76results about How to "Improve identification rate" patented technology

The invention discloses a method and system for increasing judgment accuracy of monoisotopic peaks. The method comprises the following steps: according to selected tandem mass spectrums, determining candidate isotopic peak clusters; according to the similarity and the strength ratio of the chromatographic outflow curves of the candidate isotopic peak clusters, determining the start and stop of the isotopic peak clusters; according to the start and stop of the isotopic peak clusters, determining the quality of the monoisotopic peaks; and judging whether the monoisotopic peaks of the tandem mass spectrums are not determined or not, if yes, returning and selecting a new tandem mass spectrum to determine a candidate isotopic peak cluster, and if not, ending.

The invention discloses an audio-visual cognitive event-related electroencephalogram-based identification method, which comprises the following steps of: performing wavelet packet decomposition and reconstruction on a series of filtered independent components to acquire electroencephalogram signals within electroencephalogram activity source rhythm ranges, and performing empirical mode decomposition on electroencephalogram signals of each rhythm to obtain corresponding feature mode components which reflect feature rhythm vibration modes; extracting inter-cognitive event phase locking strength features, inter-event synchronous features and driving and response coupling features from the feature mode components by adopting an inter-trial coherence index, a phase intrinsic coherence index, a phase shift intrinsic coherence index and a partial direct coherence index; extracting the power spectrum features of an audio-visual cognitive event-related electroencephalogram by adopting event-related spectral perturbation; and classifying the inter-cognitive event phase locking strength features, the inter-event synchronous features, the driving and response coupling features and the power spectrum features of cognitive events of a plurality of subjects by adopting a support vector machine multi-classifier to obtain identification results. According to the audio-visual cognitive event-related electroencephalogram-based identification method increases the identification rate.

The invention discloses an improved canonical correlation analysis-based physiological signal fusion identity recognition method, and mainly solves the problem of relatively low identification rate of an existing method. The method is implemented by comprising the steps of 1) obtaining pulse waves and respiratory signals, and preprocessing training data; 2) intercepting waveforms of the training data, and obtaining a pulse wave training set and a respiratory training set; 3) calculating intra-class and inter-class neighborhoods of the two sets; 4) according to the intra-class and inter-class neighborhoods, calculating intra-class and inter-class correlation matrixes of the two sets and constructing a regular canonical correlation analysis objective function; 5) solving regular canonical correlation analysis-based pulse wave and respiratory conversion matrixes; 6) calculating training fusion eigenvectors by utilizing the conversion matrixes; 7) obtaining pulse wave and respiratory signal test data and calculating test fusion vectors; and 8) performing type judgment on the test fusion vectors to obtain an identity recognition result. According to the method, the identity recognition rate is increased; and the method can be applied to the e-commerce and remote medical identity authentication.

Using registered one piece of frontal normal personí»s face image, the method can obtain virtual images of the personí»s face at different postures. Further, using recognition strategies including separated two stages of posture recognition and identity recognition, the method solves issue between registered image and changed postures of personí»s face at time of recognition. The method includes: 3D deformable model of personí»s face for representing structural statistical information of personí»s face; reconstruction algorithm for reconstructing 3D personí»s face from frontal personí»s face image, and two stages of posture recognition and identity recognition. Under condition of only one registered frontal normal personí»s face image, the invention still obtains higher recognition rate from side test images.

The invention provides a method for discriminating liquor odor by a liquid array taste sense bionic sensor, and the method comprises the following steps: simulating a mammal taste sense system, constructing the liquid array bionic sensor with cross response by using 8 sensitive substances, detecting liquor body directly, establishing a discrimination function scatter diagram for many different odor types of liquor by using a mode identification analysis method, carrying out mode identifications of an unknown odor type of liquor to be detected, finally comparing the mode identification result of liquor to be detected with discrimination function scatter diagrams of different odor types of liquor, finding the identification result of the liquor to be detected which approaches to the scatter range of the liquor standard sample with the same odor type most, and considering the detected liquor sample as the odor type. According to the liquid array sensor provided by the invention, an immobilized treatment of the probe is not needed to be carried out, and when the analyte is in liquid state, the analyte and the liquid array are reacted completely with rapid response advantage, accordingly the liquor odor type detectability is high.

