60 results about "Fuzzy pattern recognition" patented technology

The invention relates to an eccentric pump vibration accidence fusion diagnose method and relative vibration signal collector, wherein said invention is characterized in that: it uses the eccentric pump vibration signal collector to collect the normal state, the quantity imbalance, asymmetry rotate and loose base of eccentric pump; uses wavelet decomposition and reconstruction to extract the character of vibration signal; and inputs the character vectors into sub fuzzy neural networks I and II; to be treated and replace the relation factor matched with sensor signal function; the whole fuzzy neural network comprises data fuzzy layer, input layer, hidden layer and output layer; uses D-S theory to obtain the fused signal function distribution, realize the fusion diagnose on normal state, quantity imbalance, asymmetry rotate and loose base. The invention has simple structure and high effect.

A method based on capacitance chromatographic imaging system for measuring the two-phase oil-gas stream features that a reverse projection algorithm is used to reconfigure the oil-gas stream's medium distribution image to real-time show the stream type in pipeline, a fuzzy pattern recognization is used for in-line automatic recognization of stream type, a combined image rebuilding algorithm is used to rebuild the image for showing oil and gas distribution, and the image is processed to obtain the porosity of the oil-gas stream. Its equipment is composed of array-type capacitance sensor, data acquisition unit and computer. Its advantages are high speed and high precision.

The invention relates to a driver identity identification and driving state monitoring method based on machine learning and deep learning. Motion data of an automobile are acquired through a smart phone sensor, thereby identifying motion of a vehicle driver. Fuzzy mode identification is utilized for dividing the motion sequence of the driver to a driving operation. Then according to road traffic information and an image photographing device, front obstacles and a jam condition in automobile driving are identified through computer vision technology, and furthermore different driving scenes are divided. According to a driving operation, statistics characteristics are respectively extracted. Furthermore a characteristic vector is formed as an input of a deep neural network. The identity of the driver is identified through constructing a personal driving characteristic database and training a corresponding deep neural network model. After the identity of the driver is confirmed, the driving state of the driver at each time is identified through a recursion neural network. According to the method of the invention, multiple-signal-source data are utilized; and based on the driving operation and the driving scene, a deep learning method is utilized for improving identification accuracy.

The invention discloses an electroencephalographic-signal-based fatigue state recognizing method. The method comprises the steps that: an electroencephalographic (EEG) data acquisition instrument records electroencephalographic signals in different fatigue states from the surface of a human scalp; a signal acquisition analog circuit filters out interference factors from the signals, performs program-controlled amplification on weak electrical signals and removes level drifts to obtain analog electroencephalographic signals; a digital circuit performs analog / digital (AD) conversion on the electroencephalographic signals to obtain digital electroencephalographic signals, and the digital electroencephalographic signals are transmitted to a host and processed; and the host preprocesses the signals first, then extracts feature information from each channel of signals to establish a feature vector, and finally evaluates the fatigue degree according to the obtained electroencephalographic feature by an evaluation method based on fuzzy pattern recognition. By combining a biomedical signal processing technology and a fuzzy pattern recognizing technology, the invention provides an objectiveand feasible mental fatigue evaluating method, so the application of the electroencephalographic-signal-based detection and recognition to the field of mental fatigue evaluation is technically improved greatly.

The invention discloses a Docker container cloud platform container scheduling method based on fuzzy mode recognition. By collecting and recording the resource indexes for all Docker containers running on each server node in the container cloud platform, the historical information about the core resources, consumed by the containers, of servers such as a CPU, a memory, network IO, and a disk IO during the running of the containers can be obtained; in addition, by means of the fuzzy mode recognition, the Docker containers are classified by using the resource consumption information. When the Docker containers running the same application are scheduled again, the corresponding container classification information can be taken as the important basis for a scheduling algorithm, the Docker containers can be scheduled to proper server nodes by means of the scheduling algorithm, so that the core resource on each server node can be more balanced in occupation. The container scheduling method is simple in realization and easy to operate, has the characteristics of expandability and easy deployment, and is suitable for providing container monitoring and container scheduling services for the applications in a distributed Docker container cloud platform.

