237results about How to "Accurate and effective identification" patented technology

The invention discloses a monitoring method of parking space in a parking lot. The monitoring method comprises the following steps: a plurality of parking spaces which are marked by the use of boundary lines or boundary points and are marked with different marks therein in a parking lot, and a plurality of cameras are installed to make parking spaces to be monitored between visual angles of adjacent cameras; a server determines, from an image transmitted by each camera, coordinate position of each parking space positioned in the image; a parking space which is monitored simultaneously by at least two cameras is obtained, segmentation maps of the parking space are cut out respectively from images transferred by each camera, and a camera corresponding to a parking space segmentation map with the largest area or highest sharpness is set as the monitoring camera of the parking space; segmentation of images transmitted by each camera is carried out respectively so as to obtain corresponding parking space segmentation maps and know whether cars exist in corresponding parking spaces through image recognition; and statistical summary of empty parking spaces and corresponding parking space marks is carried out and summary information is displayed on a display screen at the entrance of the parking lot.

The invention discloses a method for extracting crown word numbers from scanned banknote images in a banknote detection module. The method comprises the following steps that the green ray of an image sensor is used for banknote collecting to form source images, the boundaries, in the source images, of the banknote images are found by adopting the mode of scanning boundary points, and then the banknote images are turned over and corrected to the horizontal state; feature points are extracted, and face value and orientation of banknotes are identified according to the obtained feature points; crown word number areas are determined, binarization processing is adopted for the crown word number areas, the obtained binarization images of the crown word number areas are searched for the boundaries containing all crown word numbers, then each single crowd word number is segmented, and a support vector machine identifies and processes the segmented and normalized single word images. The method for extracting the crown word numbers from the scanned banknote images provides a powerful support for identifying the authenticity of banknotes, crown word numbers can be identified accurately, the ability to identify the authenticity of banknotes is greatly improved, and the property safety of the financial department is effectively protected.

The invention discloses an intelligent wheelchair dynamic gesture recognition method based on Kinect depth information and relates to the field of intelligent wheelchairs. A method of dynamic gesture recognition based on Kinect depth image information is adopted, and the intelligent wheelchair dynamic gesture recognition method has strong expansibility, can solve problems of influence on gesture detection caused by complex backgrounds, large-area skin color interference and the like and can accurately and effectively separates hands. Subsequently, a method combining Camshift and Kalman filtering is adopted to fast and effectively track gestures, a fused feature extraction method is adopted, and accuracy of dynamic gesture recognition is improved. When the intelligent wheelchair dynamic gesture recognition method is applied to an intelligent wheelchair control system based on dynamic gesture recognition, fast and accurate dynamic gesture command recognition can be achieved, the intelligent wheelchairs can be safely and stably controlled, and the natural and visual man-machine interaction mode between people and the intelligent wheelchairs is achieved.

The invention relates to the text analysis field, particularly to a semantic characteristic-based text semantic similarity analysis method. According to the technical scheme, the similarity degree between texts is analyzed more accurately and effectively by calculation based on semantic relations of internal words of the texts. According to the method, shallow analysis on association relation between texts and between lexical items is performed through singular value decomposition; a lexical item-theme set is constructed by a bayesian network; the semantic similarity between the lexical items is calculated by mutual information and context; and finally, the text similarity is calculated through a graph structure. By adoption of the text semantic similarity analysis method, the semantic relation between texts can be measured and recognized more accurately and effectively.

The invention claims a response surface method for identifying the parameters of Burgers model for slope soil, comprising the following steps: determine the model parameters and the level thereof; determine the parameter combination and the test sequence of the numerical test; calculate the response equation; optimally identify the model parameters. Based on the combination of the laboratory test, the numerical simulation and the field monitoring, the response surface analysis method is adopted in the invention to determine the regression equation for the parameters of Burgers model and the slope surface displacement, and meanwhile the regression analysis is adopted to establish the optimal parameter combination.

