360results about How to "Recognizable" patented technology

The invention provides a voiceprint identification method based on a Gauss mixing model and a system thereof. The method comprises the following steps: voice signal acquisition; voice signal pretreatment; voice signal characteristic parameter extraction: employing a Mel Frequency Cepstrum Coefficient (MFCC), wherein an order number of the MFCC usually is 12-16; model training: employing an EM algorithm to train a Gauss mixing model (GMM) for a voice signal characteristic parameter of a speaker, wherein a k-means algorithm is selected as a parameter initialization method of the model; voiceprint identification: comparing a collected voice signal characteristic parameter to be identified with an established speaker voice model, carrying out determination according to a maximum posterior probability method, and if a corresponding speaker model enables a speaker voice characteristic vector X to be identified to has maximum posterior probability, identifying the speaker. According to the method, the Gauss mixing model based on probability statistics is employed, characteristic distribution of the speaker in characteristic space can be reflected well, a probability density function is common, a parameter in the model is easy to estimate and train, and the method has good identification performance and anti-noise capability.

The invention discloses a motion recognition method based on a three-dimensional convolution depth neural network and depth video. In the invention, depth video is used as the object of study, a three-dimensional convolution depth neural network is constructed to automatically learn temporal and spatial characteristics of human body behaviors, and a Softmax classifier is used for the classification and recognition of human body behaviors. The proposed method by the invention can effectively extract the potential characteristics of human body behaviors, and can not only obtain good recognition results on an MSR-Action3D data set, but also obtain good recognition results on a UTKinect-Action3D data set.

The invention discloses an automatic fruit identification method of an apple picking robot on the basis of a support vector machine. An apple orchard colored image under natural illumination is collected, and a vector median filter is adopted to pretreat the apple colored image; after pretreatment, the image is cut with the method of combining region growing with color characteristics; the color characteristics and the geometry characteristics of the cut apple colored image are respectively extracted; apple fruits are indentified with the mode identification method of the support vector machine; and finally, the fruits are accurately positioned. The identification method of the support vector machine of the invention which integrates color characteristics and shape characteristics has higher apple fruit identification precision rate than the precision rate when only color characteristics or shape characteristics are adopted and has better identification effect; the algorithm is easy to realize, the operation time is short, and the identification performance is superior to a commonly used neural network method, so that the identification method shows advantages in small sample learning.

The invention relates to a high-performance jet ink. The jet ink comprises the following components by weight percent: 20-80% of solvent, 1-45% of cosolvent, 4-20% of adhesive, 0.5-20% of colorant, 0-5% of adhesion force promoter, 0.01-2% of wax powder, 0-2% of brightening agent and 0-5% of other additives. As adhesion force promoter is added in the jet ink, the adhesion performance on base materials such as glass, metal, polypropylene or polyethylene is increased. In addition, wax is added to increase wear resistance and scratch resistance, and brightening agent is added to increase the lustrousness, brilliance and definition of the ink. The product has high adhesion force, good wear resistance, high lustrousness and definition and good compatibility, thus the defects of the existing ink such as low lustrousness, poor wear resistance and low definition can be greatly improved and the jet ink has good market prospect.

The invention relates to a multi-scale normal feature point cloud registering method. The multi-scale normal feature point cloud registering method is characterized by including the steps that two-visual-angle point clouds, including the target point clouds and the source point clouds, collected by a point cloud obtaining device are read in; the curvature of radius neighborhoods of three scales of points is calculated, and key points are extracted from the target point clouds and the source point clouds according to a target function; the normal vector angular deviation and the curvature of the key points in the radius neighborhoods of the different scales are calculated and serve as feature components, feature descriptors of the key points are formed, and a target point cloud key point feature vector set and a source point cloud key point feature vector set are accordingly obtained; according to the similarity level of the feature descriptors of the key points, the corresponding relations between the target point cloud key points and the source point cloud key points are preliminarily determined; the wrong corresponding relations are eliminated, and the accurate corresponding relations are obtained; the obtained accurate corresponding relations are simplified with the clustering method, and the evenly-distributed corresponding relations are obtained; singular value decomposition is carried out on the final corresponding relations to obtain a rigid body transformation matrix.