An efficient method of searching large databases for pattern recognition is provided. The techniques disclosed illustrate how a large database of arbitrary binary data might be searched at high speed using fuzzy pattern recognition methods. Pattern recognition speed enhancements are derived from a strategy utilizing effective computational decomposition, multiple processing units, effective time-slot utilization, and an organizational approach that provides a method of performance improvement through effective aggregation. In a preferred technique, a pattern recognition system would utilize multiple processing units to achieve an almost arbitrarily scalable level of pattern recognition processing performance.

The invention discloses a collecting method of driver's driving behavior data and a collecting device of driver's driving behavior data. The method includes steps of collecting image of a driver in a car or / and outside of the car, performing grey and binarization processing on the image and forming multiple gray pictures; performing Fourier transform, orthogonal transformation and picture segmentation operation on the gray picture, and extracting target information; identifying vehicles, road lines, passengers and traffic signboard information by a fuzzy mode identification method and a statistical pattern recognition method; according to the well set alarm level, judging the alarm, and finally sending the analysis result to a sever through the network, classifying and summarizing by the server; integrating the vehicle running state and the driver's driving behavior, and using the data as basis for the third party to analyze the driver's driving behavior, or sending to the driver or insurance company. The invention can provide comprehensive dynamic information, complete data for insurance company, related O2O of automobile, traffic management and other industries, and well use big data to optimize traffic and improve safety.

The invention relates to an online leak detection device for a steam trap of a steaming-water pipeline of a thermal power plant. The online leak detection device is composed of digital temperature sensors, a digital temperature signal collector, a 485 communication cable, an optical modem, an optical cable and a data processing computer, wherein the digital temperature sensors adopt DS18B20, and a plurality of digital temperature sensors arranged in the circumferential direction of the outer side of the downstream drain water pipeline of the same steam trap share one digital temperature signal collector; the digital temperature signal collector adopts a microprocessor ARM7, and collected signals are transmitted to the data processing computer through a data communication line to be processed; and according to the data of the downstream wall surface temperature of the steam trap and the rate of change of the data, a principle of fuzzy pattern recognition is adopted to carry out fuzzy recognition on the leakage level of the steam trap. The online leak detection device has the advantages of simple system and convenience for installation and maintenance, can monitor on line in real time, and accurately judges the leakage of the steam trap in time.

The one-dimensional bar code distinguishing and decoding method in fuzzy mode is completed in a bar code decoding system comprising laser scanning unit, signal pre-treating unit, CPU unit and scanning alarm and output unit. The method includes the following steps: converting the acquired data into electric bar / gap pulse signal; signal pre-treatment; classifying treatment of code form characteristic value; fuzziness degree calculation; polymerization degree calculation; adding the fuzziness degree and the polymerization degree; calculating and matching with the corresponding code list of distinguishable code, which has the sum of fuzziness degree and polymerization degree smaller than certain constant number, while discarding undistinguishable code; discarding code incapable of matching; repeating until finishing the sampled data. The present invention is simple, efficiency and fast.

The invention discloses a combustion monitoring and diagnosis method based on feature extraction and fuzzy C-means cluster, comprising steps of utilizing time-frequency domain analysis and comentropy analysis technology to extract a flame characteristic value reflecting the combustion condition in flame detection signals, utilizing the fuzzy C-means cluster algorithm to perform FCM cluster analysis on the flame characteristic value and calculating a fuzzy combustion index through the fuzzy recognition. The invention can obtain a quantitative value of a combustion state through the fuzzy mode recognition and calculation of the combustion index, can classify the combustion states to realize quantitative monitoring of the combustion state, can promptly and accurately reflect boiler operation performance attributes, can provide new thoughts and methods for optimizing the performance and monitoring the state of the boiler, and can provide a reference model to the advanced modules of the power plant monitoring information system (boiler operation optimization, state monitoring and fault diagnosis).

In soft measurement technology frame, the easy-to-measure reliable pressure difference fluctuation signal is adopted as auxiliary variable in detecting flow pattern of two phase flow as the main variable. The acquired pressure difference signal is wavelet analyzed, corresponding characteristic values are extracted, and the characteristic values are used in fuzzy identification of the flow pattern of two phase gas / liquid flow in horizontal pipe. The key soft measurement model is established by means of combining the wavelet analysis technology and fuzzy pattern identifying technology. Then, one flow pattern identifying system for two phase gas / liquid flow is established, and the system consists of test pipeline, pressure sampling hole, pressure sampling pipe, pressure difference sensor, amplifier, data acquiring card, microcomputer and printer. The detected pressure difference signal is processed in the system for pattern identification of two phase gas / liquid flow.

