91results about How to "Narrow the match" patented technology

The invention discloses a multiple application protocol identification method and an application protocol identification engine. According to the multiple application protocol identification method, characteristic conditions, decryption suite conditions, application protocol characteristics, application protocol hit action, various detection identification modules and corresponding rule description methods are provided on the basis of XM language; the application protocol identification engine comprises identification modules, such as a strong correlation identification module, a single packet identification module, a multi-packet identification module, an algorithm decryption and identification module and a channel extraction and identification module; a configuration module; and an application protocol characteristic matching unit, a full-flow table storage unit, a half-flow association table storage unit, and a rule base storage unit. By the use of the multiple application protocol identification method and the application protocol identification engine, flexible rule description configuration mechanisms and online, accurate and comprehensive identification of application protocols and encrypted traffic are achieved.

The embodiment of the application provides a robot repositioning and environment map construction method, a robot and a storage medium. According to the robot repositioning and environment map construction method, firstly, when performing repositioning on robot, the environmental image captured by a visual sensor is fast matched in a visual map to obtain a preliminary estimated pose by combining alaser sensor and the visual sensor, and the environmental data collected by the laser sensor on a grid map are locally matched to obtain accurate pose by combining the preliminary estimated pose. Since the matching range of the grid map is reduced by using the preliminary estimated pose, the calculation amount in the grid matching is reduced, which is beneficial to the improvement of the relocation efficiency.

Embodiments of the invention disclose a fingerprint unlocking method and a mobile terminal. The fingerprint unlocking method can comprise the steps of obtaining a fingerprint image input by a user; determining a target fingerprint group corresponding to the fingerprint image, wherein the target fingerprint group comprises a plurality of fingerprint templates; performing matching on the fingerprint templates and the fingerprint image in sequence according to a preset sequence; and when any fingerprint template in the fingerprint templates is successfully matched with the fingerprint image, unlocking the mobile terminal. With the adoption of the embodiments of the method and the mobile terminal, the target fingerprint group is firstly determined before fingerprint matching, so that the matching range can be reduced; and the matching is performed according to the preset sequence, so that the fingerprint matching frequency can be further reduced, the matching speed can be increased, and the fingerprint identification efficiency can be improved.

The invention belongs to the field of computer graphics and image processing, and specifically discloses a real-time multi-class and multi-target tracking method of a video. The tracking method comprises the following steps: s1: carrying out video frame preprocessing, such as superpixel segmentation; s2: on the basis of a superpixel block, designing a target detector, carrying out off-line training, and fully utilizing movement features so as to detect all moving targets in the video; s3: utilizing the trained detector to carry out target detection on a given video; s4: designing a target tracking model, and tracking the targets in the video; and s5: carrying out trajectory visualization. The tracking method has the following beneficial effects: 1: the complexity and the time consumption of an algorithm are greatly lowered since the video is subjected to the target detection on the basis of the superpixel block; and 2: all multiple classes of moving targets in the video can be detected, so that the tracking of all moving objects in the video can be realized under a situation of one camera, and therefore, hardware cost is greatly lowered.

The invention belongs to the technical field of power load identification, and particularly relates to a non-intrusive household appliance load identification method. Firstly, a household appliance load is regarded as an independent system, steady-state voltage and steady-state current data in the case of independent operation of each household appliance load are acquired, and according to steady-state current wave crest coefficients, loads are divided into a type of approximately linear loads and a type of nonlinear loads; the steady-state voltage then serves as input data, the steady-state current serves as output data, a system identification method is used for building models for the two types of loads, and a model library is formed; and finally, the steady-state voltage and the steady-state current of a to-be-identified load are acquired, a model library type is determined through pre-selection, and the load is identified according to a model matching principle. The actually-measured data can be used for effectively identifying the linear loads and the nonlinear loads, the algorithm is simple, the operational efficiency is quick, the method can be applied to a condition with increased loads in a home network, large-scale training is not needed when a new load is added to a power consumption network, and a model only needs to be built for the load, and the model is added to the corresponding model library.

The invention discloses an indoor WLAN fingerprint quick positioning method based on support vector regression. The method comprises the steps of 1 an off-line stage, wherein a feature fingerprint database of reference points is built, and a relation model of feature fingerprints and reference point locations is trained; and 2 an on-line stage, wherein coarse positioning is conducted through the relation model, and then precise positioning is conducted through a weighted K nearest neighbor algorithm. By adopting the indoor fingerprint quick positioning method based on support vector regression, the on-line fingerprint matching range can be narrowed, noise jamming is reduced by extracting a characteristic value of signal intensities, therefore, the positioning precision is improved, and the fingerprint matching speed is increased.

The invention relates to a binocular vision-based unmanned aerial vehicle aerial autonomous refueling fast docking navigation method and belongs to the machine vision and image processing field. The method includes the following steps that: a binocular camera system with an optical filter shoots a refueling taper sleeve with an optical marker lamp, so that a left image and a right image are obtained; gray processing, binarization and median-value filtering are performed on the left image and the right image sequentially, all communicated regions in the left image and the right image are searched and marked through adopting a regional growth method, and the centroids of each of the communicated regions are calculated and are adopted as feature points; improved haar wavelet transform is adopted to describe all the feature points, and the description vectors of all the feature points are obtained, the feature points of the left image and the right image are matched through adopting the minimum Euclidean distance method, so that the feature point pars of the left image and the right image are obtained; and the three-dimensional coordinates of the feature points are calculated by using a binocular vision principle, and the least squares method is adopted to perform spatial circle fitting on the three-dimensional coordinates, and the end surface circle curve of the refueling taper sleeve is obtained, and the center, normal vector and radius of the circle are obtained through calculation.