The invention relates to a facial expression recognition method based on a generative antagonistic network, which relates to computer vision technology. Firstly, a facial expression generation networkbased on generative antagonistic network is designed and pre-trained. The network is composed of a generator and two discriminators, which can generate a face map with random identities of specifiedfacial expressions. Then a facial expression recognition network is designed, which receives the facial expression map from the training set and the random facial expression map generated by the facial expression generation network, and uses an intra-class loss to reduce the facial expression feature differences between the real samples and the generated samples. At the same time, a real sample-oriented gradient updating method is used to promote the feature learning of the generated samples. Finally, according to the trained facial expression recognition network model, the final facial expression recognition results are obtained from the final flexible maximum classification layer of the model.

The invention discloses a digital modulation signal identification method under pulse noise. The method comprises the following steps that the fractional low-order fuzzy function of digital modulation signals is calculated; the tangent plane of zero Doppler frequency shift of the fractional low-order fuzzy function is intercepted, and the tangent plane is converted into a two-dimensional image and fills below the edge of the image so as to become a color image with color; graying, binarization, image segmentation and image size normalization are performed on the color image and translation and scale normalization are performed on the image; and the Zernike moment of the image is extracted to act as the characteristic vector of identification, and the digital modulation signals are identified by using a probability neural network classifier. The digital modulation signal identification method has great identification performance on the digital modulation signals 2ASK, 4ASK, 2FSK, 4FSK and BPSK under standard distribution pulse noise.

The invention provides a telemetry data driven unmanned plane flight state identification method, and provides a Chebyshev-random forest algorithm (C-RF) unmanned plane state identification method. The telemetry data of an unmanned plane undergoes characteristic extraction and dimension reduction through adopting a Chebyshev fitting technology, and the adaptive classification of the flight state is realized by using a random forest algorithm. The method combines the advantages of simple calculation and optimum fitting approaching of a Chebyshev fitting coefficient with the advantages of fast training speed, high classification accuracy and strong noise resistance, can cover various samples of the unmanned plane and avoid an over-fitting problem, and realizes effective identification of the flight state of the unmanned plane. Real unmanned plane flight telemetry data is adopted to carry out identification, so the total identification accuracy is higher than 90%, and a small amount of kinds of the samples is accurately identified, so the effectiveness and the practicality of the method are proved.

The invention relates to a Chinese-English bilingual speech recognition method based on phoneme confusion. The method comprises the following steps: unifying a Chinese-English phoneme set by adopting a twice phoneme clustering method, obtaining a Chinese-English mixed acoustic model by retraining, correcting corresponding bilingual pronunciation dictionaries, and realizing Chinese-English bilingual recognition by a decoder based on the preceding steps. The Chinese recognition rate of a Chinese-English bilingual speech recognition system based on TCM phoneme confusion can be comparable to that of an independent Chinese speech recognition system; on the premise that English data with Chinese accent is unavailable and standard Chinese-English training data are used only, compared with the relatively independent English speech recognition system, recognition of English fragments with the Chinese accent is obviously improved; meanwhile, the Chinese-English bilingual speech recognition system based on the TCM phoneme confusion also has a better recognition performance than the existing common bilingual recognition system which performs phoneme clustering by virtue of a logarithm likelihood criterion, and has very high practicability.

The invention discloses a laser radar and camera fusion device and a calibration method. The laser radar and camera fusion device comprises a vertical lifting platform and a calibration plate bracket,a working table is arranged on a working guide rail on the vertical lifting table; a camera and a laser radar are fixedly mounted on the working table; a calibration plate is fixedly mounted on the calibration plate bracket; the laser radar scans the calibration plate to obtain a horizontal straight line, finds left and right end points of the straight line on the boundary of the calibration plate, then moves for multiple times at the same interval to obtain multiple groups of end points on the left and right side lines, and performs fitting to approximately obtain a linear equation of the left and right boundaries so as to obtain the end points of the calibration plate; pixel points at the positions where the upper end points of the camera pictures are located correspond to the end points, so that a pose relation expression of the camera and the radar can be obtained; by applying the laser radar and camera fusion calibration method, the characteristics under a radar point cloud sparse scene can be obtained, the joint expressions of internal and external parameters can be solved, and the internal parameters of the camera do not need to be measured in advance.

