverter Linear effects are precisely compensated.

The invention discloses a robot obstacle recognition method based on gradient histogram and support vector machine. The method mainly includes two parts: first, in the feature extraction stage, the feature of the transmission line obstacle of the principal component gradient histogram is proposed extraction algorithm. Utilizing the characteristics of typical obstacles with different structural compositions and spatial layouts, by calculating the statistical characteristics of common online obstacles and using the feature extraction of the HOG algorithm, feature points that are independent of illumination and scale changes can be obtained, which can effectively remove interference. At the same time, the principal component analysis is used to further reduce the dimensionality of the obtained eigenvectors to obtain the principal component gradient histogram, which can effectively reduce irrelevant features, reduce the amount of calculation, and use the least amount of features to establish a feature set corresponding to obstacles. One-step target identification provides good support. Second, in the stage of target recognition, a linear support vector machine (SVM) is used for recognition, and a very good recognition effect is obtained.

A multipoint recognition method and system of an infrared touch screen. Optical path data is first collected; an image / data structure in a predetermined proportion to the size of the infrared touch screen is generated; then, an attribute of association between a pixel / data element and an optical path on the infrared touch screen is established; finally, touch point information is extracted according to the attribute of the association between the pixel / data element and the optical path on the infrared touch screen and the optical path data. Through the optical path data, the number of optical paths associated with image pixels / data elements is changed. When the number of optical paths associated with a certain image pixel / data element turns to be zero, the image pixel / data element is marked as a target point, and target points are then clustered according to the density of the marked target points. The target points in the same cluster form one touch point. The target points are determined based on the order of magnitude of the image pixels or the data elements, thereby affectively eliminating ghost points and noise points, and achieving high recognition precision.

The invention discloses a rotation speed low-pass filter method and device for a bar-wire mill, and the rotation speed low-pass filter method comprises the steps: controlling a set rotation speed value of the bar-wire mill to rise or fall to a target rotation speed value in a process that the bar-wire mill works according to the set rotation peed value, so as to present a step response process ofthe actual rotation speed value of the bar-wire mill; collecting an actual rotation speed feedback value of the bar-wire mill in a stable stage; adopting a digital low-pass filter for filtering the actual rotation speed feedback value of the bar-wire mill in the stable stage; and determining the actual rotation speed value obtained after filtering processing as the actual rotation speed value of the bar-wire mill in the stable stage. The accuracy of transfer function parameter identification can be improved, so that a transfer function which conforms to the actual rotation speed of the bar-wire mill is obtained, and thus, the bar-wire mill is controlled to work, the situation that bars and wires on the mill are accumulated is effectively prevented, and the bar and wire production efficiency is improved.

The invention provides an action recognition method and its neural network generation method, device, and electronic equipment, which relate to the technical field of image recognition. The neural network generation method for action recognition includes: extracting a target image to obtain optical flow features; The optical flow feature obtains convolution kernel bias information; according to the convolution kernel bias information, a deformed convolution neural network is generated based on the initial convolution neural network, which solves the problem of image recognition neural network in the prior art for action recognition The less effective technical issues.