The embodiment of the invention discloses a bill discrimination method, device and equipment, and a storage medium. The method comprises steps of carrying out binarization on a fluorescence region of a fluorescence image of a to-be-discriminated bill so as to obtain a binarization fluorescence region; blocking the binarization fluorescence region so as to obtain multiple binarization fluorescence region blocks; matching the binarization fluorescence region blocks with corresponding blocks of detection templates, and determining the authenticity of the to-be-discriminated bill according to matching results. According to the invention, by carrying out binarization on the fluorescence image of the to-be-discriminated bill, blocking the binarization image and carrying out matching by use of multiple detection templates, time complexity of an algorithm can be reduced and discrimination rate of the bill authenticity is increased, there is no need to re-check the bill, high-speed reorganization requirements of the bill can be met and bill discrimination efficiency can be improved.

The invention provides a polarization discrimination method for active false targets in a main lobe suppressing interference background. The polarization discrimination method comprises the steps of: receiving signals by using an H polarization channel and a V polarization channel of a radar, carrying out matched filtering according to sub-pulse waveforms at first after synchronous sampling, carrying out interference polarization estimation and polarization cancellation on the two paths of output signals, processing one path of output signal by adopting an M point sliding window method to realize non-coherent accumulation and further adopting a CFAR method to realize target detection, extracting target features of a detected sampling position, and carrying out discrimination of real and false target attributes by utilizing the features. According to the polarization discrimination method, the radar adopts a transmission variable polarization method, and can perform active false target discrimination through extracting signal amplitude variation features.

The invention provides a tandem mass spectrogram identification method which is characterized by comprising the following steps: 1) conducting restrictive searching in a global sequence library on each spectrogram in a spectrogram dataset to be identified, so as to obtain a peptide fragment I matched with the spectrogram; 2) establishing a local sequence library according to the peptide fragments I obtained in step 1), and conducting open search in the local sequence library on the spectrogram in the spectrogram dataset to be identified, so as to obtain modified peptide fragments II matched with part of the spectrograms, and obtain the modification mass and error burst; 3) setting a restrictive searching interval for the spectrogram in the spectrogram dataset to be identified according to the matched modification mass and error burst in the step 2) and the mass of the current spectrogram to be identified, searching in the global sequence library, and obtaining a final matching result. The tandem mass spectrogram identification method has the advantages that the identification rate and accuracy are improved; the searching speed is higher.

The application discloses transfer paper with an anti-counterfeiting fuzzy holographic pattern, which comprises a bottom layer, a glue layer, an aluminum plating layer, a surface release layer and a water-based varnish layer which are sequentially compounded, wherein holographic fuzzy characters are arranged on the surface release layer. The manufacturing method of the transfer paper comprises the following steps of coating, plate making, mould pressing, aluminum plating, transferring and water-based varnish coating. The invention has reasonable production process and high production efficiency and can ensure product quality. The product of the invention has good anti-counterfeiting performance, and consumers can directly distinguish with the naked eyes.

The invention provides a method for analyzing data of prokaryotic proteogenomics rapidly and automatically. Users only need to provide mass spectral data and a corresponding database file and set a simple retrieval parameter; retrieval of data of proteogenomics can be completed; simultaneously, a user defined data retrieval result can also be compatible; therefore, the identification coverage rate of the data of proteogenomics is increased; in the method disclosed by the invention, library search engines of different algorithms are integrated in advance; disadvantages of a single retrieval method are made up; a user defined library search result can also be compatible; the method has good compatibility; the peptide fragment identification coverage rate is increased to the most extent; in the method disclosed by the invention, functional annotation of new genes is automatically completed; large-scale identification of protein post-translational modification and analysis of non-labelled quantitative proteomics are realized for the first time; and automatic and rapid deep analysis of the data of proteogenomics is really realized.

The invention discloses a display method of tempered martensitic steel carbide colour metallography, comprising the following steps of: (1) grinding and polishing the surface of a tempered martensitic steel sample to obtain a tempered martensitic steel metallographic sample; (2) heating the tempered martensitic steel metallographic sample in the oxidation atmosphere to 400-500 DEG C, keeping warm for 6-15 minutes, and forming an interference oxide-film, which makes the microstructure of metal and alloy to present different colors by the action of light, on the surface of the tempered martensitic steel sample, so as to display the tempered martensitic steel carbide color metallography. The method provided by the invention can be used to effectively display and raise the contrast between the T91 steel carbide and the matrix microstructure, so as to effectively observe the microstructure form of the martensitic steel.