The invention discloses a coumarin-thiacalix [4] arene fluorescent reagent as well as a preparation method and an application thereof, which belong to the chemical field of organic synthesis and analysis. The invention prepares an excellent coumarin-thiacalix [4] arene derivative which can be used for respectively detecting performances of fluorine ions and zinc ions. The preparation method of the coumarin-thiacalix [4] arene fluorescent reagent comprises the following steps of: respectively connecting two coumarin fluorescent groups at the lower 1,3-site of the calixarene and connecting two methoxyl ethyl at the 2,4-site through step by step reaction by using coumarin-thiacalix [4] arene and 7-hydroxycoumarin as starting materials to prepare the coumarin-thiacalix [4] arene fluorescent reagent. Under a special medium condition, the compound can be used as a dual-functional fluorescent or colorimetric reagent for detecting micro fluorine ions and zinc ions with high sensitivity and high selectivity by utilizing fluorescence spectrum and ultraviolet absorption spectrum.

The invention relates to a preparation method of a fly ash superbead magnetic composite material surface imprinting adsorbent. The invention belongs to the technical field of environmental material preparation. Through a microemulsion method, cross-linked chitosan is used for coating nano-grade gamma-Fe2O3 and micrometer-grade ball-shaped fly ash superbeads, such that the fly ash superbead magnetic composite material is obtained; the fly ash superbead magnetic composite material is adopted as a substrate, molecularly imprinted polymer modification is carried out on the surface of the substrate through a suspension polymerization method; template molecules bisphenol A are eluted, such that the fly ash superbead magnetic composite material surface imprinting adsorbent is obtained. The ball-shaped imprinting adsorbent has substantial thermal and magnetic stabilities. 1H-NMR shows that hydrogen bonding among the molecules is an identification framework of the imprinting adsorbent. As a result of a static absorbing experiment, the fly ash superbead magnetic composite material surface imprinting adsorbent provided by the invention has relatively high adsorption capacity, rapid adsorption dynamic property and substantial BPA molecule identification performance.

The invention discloses a multiple electric appliance type-supportable high-precision NILM (Non-intrusive Load Monitoring) realization method, so as to recognize multiple types of load electric appliances in the NILM system and achieve the purposes of expanding the compatibility breadth of the system and improving the precision. The method comprises steps: data are acquired; the system prepares training data according to a selected load electric appliance; the system uses a specific architecture based on a deep neural network to complete training and modeling on the type of the load electric appliance; and the system completes the load electric appliance recognition process through the well-trained architecture. A complete training data set is selected, and good performance is achieved inthe modeling of electric appliances; in consideration of the compatibility of recognition of most electric appliances, the recognition performance of a two-state electric appliance is enhanced; and the method can be promoted to be used in a system in no need of high-end hardware to acquire data.

The invention provides an SAR target recognition method based on range profile time-frequency image identification dictionary learning. An SAR range profile time-frequency image is used as a recognition feature, so that influence of defocusing caused by target movement or low image quality caused by a factor such as a signal noise ratio on a target recognition effect is avoided; and identification dictionary learning is combined with dictionary learning and classifier training, so that feature information of radar target range profile time-frequency data can be effectively extracted. The invention is conducive to reducing a number of atoms in a dictionary and decreasing operation complexity in a sparse encoding process, and is also conducive to improving precision of sparse encoding, so as to improve recognition accuracy of a radar target. Further, azimuth angle estimation does not need to be performed on an SAR image target in a whole recognition process. Therefore, a recognition complex degree is reduced, and dependency of recognition accuracy on target azimuth angle estimation is also reduced. In addition, the invention has excellent recognition performance, and facilitates improving robust performance of radar target recognition.

The invention provides a control method of an electronic cigarette. The method comprises the following steps: S100, obtaining a resistance value of an atomization assembly (20), determining whether the atomization assembly (20) has a magnetic material (21), if so, displaying a first display interface, and otherwise, displaying a second display interface; S200, according to the obtained resistance value of the atomization assembly (20) and a preset corresponding table, outputting corresponding power; and S300, determining whether output power / voltages are adjusted, if a determining result is yes, recording the output power after adjustment, and re-generating a corresponding table, and if the determining result is no, ending operation. Further provided is an electronic cigarette using the control method. According to the invention, user operation can be simplified, and the electronic cigarette is better customized.