The invention discloses a hybrid current criterion based non-invasive non-variable frequency air conditioner starting identification method. According to the method, the non-variable frequency air conditioner starting can be accurately, effectively and rapidly identified according to actual voltage and current and according to average power increment, starting inrush current multiple and 2nd harmonic proportion coefficient of a user incoming end; and when the steady-state average power increment of the user incoming end is greater than 500W and the user incoming end has 5 to 9 times of starting inrush current multiples and 5-15% of 2nd harmonic proportion coefficient, the non-variable frequency air conditioner starting is judged to exist. The invention provides an effective judgment method for identification of non-variable frequency air conditioner starting. By the method, the non-variable frequency air conditioner starting can be accurately identified in no need of entering a residential home, the method has the advantages of simplicity, economy, easiness in promotion and application and the like, an identification result helps a user to know the starting running characteristic and energy consumption condition of a large-power non-variable frequency air conditioner, and the method can be used for management of a power grid company on a demand side of the air conditioner.

The invention relates to an SCR catalyst crystallization detection method, a detection device and a clearing device, and relates to the technical field of engine after-treatment control. Real-time acquisition and accumulation of the actual intake air volume of an intake and exhaust system and the pre-calibrated intake air volume in an uncrystallized state are carried out, the crystallization stateof an SCR catalyst can be recognized according to an intake air volume deviation value calculated according to the actual intake air volume total value and the pre-calibrated intake air volume totalvalue, and when it is detected that the SCR catalyst has a light crystallization fault, an exhaust temperature management unit can be used for adjusting own control parameters of an engine to roughlyremove crystals, and when it is detected that the SCR catalyst has the severe crystallization fault, the crystals can be thoroughly removed through a DPF active regeneration unit, so that the crystallization fault in the SCR catalyst can be effectively recognized, and different removing measures can be taken for crystallization states of different degrees to prevent the SCR catalyst from being subjected to severe influence caused by crystallization.

The invention discloses a rolling bearing fault diagnosing method and equipment based on CEEMDAN and CFSFDP, belonging to the field of fault diagnosis of rotary machines. The method comprises the steps of acquiring vibration signals of a bearing in a normal state and different fault mode states so as to obtain sample points of the vibration signals of different states, decomposing by using CEEMDANto obtain time and frequency domains characteristics of bearing diagnosis and screening out bearing state representation parameters along with the time domain and frequency domain characteristics, dividing the representation parameters into training samples and a test samples, then using a table CFSFDP algorithm as a bearing fault diagnosis model, inputting the training samples into the bearing fault diagnosis model, clustering output results, obtaining clustering amount, clustering center points of each type, and state types corresponding to the clustering center points, and inspecting the trained diagnosis model by using test samples. The method and equipment can identify different bearing fault types and fault degrees accurately and effectively.

The invention relates to the technical field of power systems, and discloses a transformer area user identification method based on principal component analysis and k-means clustering. The method comprises the steps of obtaining voltage data of a low-voltage side of a transformer and a user ammeter; performing principal component analysis on the voltage data to obtain a low-dimensional analysis object; determining the number k of the K-means clusters; calculating the data density of the to-be-clustered object, and determining an initial clustering centroid; Using k-means clustering algorithm for clustering objects to be clustered, and then the transformer area user relationship is identified. By the adoption of the scheme, identification reliability is improved, the problems that efficiency is low and potential safety hazards exist are solved, and hardware and labor cost is reduced; Multi-dimensional voltage is digitalized into few main components for analysis through a dimension reduction technology, so that the calculation amount of the algorithm is reduced, and the efficiency of the algorithm is improved; Regular k The means algorithm is improved, the transformer area and the phase of the user can be accurately and effectively identified, and the cross-transformer-area user attribution problem can be thoroughly solved.

The invention provides an identification method of a lighting shielding failure of an electric transmission line. The identification method is based on waveform characteristics of traveling wave current in on-line monitoring electric transmission line phase conductors and mainly includes performing on-line monitoring to the traveling wave current, extracting data of lightning stroke failure traveling wave current on a guide line, analyzing the waveform characteristics of the lightning stroke failure traveling wave current and determining that the electric transmission line has a lightning shielding failure if a section prior the maximum traveling wave current amplitude value responds to a like polarity small pulse which is directly connected with a main discharging pulse and the pulse response duration and the current amplitude value are within certain ranges. Compared with identification methods in prior art, the identification method is explicit in judgment thoughts and tested and verified by field data, so that the identification method is capable of accurately and effectively identifying whether a waveform current signal is a signal of the lighting shielding failure electric transmission line, effectively helping electric transmission line operating maintenance departments to judge lighting stroke failure properties of the line and know lightning stroke failure reasons, and scientifically guiding line lightning protection reconstruction.