The invention discloses a PSE meat screening method based on a near infrared spectrum. The method comprises the following steps: acquiring a near infrared spectrum of a to-be-detected meat sample of the slaughtered pork longissimus dorsi by using a near-infrared spectrograph, and preprocessing the acquired near infrared spectrum of the to-be-detected meat sample; and screening the PSE meat by virtue of the average spectrum of a normal meat sample, the average spectrum of a PSE meat sample, the normal meat sample threshold value and the PSE meat sample threshold value in a PSE meat qualitative analysis model based on the near infrared spectrum, wherein the meat sample is a normal meat sample when the distance between the spectrum of the to-be-detected meat sample and the average spectrum of the normal meat sample is smaller than the normal meat sample threshold value, and the meat sample is a PSE meat sample when the distance between the spectrum of the to-be-detected meat sample and the average spectrum of the PSE meat sample is smaller than the PSE meat sample threshold value. The method is high in speed, high-efficiency and high in identification rate, and the meat is not damaged. A novel scientific, standard and rapid method for screening PSE meat on line in a slaughter house is provided.

The invention discloses a target crowd tracking and monitoring method based on a limited range scene, and the method comprises the steps: S1, designing and achieving a whole set of camera placement and collection device based on a limited visual field range, so as to meet the demands of a target crowd activity scene and the full coverage of a visual field; S2, modifying the size and proportion ofan anchor frame of a detection network model according to the body type and face type proportion of the target crowd, and modifying a loss function of the detection network to improve the detection accuracy; S3, recognizing the human face by adopting a specific recognition network while detecting the human face, so as to improve the detection and recognition precision while not reducing the speedunder the condition that the detection network and the human face recognition network are used at the same time; and S4, improving a multi-target tracking monitoring model, and improving the model adaptability of target crowd tracking monitoring through extraction of specific features and pruning of a specific scene. According to the design scheme, in a visual scheme of nursing the target crowd and preventing accidents, the monitoring accuracy and the positioning speed are improved while no leakage and real-time performance are realized.

The invention provides an authentication method based on object doppler velocity estimation and aims at addressing the problem of deteriorating authentication effects along with the increasing intensity with the false objects in the course of authenticating intense distance false object interference by a radar. The method includes the following steps: firstly selecting a standard radar and conducting measurement grouping pre-processing, converting other networking radar object measurement to a standard radar local coordinate system and conducting measurement association based on position information in order to rule out irrelevant false measurement and noise wave and at the same time reduce calculation; then estimating an object doppler velocity based on the position information measured after the coordinate conversion, conducting authentication in accordance with the standard radar doppler velocity measurement so as to authenticate the false object. Compared with classical authentication methods which associate based on position information, according to the invention, the method herein can guarantee high authentication rate of true and false object under circumstances when intensity from the multiple false object is greater.

The invention relates to the technical field of identification of traditional Chinese medicines, and discloses a method for tracing to production places of market saffron via ATR-FTIR in combination with a RBF neural network. The method comprises the steps of 1, pretreating a sample; 2, collecting infrared spectrum data; 3, pretreating spectrum data; and 4, putting data into a model: putting dataacquired in the step 3 into a RBF neural network tracing model, and judging the production place of the sample according to a result. According to the method, the 129 collected saffron samples produced by 7 different countries are used as research objects, and a saffron production place judgment model having less limitation on the analysis samples is built by using an ATR-FTIR united quantitativeanalysis technique; the method provided by the invention is high in accuracy, high in analysis speed, less in sample consumption and low in detection cost, and does not need standard substances.

The invention discloses an MIR spectral analysis model and an identifying method for excessive cadmium content of rice. The analysis model is established by use of the steps of sample selection and judgment of qualified rice, acquisition of sample spectral data, processing of the sample spectral data, screening of the sample spectral data, establishment of a preliminary analysis model, verification of the preliminary analysis model and the like, and the MIR spectral analysis model for identifying excessive cadmium content of rice is determined. The invention also provides a method for identifying whether the cadmium content of rice is excessive by use of the MIR spectral analysis model. According to the invention, on the basis of identifying whether the cadmium content of samples of a large amount of rice is excessive by use of a chemical method, the MIR spectral analysis model for identifying whether the cadmium content of rice is excessive based on mid-infrared spectroscopy and a chemometrics method is established by acquiring the mid infra-red spectrum of the rice sample in combination with the least square support vector machine classification method, and the accuracy rate is more than 93.1%; the MIR spectral analysis model and the identifying method for excessive cadmium content of rice have the advantages of simplicity in operations, quickness in detection, safety, environmental protection and the like.