The invention discloses an indoor location and landmark semantic identification method based on behaviors. The method comprises steps as follows: on the basis of action modes of an indoor person, data of the designed action modes is collected by a built-in inertial sensor arranged in a smart phone, the data is subjected to preprocessing and characteristic extraction, and action sequences are identified by a support vector machine and are recognized as behavior patterns with a fuzzy pattern recognition algorithm; an initial locating point is determined according to WIFI signals, and an approximate walking track is determined with a dead reckoning method; the behavior patterns and locations are analyzed respectively; more accurate indoor location is realized by continuously updating landmarks, and the landmarks are subjected to semantic identification according to attributes of the landmarks. When the indoor pedestrian is located, drift errors produced through dead reckoning are corrected according to the behavior patterns and a landmark set. According to the method, the location errors can be corrected by the landmarks, disturbance of environmental factors and noise of the sensor are effectively treated, and precision of indoor location is remarkably improved.

The invention provides a BCG heart rate extraction method and system based on fuzzy pattern recognition. The method includes the steps that BCG wave signals are sampled to obtain one-dimensional BCG data; modulus maximum processing is carried out on the one-dimensional BCG data to obtain the positions of all effective extreme points in the one-dimensional BCG data; selecting a point, with the amplitude larger than a preset maximum value, in all the effective extreme points to serve as an inflection point J, wherein an inflection point adjacent to the left side of the inflection point J is a point H, an inflection point adjacent to the right side of the inflection point J is a point N, and an HJN wave group set is constructed; using relative distances between every two of the point H, the point J and the point N in an HJN wave group as a generality judging criteria of the HJN wave group; selecting the HJN wave group, with the generality, in the set in a matched mode to serve as the heart beat occurrence position. The BCG heart rate extraction method and system have no requirement for BCG specific waveforms; periodic operation is carried out; the BCG heart rate extraction method and system are suitable for people of all ages and both sexes; the adaptability is high; the instantaneity is high; the operation amount is small; operation verification is carried out after the heart rate is calculated; erroneous judgment is reduced; and the reliability is high.

The invention discloses a traditional Chinese medicine cataplasma forming device with automatic detection function. The device comprises: a storage silo, a transverse feeding medicine film forming pump, a patch composite molding device, a patch longitudinal trimming device, a patch transverse cutting device, a patch dehumidifying device, a weighing device and an automatic online detection device based on FPGA (Field Programmable Gate Array) fuzzy pattern recognition, wherein the patch longitudinal trimming device is formed by a combined cutting tool magazine; the patch transverse cutting device adopts an electrostatic adsorption quantitative feeding structure; and the automatic online detection device based on FPGA comprises a plurality of cameras respectively used for stock level detection, coating film detection and cloth substrate detection, a camera and a backlight used for detection of uniformity degree, a photoelectric detector used for transverse feeding detection, a video A / D (Analog to Digital) converter ADV7181B, an SDRAM (Synchronous Dynamic random access memory) cache with the capacity of 16M, and a programmable logic device FPGA. The traditional Chinese medicine cataplasma forming device can monitor the humiture of materials and the indexes such as uniformity, size, dehumidification, weight and the like of auxiliary materials and patches.

The invention discloses a sound-wave-based square wood hole-defect recognition system and a sound-wave-based square wood hole-defect recognition method, which mainly solve the problem of difficulty for nondestructive detection of a hole defect of square wood. The system comprises a small hammer, a sound wave signal acquisition module, a sound wave signal processing module, a NiosII processor, an external storage module, a USB (universal serial bus) communication module and a principal computer processing module. The sound wave signal acquisition module is used for completing the sound wave acquisition, AD conversion, amplification and low-frequency filtering; the sound wave signal processing module is used for driving the acquisition module and processing a sound wave digital signal; the external storage module comprises an SDRAM (synchronous dynamic random access memory) and an EPCS (electronic propulsion control system) configuration memory; the NiosII processor is used for processing and caching the signal and calling the USB communication module to upload the data to the principal computer processing module; a sound wave signal time-domain and frequency-domain waveform characteristic value is extracted from the principal computer processing module, and the defect is recognized by adopting a fuzzy model recognition method. The system and the method are pollution-free, harmless, simple and convenient to use, high in recognition rate and suitable for detecting the hole defect of the square wood, and the detected wood is not damaged.