The invention discloses a certificate image text recognition method and system based on deep learning, belongs to the field of certificate image recognition, and aims to solve the technical problem ofhow to realize effective recognition of texts in a certificate image in a natural scene. The method comprises the following steps: preprocessing a certificate image to remove noise to obtain a preprocessed image; performing text detection on the preprocessed image based on a CTPN algorithm to obtain a text area of the certificate image; screening the text area of the certificate image through theimage position template to obtain a target text area of the certificate image; taking the target text area of the certificate image as input, and training the character recognition model through a TensorFlow Slim algorithm to obtain a trained character recognition model; and recognizing the font to be recognized through the trained character recognition model. The system comprises a preprocessingmodule, a text detection module, a text region module, a model training module and a test module.

The invention brings forward a rolling bearing fault diagnosis method based on improved fractal box dimensions and a self adaption gray correlation theory. Fault characteristics (including useful and easy-to-differentiate information capable of reflecting different working states of a bearing) are extracted from bearing vibration signals via the improved fractal box dimensions, and fault types and different orders of severity of bearings can be automatically identified via self adaption gray correlation algorithms. The rolling bearing fault diagnosis method is used for solving a problem that work and health states of rolling bearings cannot be easily assessed via traditional time domain and frequency domain methods; via the rolling bearing fault diagnosis method, fault types and different orders of severity of different bearings can be accurately and effectively identified.

The application discloses a switch cabinet fault feature selection method and apparatus. The method comprises: based on a sorting principle of importance degree of fault features, performing sorting on N fault features of a fault feature set to obtain to-be-selected fault feature subsets; based on mRMR criterions, performing screening on the to-be-selected fault feature subsets with an increment search method; and calculating the classification accuracy of each candidate fault feature subset in N candidate fault feature subsets obtained after screening, and determining the candidate fault feature subset with highest classification accuracy as an optimal candidate fault feature subset. According to the switch cabinet fault feature selection method and apparatus, the fault features in the fault feature set are sorted in advance, so that the dimension reduction of the fault features is realized; and then the screening is performed based on the mRNR criterions, so that noise components in a fault feature sample are effectively reduced, the feature dimensions are further reduced, the screening effect is improved, and the degree of matching between the subsequently obtained optimal fault feature subset and an actual fault reason is increased.

The invention relates to the field of data analysis, and discloses a method for carrying out transformer area user identification based on optimized supervised learning. The method comprises the following steps of determining a user with a known station user topological relation and a station area and a phase to which the user belongs, determining a corresponding tag of user data according to thestation area and the phase to which the user belongs, establishing a training set, a verification set and a test set, determining k parameters in a KNN model by adopting a cross verification mode, andcompleting model training; and identifying and classifying the voltage data to be identified by adopting the trained model and the determined k value, thereby realizing the identification of the users in the transformer area to be identified. According to the invention, conversion from unsupervised learning to supervised learning is realized; a training set, a verification set and a test set arereasonably set, and k parameters are determined by adopting a cross verification mode, so that the transformer area and the phase of a user are accurately and effectively identified, the problem of cross-transformer-area user ownership is thoroughly solved, and a foundation is laid for comprehensively guiding work in the fields of operation, maintenance, first-aid repair, technical improvement, planning and the like of a low-voltage transformer area.

The invention disclose an earth fault phase identification method based on an injection method. The earth fault phase identification method includes the steps of (1) determining a phase angle Theta nor and the compensation state of an arc extinguishing coil based on system parameters (system total capacitance and damping ratio); (2) injecting current to a neutral point to obtain a compensated neutral point voltage when a power grid has single-phase earth faults; (3) determining the range of the compensated neutral point voltage phase angle a varying along the fault resistance when each phase has the single earth faults; and (4) determining the phase angle a falling into which variation range in step (3) based on the measured phase angle a of the neutral point voltage to determine whether the phase corresponding to the variation range has earth faults or not. The earth fault phase identification method based on the injection method can effectively realize the system earth fault phase identification, particularly realize the high-resistance earth fault phase identification, thereby overcoming the limitation of a conventional fault phase identification method and having better application prospects.