The invention belongs to the field of radar data processing, and aiming at a difficult problem of autonomous recognition and steady jamming suppression of dense false target jamming, the invention designs a data layer radar dense false target jamming resisting method based on multi-dimensional feature extraction. The method comprises the following steps: firstly, aiming at a dense false target jamming signal and motion feature difference of target echo, performing primary identification for the dense false target jamming through a radial velocity contrast method; secondly, based on the dense false target jamming signal and the spatial feature difference of target echo, performing comparative analysis for the target position and the obtained echo trace point, thereby eliminating infinitelydense echo trace points and flight paths; then, comprehensively identifying the dense false target jamming through spatial and motion information of various moments, performing identification and suppression for the determined dense false target jamming, and retaining an uncertain target echo signal; finally, performing final identification and suppression for the dense false target jamming through a method of multi-radar data association.

The invention provides a spiral photoelectric yarn cleaner. The spiral photoelectric yarn cleaner comprises a monitoring ring provided with a reflecting coating; a yarn needing to be monitored passes through the center of the monitoring ring; a rotatable optical emission head sweeping over the plane of the monitoring ring is arranged on the inner wall of the monitoring ring; the monitoring ring is slidingly arranged on an annular support and driven to rotate by use of a driver; optical monitoring devices for monitoring light emitted by the optical emission head are arranged at the two sides of the plane over which the light emitted by the optical emission head sweeps. Light emitted by the optical emission head of the spiral photoelectric yarn cleaner is reflected continuously on the inner wall of the monitoring ring so that a similarly circular optical trajectory is formed; when the optical trajectory is blocked by the yarn, the diffuse reflection of the light is formed on the surface of the yarn and monitored by the optical monitoring devices; as the diameter of the yearn is monitored by virtue of the optical trajectory in an annular form, the optical trajectory is applicable to yarn defects of any shapes, and consequently, the discrimination rate of the yarn defects is improved.

The invention discloses a method for simultaneously identifying short-cut denitrifying bacteria and triclosan degrading bacteria based on a DNA stable isotope nucleic acid probe technology, and belongs to the technical field of sewage biological treatment technologies and microbial ecology. The method comprises the steps of (1) starting a short-cut denitrification system based on granular sludge and biologically removing triclosan in the system; (2) carrying out 13C-propionic acid sodium and 13C-triclosan isotope microcosm experiments; (3) separating heavy-layer DNA by adopting an ultra-high-speed centrifugation method based on a cesium chloride density gradient solution; and (4) carrying out DNA marking degree evaluation and flora structure analysis. According to the method for simultaneously identifying active short-cut denitrifying bacteria and triclosan degrading bacteria in the short-cut denitrification granular sludge system based on the DNA stable isotope nucleic acid probe technology, a microorganism capable of simultaneously realizing short-cut denitrification and triclosan removal can be explored and disclosed, and a theoretical support is provided for an actual process of removing triclosan by short-cut denitrification in a sewage treatment plant.

The invention discloses a method for obtaining a lot of target nucleic acids form mixed nucleic acid, the mixed nucleic acid comprises the target nucleic acid and non-target nucleic acid. The method comprises the following steps: gathering the target nucleic acid from the mixed nucleic acid, wherein the gathering means usage of methyl binding domain protein for binding the non-target nucleic acid, removing a formed methyl binding domain protein- non-target nucleic acid bonder; amplifying the enriched target nucleic acid, wherein the amplifying means usage of a N6 random primer; wherein, a lot of target nucleic acids means that the target nucleic acid amount reaches a minimum nucleic acid amount which exceeds sequencing requirement, and the N6 random primer is a sequence with the length being 6 bp combined by A,T,C and G with random arrangement. The invention also discloses a method for identifying low-abundance strain in an anthropogenic genome sample.