The invention mainly researches eyebrow detection and final face detection in a static colorful front image, mainly utilizing a detection method combining biomimetic pattern recognition with fuzzy pattern recognition, wherein, eyebrow characteristic vectors are extracted by adopting a Legendre moment invariant. The result of the invention proves that eyebrow can be used as an effective biologic characteristic like eye to detect face, thus expanding a new direction for the face detection method and having greater application value.

The present invention discloses a fuzzy mode identification-based blast furnace hot-blast stove energy consumption state multi-dimension assessment method. The method comprises the steps of (1) acquiring the hot-blast stove operation data in a steel enterprise energy management and control system, and calculating the energy consumption indexes of three dimensions; (2) obtaining an energy consumption index target value and an energy consumption index threshold value according to the historical data statistical analysis; (3) obtaining an energy consumption assessment index and the energy consumption assessment characteristic vectors of multiple dimensions; (4) circularly executing the steps (1), (2) and (3), and obtaining the typical energy consumption assessment characteristic vectors to form a complete state sample set; (5) clustering and classifying the sample set by a fuzzy C-means (FCM) clustering algorithm; (6) acquiring the to-be-assessed operation data of the blast furnace hot-blast stove to obtain a to-be-assessed energy consumption assessment index, adopting an evidence theory-based fuzzy mode identification method to match the characteristic vectors of the energy consumption assessment index with a symptom set in the step (5), thereby obtaining a final assessment result of the energy consumption state of the hot-blast stove.

The invention discloses a marine diesel-electric hybrid propulsion system regular energy management control method based on working condition identification. Data collected and monitored by each partcontrol unit system of a diesel-electric hybrid propulsion system are used for extracting characteristic parameters used for working condition identification, and a fuzzy pattern identification modelis adopted to identify sailing working conditions and determine standard sailing working conditions to which the sailing working conditions belongs; and then, according to the analysis and research onthe characteristics of operating conditions and the characteristics of energy transmission of the diesel-electric hybrid propulsion system, a working mode switching rule diagram of the hybrid propulsion system is established, and the switching of a working state or a working mode of each part control unit system of the diesel-electric hybrid propulsion system is completed by adopting a rule-basedenergy management control strategy. Through the marine diesel-electric hybrid propulsion system regular energy management control method based on the working condition identification, a more economical operation mode can be automatically selected and provided in the marine sailing process, and the purposes of improving the working efficiency of a diesel engine and reducing the fuel consumption isachieved.

The invention discloses a soft segmentation method of financial OCR system handwritten numerical strings. Automatic recognition processing of financial paper realizes automatic input and rechecking of financial paper, seamlessly integrates the whole process of image processing, layout analysis and intelligent identification, and comprises automatic classification of financial paper images, image preprocessing of the financial paper, identification, supervision and checking of elements in the financial paper, and the like. OCR technology is the core part in a financial paper automatic recognition processing system, requires cutting connected character strings required for automatic processing of financial paper elements into single characters, and performing character recognition. The character recognizer at the present stage has high accuracy rate, so that the overall recognition rate of the OCR system depends on the accurate rate and the acceptability of character string segmentation. The method aims at solving the technical problem of a soft segmentation method for realizing connected numeric strings based on a fuzzy pattern recognition theory, so as to improve the accurate rate of the integral segmentation flow and reduce the false rejection rate of the system, and improve the overall performance of the recognition system.

The invention discloses a short text classification method based on part-of-speech and fuzzy pattern recognition combination. for a user request text with failed domain classification, extracting feature words with different parts of speech from the historical data with correct domain classification to form basic domain features, and extracting entities in related domains in combination with a knowledge graph to form extended domain features; performing extended part-of-speech tagging on the basic domain feature and the extended domain feature to form a user-defined dictionary; pn the basis ofthe thought of first coarse classification and then subdivision, part-of-speech mode matching and the maximum membership degree principle are combined to carry out domain classification on the to-be-classified text, and finally acquiring a short text classification result with high accuracy. The method provided by the invention can be used for performing domain classification on the user requesttext in the human-computer interaction process, so that the accuracy and efficiency of short text classification are improved.