The invention discloses a fault tolerance control method and device of a multiphase motor. The fault tolerance control method comprises the following steps: when at least one phase of the multiphase motor is subjected to a phase loss fault, obtaining a phase voltage and phase current of a normal phase that is not subjected the phase loss fault; transforming the phase voltage and the phase currentof the normal phase to obtain stator current and a stator voltage in a two-phase stationary coordinate system; obtaining an amplitude and a phase of a rotor flux linkage and a rotating speed parameterof the multiphase motor according to the stator current and the stator voltage in the two-phase stationary coordinate system; transforming the stator current in the two-phase stationary coordinate system according to the phase of the rotor flux linkage to obtain the stator current in a two-phase rotating coordinate system; obtaining a reference phase voltage of the normal phase according to a setvalue of the amplitude of the rotor flux linkage, a set value of the rotating speed parameter, the amplitude of the rotor flux linkage, the rotating speed parameter, the phase of the rotor flux linkage and the stator current in the two-phase rotating coordinate system, and controlling the multiphase motor according to the reference phase voltage of the normal phase.

The invention relates to an information pushing method and device based on human-computer interaction and computer equipment. The method comprises the following steps: receiving session information sent by a user terminal, and carrying out preprocessing and word segmentation processing on the session information to obtain a plurality of session texts; inputting the plurality of session texts intoan intention recognition model, and outputting an intention type corresponding to the session information; inputting the plurality of session texts into a trained information extraction model, and calculating the matching degree of the plurality of session texts and a plurality of structured texts in a structured corpus to obtain target field information corresponding to the plurality of session texts; generating corresponding target consultation information according to the intention type and the target field information; and matching corresponding target push information according to the target consultation information, and sending the target push information to the corresponding user terminal. By adopting the method, the accuracy of identifying the structured text in the specific fieldin the session information can be effectively improved, so that the matched information can be accurately and effectively pushed to the user.

The invention provides a malicious traffic detection method and system, a computer and a medium. The method comprises the following steps: acquiring traffic data to be detected; extracting a to-be-analyzed field information set of the to-be-detected traffic data, and dividing the to-be-analyzed field information set into a training set and a test set according to a preset proportion; determining to-be-analyzed features of the training set and the test set; inputting the to-be-analyzed features of the training set into a plurality of preset classifiers for training, and integrating the plurality of preset classifiers through a soft voting method to obtain a malicious traffic detection model; and inputting the to-be-analyzed features of the test set into the malicious traffic detection model for testing to obtain a prediction result. The technical effect that malicious traffic can be accurately recognized without decryption is achieved, the generalization ability and safety of the malicious traffic detection model are improved, and the invention can be deployed on different firewalls, intrusion detection systems and intrusion prevention systems and has good universality.

The invention provides an Internet of Things terminal abnormity detection and response method and system, relates to the technical field of Internet of Things, and can solve the technical problem that the security problem of the Internet of Things terminal is serious. According to the specific scheme, the method comprises steps of receiving terminal information sent by a to-be-detected Internet of Things terminal; determining a communication abnormal port according to communication port information in the terminal information; inquiring a process file corresponding to the communication abnormal port from file information in the terminal information; detecting the process file to obtain an abnormity detection result; and sending a response instruction to the to-be-detected Internet of Things terminal according to the abnormity detection result.

The invention discloses a charging pile harmonic wave detection algorithm based on a windowing interpolation FFT and a wavelet packet; the algorithm comprises the following steps: windowing interpolation FFT transformation, using a Blair window and double-spectral line interpolation to carry out FFT transformation for an obtained signal, thus obtaining various harmonic waves in the signal; selecting a to-be analyzed frequency range, extracting harmonic waves having big influence on power grid characteristics, and neglecting harmonic waves with almost no influence, thus improving detection instantaneity; using the wavelet packet to decompose and reconstruct the selected frequency range , applying DB3 wavelet packet transformation on the signal, thus obtaining start-stop times of various generated harmonic waves. The algorithm can fast and accurately detect each harmonic wave of the electric vehicle charging pile, and can determine harmonic wave generation start-stop times.