The invention discloses a preparation method of a temperature sensitive adsorbent of halloysite magnetic composite material surface blotting, belonging to the technical field of environment functional material preparation. The preparation method comprises the following steps: preparing a ferroferric oxide / halloysite nanotube (Fe3O4 / HNTs) magnetic composite material through a simple and effective solvent thermosynthesis method; and then, performing ethylene modification on the magnetic composite material; and preparing the temperature sensitive adsorbent of halloysite nanotube magnetic composite material surface blotting through a radical polymerization process, and utilizing the adsorbent in selective identification and separation of 2,4,5-trichlorophenol in aqueous solution. The prepared temperature sensitive blotting adsorbent has excellent thermal and magnetic stability, sensitive magnetic and thermal induction effect, high adsorption capacity, excellent reversible adsorption / release function with temperature and obvious TCP molecular identification performance.

The invention relates to a various illumination face identification method based on small sample emulating and sparse expression and belongs to the technical field of image processing. The method comprises the steps of: creating an image illumination collection and combining into a virtual combined sample with an entropy image method, acquiring the sparse expression of the individual to be identified on a complete face base from the virtual combined sample by using a sparse expression method, rebuilding an original detected image after substituting the sample rebuilding factor in a classification band-pass function so as to acquire a residual error between each type of rebuilt sample and the original detected image, and guiding the residual error into a category identifying formula to acquire an indentifying result. The invention solves the problem that the traditional sparse frame requires huge learning sample and achieves an excellent identifying property. The method of the invention obtains a better robustness under shielding or various illumination conditions and acquires better properties of sampling requirement, identifying ratio and man-machine interaction efficiency.

The invention relates to a partial discharge signal pattern recognition method and system. The method is mainly technically characterized in that a large data sample of a partial discharge signal is acquired; a deep convolutional neural network model is constructed; the deep convolutional neural network model is trained based on the large data sample of the partial discharge signal; and the pattern of the partial discharge signal to be recognized is determined based on the trained deep convolutional neural network model. According to the invention, the design is reasonable; the deficiencies ofpartial discharge detection of a GIS in the substation field in the prior art are overcome; pattern recognition can be carried out on partial discharge of large number and complicated sources of theGIS in the substation field; the accuracy of pattern recognition is higher; and the method and system have better recognition performance, and are more suitable for engineering applications under a big data platform.

Provided are: an original plate for a lithographic printing plate, which can provide a visible image having excellent visibility when exposed to light, and which can be developed on a machine; and a method for printing the original plate. An original plate for a lithographic printing plate, which comprises at least a white base plate and an image recording layer, said original plate being characterized in that an image-recording-layer-formed side of the white base plate has a reflection density of 0.25 or less and the image recording layer comprises an infrared ray absorbent, an onium salt-type polymerization initiator, a polymerizable compound and a color-developable compound and can be removed with at least a printing ink and / or a dampening solution. It is preferred that the white base plate comprises an aluminum support and a white layer arranged on the aluminum support and comprising at least a fluorescent whitening agent and / or a white pigment or comprises an aluminum support of which the surface has been roughened so that the degree of sharpness (a45) can become 30% or less and has also been anodically oxidized.

The invention discloses a defect identification method of partial discharge signal, which comprises the following steps: (1) acquiring training samples of partial discharge signals representing several partial discharge insulation defect types, extracting phase-resolved pulse sequence data based on the training samples, and normalizing the phase-resolved pulse sequence data; (2) the sparse self-encoder model being trained by the normalized phase-resolved pulse sequence data, and the dimension-reduced training eigenvectors being obtained based on the trained sparse self-encoder model; 3) training the extreme learning machine model by using the training eigenvector; (4) the PD signal to be identified being input into the trained sparse self-encoder model to extract the test feature vector, and the test feature vector being input into the trained limit learning machine model to obtain the defect types represented by the PD signal to be identified. The invention also discloses a defect identification system of a partial discharge signal.