The invention discloses a user real intention-based semantic fuzzy recognition method, which comprises the following steps of: carrying out feature extraction on a large amount of historical data withcorrect field classification through a Chinese word segmentation tool and a word frequency matrix to form a feature word list; performing word segmentation on the single request text data of the sameuser and a plurality of pieces of request text data in a preset time period to obtain a word segmentation list; respectively constructing membership functions for different fields, wherein the membership functions are used for performing fuzzy pattern recognition on request text data which fail to be classified; and respectively calculating membership degrees of the request text data failed in classification to different fields, and performing field classification on the request text data failed in classification according to a principle of a maximum membership degree. According to the method, fuzzy pattern recognition is carried out on the request text data failed in classification through the maximum membership degree principle for the user request text data failed in semantic analysis,so that domain classification is carried out, the classification accuracy is improved, and the semantic analysis accuracy is further improved.

The invention discloses an image retrieval method and system based on k-nearest neighbor and fuzzy pattern recognition. The method comprises the steps that colors and texture feature vectors are extracted respectively aiming at query images and retrieved images, fuzzy normalization processing is conducted, and fusion is conducted on fuzzy colors and texture features to obtain comprehensive featurevectors of corresponding images; K near images of the query images are searched aiming at the obtained query images and the comprehensive feature vectors of all of the retrieved images; the similarity between the query images and the k near images is calculated, and the similarity among each retrieved image and the k near images of the query images is calculated to obtain corresponding k-dimensional fuzzy feature vectors of the query images and each retrieved image; the fuzzy similarity among the corresponding k-dimensional feature vectors of each retrieved image and the k-dimensional fuzzy feature vectors of the query images is calculated; the retrieved images are fed back to a user in the order from high to low according to the fuzzy similarity; whether or not the image retrieval process is stopped is judged according to the satisfying degree of the user.

The invention discloses an image retrieval method and a system based on image segmentation and fuzzy pattern recognition, comprising the following steps: image segmentation: a query image is segmentedinto a plurality of image regions, and each retrieved image is also segmented into a plurality of image regions; feature extraction: extracting comprehensive feature vectors of each image region of the query image; extracting an integrated feature vector for each image region of each retrieved image; obtaining k nearest neighbor images of the query image; obtaining k nearest neighbor images of each retrieved image; for the k nearest neighbor images of the query image and the corresponding k nearest neighbor images of each retrieved image, the similarity of the k nearest neighbor images is calculated, and the similarity of the k nearest neighbor images is weighted and summed to obtain the final similarity between the query images and each retrieved image, and the final similarity is sortedin the order from large to small, and the retrieved images are outputted in the order from large to small.

The invention discloses a quantitative microscopic imaging testing and evaluating method of microbial aggregate property. The method is characterized by comprising the following steps: firstly, demarcating a standard relation curve of grey scale values and specific gravity values of a to-be-tested microbial aggregate sample; then obtaining an image of a to-be-tested slide made of a mixed solution containing to-be-tested microbial aggregates by using a digital microscope, calculating grey scale values and areas of all the to-be-tested microbial aggregates from resolutions, R values, G values and B values of all the microbial aggregates in the image, and calculating specific gravity values from the grey scale values as well as the standard relation curve of grey scale values and specific gravity values; and finally performing comprehensive evaluation on the areas and the specific gravity values of the to-be-tested organism aggregates by using a fuzzy pattern recognition method, to obtain a level characteristic value vector as an evaluation result of the to-be-tested organism aggregates. According to the high-efficiency and precise method, quantitative analysis and comprehensive evaluation on the microbial aggregates are achieved.

The invention discloses a human body-behind-wall multi-state target detection method based on fuzzy pattern recognition and a genetic algorithm. The human body-behind-wall multi-state target detection method comprises processing a received signal of a P410 radar, and extracting a characteristic parameter of the received signal; and forming a membership function set by means of the extracted characteristic parameter and multiple states behind a wall. A gaussian function is selected as a sub-membership function, the mean value and variance in the sub-membership function are optimized by means of the genetic algorithm, and the membership function set is constructed; a human body-behind-wall target prediction function is established in dependence on a fuzzy pattern recognition theory, and through calculation, which kind of state the detected data belongs to can be obviously recognized on the basis of a maximum membership degree principle. The human body-behind-wall multi-state target detection method based on the fuzzy pattern recognition and the genetic algorithm is mainly applied to the disaster rescue field and the anti-terrorism criminal investigation field in order to guarantee that a target under which a living body is buried is detected and rescued and the personal safety of a hostage is guaranteed when the hostage is seized in the anti-terrorism action.