The invention relates to an advertising image identification method, an electronic device and a readable storage medium. The method includes: performing optical character recognition on a to-be-analyzed image to recognize text; performing word segmentation on the recognized text; matching each word with each advertising keyword in a preset advertising keyword library to obtain the words matched with the advertising keywords in the advertising keyword library; allocating corresponding a corresponding keyword matching score according to the matching result and a preset matching scoring rule; identifying the different font sizes of the text in the to-be-analyzed image, and allocating corresponding word scores according to the matched word font sizes and a preset font scoring rule; using a preset rule to judge whether the to-be-analyzed image is an advertising image or not according to the keyword matching score and the font scores. By the method, whether the to-be-analyzed image is the advertising image or not can be judged accurately and effectively, manual detection is not needed, automatic advertising image identification can be achieved, and identification efficiency can be increased effectively.

The invention relates to the field of data analysis, in particular to a transformer area user identification and discrimination method based on a supervised learning method. The method comprises the steps of determining a corresponding tag of user data according to a transformer area and a phase to which a user belongs, establishing a training set, a verification set and a test set, and determining in a cross verification mode; Identifying the voltage data of the to-be-identified user by adopting the trained training model; And establishing a quantitative evaluation index of the reliability ofthe transformer area user identification result, and calculating the reliability of the primary identification result of the transformer area user. According to the invention, conversion from unsupervised learning to supervised learning is realized, the hardware and labor cost is reduced, and the identification result is more reliable; Meanwhile, a quantitative evaluation index of a transformer area user household change relation identification result is established, accurate identification of the dispute user is achieved, the transformer area to which the user belongs and the phase are accurately and effectively identified, the cross-transformer-area user ownership problem is thoroughly solved, and a foundation is laid for comprehensively guiding work in the fields of operation, maintenance, first-aid repair, technical improvement, planning and the like of the low-voltage transformer area.

The invention discloses an MSC-CFS-ICA-based apple slight damage hyperspectral detection method. The MSC-CFS-ICA-based apple slight damage hyperspectral detection method comprises the following stepsof acquiring an original hyperspectral image of an apple sample; extracting average spectral reflectivity data of the original hyperspectral image; preprocessing the average spectral reflectivity of an original hyperspectrum by using a multiplicative scatter correction algorithm; for the preprocessed spectral data, adopting a feature selection algorithm based on correlation, and selecting out a characteristic wavelength with high correlation with apple damage classification; and based on the characteristic wavelength, performing conversion by using independent component analysis to obtain a damage image, and obtaining a damage area through an adaptive threshold segmentation algorithm, so that the detection of apple damage is completed. The method has the beneficial effects that an AmericanSOC710VP hyperspectral imager is used for collecting hyperspectral images of Fuji apples with the waveband range of 400-1,000 nm (128 wave bands in total) in normal and different damage time.

The invention discloses an improved fractal box dimension algorithm and self-adaptive grey relation theory algorithm-based rolling bearing fault diagnosis method. The method comprises the following steps of: firstly extracting fault features (comprising information which is capable of reflecting different working sates of a bearing, and is more useful and easier to distinguish) from bearing vibration signals through an improved fractal box dimension; and automatically recognizing fault types of the bearing and different severity degrees through a self-adaptive grey relation theory algorithm. The invention aims at solving the problem that the traditional time domain and frequency domain methods are difficult to correctly estimate the working health conditions of rolling bearings, so that different rolling bearing fault types and fault severity degrees can be correctly and effectively recognized.

The invention provides a blood pressure measuring device, a method for effectively improving the blood pressure measuring accuracy on the basis of the measuring device and a method for judging moments of the closed state and the open state of an artery blood vessel. A blood pressure value is accurately obtained through continuous collection of Korotkoff vibration signals, detection of heartbeat periodic waveforms and extraction of fixing belt pressure signals. The problems that an existing sphygmomanometer is easily disturbed and the detection reliability is limited are effectively solved, a functional relationship of the Korotkoff vibration signals and the heartbeat period of the human body can be accurately and effectively recognized, and therefore the human body blood pressure accuracy value is accurately determined. The device can collect multiple physiological parameters at the same time, and provides data supports for related medical diagnosis and scientific research. The measuring device and method are applicable to the multiple fields of conventional blood pressure measurement, clinical diagnosis of related diseases of the heart, blood vessels, blood and the like, family health care and the like.