The invention discloses a di-pyrene modified perylene bisimide derivative fluorescent probe and a synthesis method and application thereof, and belongs to the technical field of protein detection. The preparation of the fluorescent probe comprises the following steps that in a protective atmosphere, 1-methyl-2-pyrrolidinone solution of a pyrene sulfonyl derivative is added into a 1-methyl-2-pyrrolidinone solution of perylene; then, anhydrous zinc acetate is added to obtain a mixed solution; next, the mixed solution is stirred; heating reflux is performed for 10 to 15h at 150 to 200 DEG C; after the reaction liquid obtained after the reflux is cooled to the room temperature, filtering is performed to obtain filter liquid; the filter liquid is subjected to acid washing; precipitates occur; suction filtration is performed; filter cake is obtained; the filter cake is washed and dried to obtain the di-pyrene modified perylene bisimide derivative fluorescent probe. The fluorescent probe provided by the invention has the advantages that the chemical stability is high; the response speed is high; the sensitivity is high; the discrimination performance is good; the sample consumption quantity is small; the data collection is simple, and the like. The instant and on-line detection of the protein can be realized. The fluorescent probe provided by the invention can be used for respectively realizing the distinguished detection of various proteins.

The invention discloses a foot-to-ground acting force-based gait feature extraction method and a foot-to-ground acting force-based gait identification system. The system comprises a gait data acquisition module and a gait data processing and identification module, wherein the gait data acquisition module acquires the foot-to-ground acting force during walking in a hidden and non-intrusive way; and the gait data processing and identification module extracts gait features and performs gait identification from the foot-to-ground acting force. The gait feature extraction method comprises the following steps of: denoising the foot-to-ground acting force by a wavelet transform hard threshold method; detecting a key point and calculating gait time-domain features by a first-order difference algorithm; and representing the gait frequency-domain features by adopting a wavelet decomposition coefficient after performing waveform alignment on the foot-to-ground acting force. When the system is used for identifying, an optimal gait feature set is selected by a fuzzy C-means algorithm; and the classification or the identification can be performed on the gait based on a support vector machine. The gait information acquired by the method and the system is natural, true and not easy to simulate; the extracted gait features are full and accurate; and the requirements on gait analysis, classification and long-distance gait identification can be met better.

The present invention relates to a method for preparing a porous magnetic imprinting absorption agent through emulsion polymerization, and belongs to the technical field of environment function material preparation. The method comprises: preparing magnetic halloysite, carrying out vinyl modification on the halloysite, adopting a dispersion solution of the modified magnetic halloysite and 2,2'-azobis(2-methylpropionamidine) dihydrochloride as a water phase, adopting a mixture of a template molecule cyhalothrin, a function monomer methacrylic acid, a cross-linking agent ethylene glycol dimethacrylate, a pore-forming agent and an organic solvent chloroform as an oil phase, mixing the water phase and the oil phase, carrying out an ultrasonic treatment to prepare stable Pickering emulsion, carrying out polymerization to prepare porous magnetic imprinting, adopting a methanol and acetic acid mixing solution to wash to remove the template molecule, and carrying out vacuum drying on the obtained porous magnetic imprinting at a temperature of 50 DEG C, wherein the obtained product is used for selective recognition and separation of cyhalothrin in an aqueous solution. The prepared porous magnetic imprinting absorption agent has characteristics of significant thermal stability, significant magnetic stability, high absorption capacity, and significant cyhalothrin molecule recognition property.

The invention discloses a face identification method based on Gabor wavelet and SB2DLPP. The face identification method based on the Gabor wavelet and SB2DLPP mainly includes pre-treatment, feature extraction, feature dimension reduction and classification identification, and to be specific, the face identification method includes that (1) pre-treating all the face images in a known face database, wherein the pre-treatment includes scale normalization and histogram equalization; (2) using the Gabor wavelet to extract features of the pre-treated face images; (3) leading in class information, and using a supervised bidirectional two-dimensional local preserving projection (SB2DLPP) algorithm to reduce the dimensions of the high-dimensional image features extracted through the step (2) to extract feature matrices mapped to a low-dimensional sub-space; (4) using a nearest neighbor classifier to perform classification identification. The face identification method based on the Gabor wavelet and SB2DLPP uses the Gabor wavelet and improved SB2DLPP algorithm to identify images, the problems that a traditional face identification method is easy to be influenced by light, expression and the like external factors are overcame, and the face identification rate is effectively improved.