The invention discloses an air passage fault diagnosis method based on model and fuzzy mode identification, wherein the air passage fault diagnosis method belongs to the field of engine fault test designing. Through performing mode algorithm matching on an unknown fault with a known fault database, a most possible fault type is found out. The air passage fault diagnosis method overcomes excessive dependence of an existing model-based method to an engine model and furthermore expands application range of a model-based air passage fault diagnosis method. On the condition of low calculation precision of the engine model, a model-based calculation result can be effectively corrected, thereby improving air passage fault diagnosis precision.

The invention relates to the fields of computer technology, automatic control technology and drug coating. In order to realize full automation of the process of controlling and detecting the cataplasm coating uniformity and thickness, improve the practicability and reliability of a system, lower the blindness in the production process and enhance the product quality, the invention adopts the technical scheme that an automatic cataplasm coating control system based on FPGA (field programmable gate array) fuzzy pattern recognition comprises a camera CCD (charge coupled device), a high capacity cache SDRAM (synchronous dynamic random access memory) which is taken as an external cache to store a digital video image signal and a programmable logic device FPGA, wherein the interior of the FPGA is provided with an SDRAM controller, a pattern recognition module comprises a division auxiliary functional module and is used for judging and recognizing a signal by applying the fuzzy pattern recognition principle and providing a control feedback signal, and the interior of the FPGA is also provided with a configuration initialization module used for initialization configuration on a video decoder. The invention is mainly applied to drug coating.

In soft measurement technology frame, the easy-to-measure reliable pressure difference fluctuation signal is adopted as auxiliary variable in detecting flow pattern of two phase flow as the main variable. The acquired pressure difference signal is wavelet analyzed, corresponding characteristic values are extracted, and the characteristic values are used in fuzzy identification of the flow pattern of two phase gas / liquid flow in horizontal pipe. The key soft measurement model is established by means of combining the wavelet analysis technology and fuzzy pattern identifying technology. Then, one flow pattern identifying system for two phase gas / liquid flow is established, and the system consists of test pipeline, pressure sampling hole, pressure sampling pipe, pressure difference sensor, amplifier, data acquiring card, microcomputer and printer. The detected pressure difference signal is processed in the system for pattern identification of two phase gas / liquid flow.

The invention discloses an energy efficiency multi-dimensional evaluation method, device and system. The method comprises the steps that firstly, a sample library of energy efficiency evaluation characteristic quantities based on energy consumption indexes and thermal comfort indexes is established, a typical characteristic set is obtained through clustering analysis, and on the basis, a fuzzy pattern recognition method is adopted to intelligently match an energy efficiency characteristic vector to be evaluated with the typical characteristic set. Through the method provided by the invention,the thermal comfort of the terminal building passengers can be evaluated in real time, meanwhile, the influence of factors such as climate and airport scale on the terminal building energy consumptionis comprehensively reflected, the multi-dimensional comprehensive objective evaluation on the terminal building energy efficiency is realized, and the terminal building system energy consumption is reduced while the comfort of the passengers is ensured.

The invention relates to the technical field of P2P flow detection, and particularly to a P2P flow detecting system based on combination of flow statistical characteristics and fuzzy pattern recognition. The detecting process of the P2P flow detecting system particularly comprises the following detecting steps: a first step, acquiring a data packet, performing statistical analysis on the acquired data message through the acquired data packet; a second step, performing characteristic selection by means of an FCBF algorithm; and a third step, setting a model determining rule according to a maximum subordination principle, and according to the extracted flow characteristic, determining input modes according to a maximum subordination principle classifying method in fuzzy recognition, thereby classifying the input modes. According to the P2P flow detecting system, a fuzzy recognition model in an intelligent system is combined with the flow statistical characteristics of network data; and the fuzzy recognition model and the flow statistical characteristics of the network flow are applied in P2P flow detection, thereby realizing high feasibility and good recognition effect.