The invention discloses a node voltage sag severity multi-index evaluation method, which selects a plurality of evaluation indexes to establish an evaluation index set, overcomes the defect that onlyone aspect of the power supply side or the power consumption side is considered in the traditional evaluation method, and combines the two together to make the evaluation result more comprehensive. Atthe same time, the comprehensive weighting method based on AHP and CRITIC method is used to give new weight to the indexes, which not only fully considers the deep internal correlation among the evaluation indexes, but also reflects their relative importance, so that the evaluation results are more reasonable. The simulation results show that the method can reflect the actual situation of distribution network more truly, can accurately and effectively identify the weak nodes and vulnerable areas in the network, and lay the foundation for the treatment of voltage sag.

The invention discloses a method for identifying key proteins based on a cuckoo search algorithm, and belongs to the technical field of biological information. A protein interaction network is converted into an undirected graph, denoising processing is conducted on the protein interaction network to acquire gene expression data, subcellular localization information, standard protein complex information and GO annotation information corresponding to the protein, weighting processing is conducted on edges and nodes of the purified protein interaction network to generate a bird nest and cuckoos and update a bird nest adaptive value to generate the key proteins. The method can accurately identify the key proteins; simulation results show that the indexes such as sensitivity, specificity, F measure, a positive predictive value, a negative predictive value, correct rate and the like are superior; compared with other existing identification methods, the optimizing characteristics of the cuckoo search algorithm and topological characteristics of the protein interaction network are combined to identify the key proteins, and the recognition accuracy of the key proteins is improved.

The invention provides prompting message sending method and system for phone fraud, and belongs to the technical field of network communication. The prompting message sending method for the phone fraud comprises the steps of A, indicating a calling MSC (Mobile Switching Center) for call connection through an SCP (service control point) after a calling from a calling user is triggered to the SCP by the calling MSC, so as to build a voice channel between the calling user and the called user; B, determining whether the calling number of the calling message is included in a phone fraud number list when the SCP receives the calling message between the calling user and the called user; and if so, sending the prompting message for the phone fraud to the called user. The prompting message sending method for the phone fraud has the advantages that the user can be prompted on real time for the phone fraud received.

The invention relates to a session interaction processing method and a device thereof, computer equipment and a storage medium. The method comprises the following steps: acquiring voice information input by a user terminal; performing voice endpoint detection and voice recognition on the voice information to obtain voice endpoint information and session information corresponding to the voice information; recognizing silence duration according to the voice endpoint information, and determining a current context according to the context session information; obtaining matched response strategy information according to the silence duration and the current context, and sending the response strategy information to a user terminal; and continuously matching corresponding response strategy information for interaction according to the silence duration of the voice information input by the user terminal and the current context until the preset response ending strategy information is matched, andsending the response ending strategy information to the user terminal. By adopting the method, the matching accuracy of the response information can be effectively improved, so that the interaction efficiency is effectively improved.

The invention provides a vehicle collision early warning processing method and device, and the method comprises the steps: obtaining vehicle information and road network information of a current vehicle and a target vehicle, and carrying out the road network matching of the current vehicle and the target vehicle according to the vehicle information and the road network information; if a road network matching result is that the target vehicle is in front of the current vehicle and is located in the same road section lane, establishing a vehicle body coordinate system of the current vehicle, anddetermining a coordinate set of a predicted trajectory of the current vehicle within a predetermined time; determining that the current vehicle has a collision risk with the target vehicle accordingto a coordinate set of the predicted trajectory of the current vehicle; according to the method, the problem that the way of determining collision risk by calculating the safe vehicle distance betweenthe vehicle and the surrounding adjacent vehicles based on the positioning information of the vehicle in the prior art cannot be suitable for forward collision early warning in a curve scene can be solved, and the method can be well compatible with forward collision scenes of a straight road and a curve.