The invention discloses a high-selectivity magnetic core-hollow porous molecular imprinting polymer satellite assembly applied to determining the residual quantity of trace macrolide antibiotics in animal-derived foods and a preparation method of the assembly. With mesoporous silicon as a sacrificial support template, methacrylic acid as a functional monomer and spiromycin as a template molecule,hollow porous molecular imprinting polymers (HPMIPs) are prepared by means of a thermal initiation polymerization method; ferroferric oxide magnetic nanoparticles (Fe3O4NPs) are prepared by means of asolvothermal method. By utilizing the self-polymerization properties of dopamine, polydopamine-coated ferroferric oxide magnetic nanoparticles (Fe3O4@polyDA) are prepared, the HPMIPs are grafted on the surfaces of the polydopamine-coated ferroferric oxide magnetic nanoparticles (Fe3O4@polyDA), synthesized magnetic hollow porous molecular imprinting polymers (Fe3O4@polyDA-HPMIPs) are obtained andserve as a magnetic dispersion solid phase extraction adsorbent, macrolide drugs are selectively enriched and separated, and in combination with HPLC-MS / MS detection, the effect is ideal. The preparedmaterial has the advantages of high affinity, high adsorption capacity, a good magnetic response effect and the like and has broad application prospects in the field of food detection and the like.

The invention relates to a preparation method of molecularly imprinted polymer of penicillin antibiotics and intermediate of penicillin antibiotics. The molecularly imprinted polymer is prepared from 6-aminopenicilanic acid (6-APA) as template molecule, methacrylic acid (MAA) as functional monomer and ethylene glycol dimethacrylate as cross-linking agent by dissolving, polymerization, elution, and the like. The molecularly imprinted polymer is used as filler of solid phase extraction (SPE) column, and the prepared solid phase extraction column has good recognition on penicillin antibiotics, such as penicillin-G potassium (PEN), ampicillin (AMP), amoxicillin (AMO) and intermediate 6-APA of penicillin antibiotics, and can be used for sample pretreatment in detection of penicillin antibiotics residue in animal-derived food.

The invention discloses a metal plate strip product label information identification method based on computer vision. the position of a product label area is obtained through segmentation of a lightweight network; the coordinate information of the product label is obtained through an image processing means; correction to enabline see-through transformer is realized, the VGG16 is used for identifying the rotating text; character rotation small-angle registration is carried out by using a variance method; the text position detection precision and the text recognition precision are effectively improved; YOLOv3 and ENet are adopted, so that text correction and position acquisition are faster and more accurate; the loss of the computer and the requirement on the performance of the computer areeffectively reduced; the detection of the text with the uncertain length is realized by utilizing the characteristics of the LSTM in the CRNN; The detection performance is effectively improved, the good recognition performance is achieved in natural scenes such as non-uniform illumination, complex backgrounds, multi-language mixing, text complex formats, product label picture rotation, affine distortion and perspective distortion, and convenience is provided for inputting of label information of metal plate strips.

The invention provides a radar target recognition method based on a sparse feature. The method is used for improving the pertinence of data recognized by a radar target and reducing the computation burden of data processing by taking the sparse feature of a radar target image as a radar target image training sample and the recognition feature of a radar target to be measured; then, a sparse linear equation of the radar target to be measured is established according to the sparse feature of the training sample; then, the sparse linear equation is solved by using a Bayes compressed sensing algorithm, and the radar target to be measured is recognized by means of a compressed sensing theory instead of target azimuth estimation, so that the recognition complexity is lowered, and the recognition accuracy is prevented from depending on the target azimuth estimation; and meanwhile, the radar target recognition based on the compressed sensing theory also has favorable recognition performance under a noise environment, thus the problems that a radar target recognition system in the prior art is more complex and limited in recognition accuracy are solved, and the aims of increasing the processing efficiency for radar target recognition and improving the recognition accuracy are achieved.