The invention discloses an auxiliary driving obstacle detection method based on binocular stereoscopic vision, and belongs to the technical field of environmental perception. The method mainly comprises the steps of synchronously collecting two frame images output by two CCD cameras respectively, wherein the two CCD cameras are arranged right ahead of a vehicle; carrying out distortion correction on the collected images; calling a stereo matching sub-process to carry out stereo matching on the two corrected images to obtain the corresponding relations of all pixel points; respectively calculating the projection matrixes of the two CCD cameras according to camera parameters; finding all corresponding pixel point pairs in the two corrected images; carrying out scene three-dimensional reconstruction on the images collected in real time according to the projection matrixes to obtain the space coordinate of an obstacle scene, screening obstacle areas, and outputting an obstacle position information image. The method achieves the function of automatically detecting road obstacles in the static or advancing process of the vehicle, and has the advantages of being capable of achieving the unattended operation and high in detection result precision.

The invention discloses an obstacle detection method and system. The obstacle detection method comprises the following steps: acquiring laser radar point cloud data, wherein the laser radar point cloud data is detected by a single-line laser radar installed on a train; clustering the laser radar point cloud data according to the sequence number of the radar points in combination with a set maximumsearch range to obtain a plurality of clustering sets; extracting feature parameters of a target corresponding to each clustering set of the current frame according to the clustering sets, wherein the characteristic parameters comprise the number of radar points, an average azimuth angle, an average distance, a centroid position, average reflection intensity and reflection intensity variance; matching all the targets of the current frame according to the feature parameters of the targets corresponding to each clustering set of the previous frame; and when the matching succeeds, determining the target of the current frame as an obstacle target. According to the invention, the obstacle detection accuracy can be improved.

The invention provides a three-dimensional positioning method of lightning based on stereoscopic vision, which is applicable to space reconstruction and three-dimensional positioning of lightning and relates to the fields of computer vision and aerospace and the like. The three-dimensional positioning method provided by the invention adopts a stereoscopic vision technology to carry out three-dimensional positioning on the lightning in the space and displays the development process of the lightning intuitively according to a two-dimensional image of the lightning so as to make further exploration and early warning for the lightning activity. The three-dimensional positioning method comprises the following steps of: (A) setting up a binocular stereoscopic-vision hardware system, and calibrating relevant parameters of a camera; (B) collecting and pre-processing lightning images; (C) segmenting and extracting a lightning skeleton line; (D) matching lightning sequence points of left and right images; (E) calculating a space three-dimensional coordinate of the lightning by utilizing a binocular stereoscopic-vision measurement model.

The invention discloses a hearing aid self-fitting method. The method includes the steps that firstly, a knowledge base is established, a historical patient most similar to a current present is found out, and the optimal hearing aid algorithm parameters of the historical patient are obtained; the current patient conducts satisfaction evaluation on the sound quality, the hearing aid algorithm parameters are updated in an iteration mode according to evaluation indexes, and interaction with the knowledge base is conducted; test sounds are generated for the current patient according to the updated hearing aid algorithm parameters until the current patient feels satisfied. The knowledge base is introduced into the hearing aid self-fitting method, the matching range of the hearing aid algorithm parameters is greatly narrowed, and the fitting efficiency of a hearing aid is obviously improved. The expected value of the hearing aid of the current patient is met to the maximum degree due to the fact that the hearing aid algorithm parameters are updated in the iteration mode according to the evaluation indexes, and the accuracy of the hearing aid algorithm parameters is improved.

The invention provides an electrocardiogram classifying method based on fuzzy inference and weighted similarity measurement and relates to an electrocardiogram classification method. The electrocardiogram classifying method aims to solve the problem that by the adoption of an existing fuzzy inference classification method, due to the fact that an electrocardiogram knowledge base cannot be established, the influence of electrocardiogram knowledge and different combinations of different wave band forms on classification is omitted, and as a result, the error rate of classification is high and to solve the problem that by the adoption of the existing fuzzy inference classification method, due to the facts that attribute concepts are not screened and comparison of membership degrees of the attribute concepts is directly used for classification, the error rate of classification is high. The electrocardiogram classifying method comprises the following steps that, firstly, electrocardiosignals are preprocessed; secondly, feature parameter extraction is conducted on all wave bands; thirdly, a classification feature attribute value vector Yi=[yi1, yi2, yi3, yi4 and yi5] and a to-be-detected feature attribute value vector X=[x1, x2, x4, x4 and x5] are established, and an ECG body ecg.owl is established according to electrocardiogram knowledge; fuzzy concept lattices are established; fuzzy attributes are converted into specific membership degree values, and effective screening is conducted on the specific membership degree values; final classification is conducted through a weighted classification method. The electrocardiogram classifying method is suitable for electrocardiosignal classification.

The invention discloses a message classifying method based on rule information entropy. The message classifying method based on the rule information entropy includes: calculating regular stray attribute subsets on an attribute domain with regular classified messages through preprocessing on classification rules, using a regular attribute weight vector to calculate weighting distance afterwards, and then analyzing subspace straying influence factors of regular weighting neighbor domains, and generating frequent matching subsets for matching messages by comparing the subspace straying influence factors with straying factor threshold value. According to the message classifying method based on the rule information entropy, subsequent message matching had the regular frequent matching subsets with no-straying attributes by being compared with prior message matching, matching efficiency is improved, and simultaneously time expenditure for generating the frequent matching subsets through the preprocessing of the classification rules is smaller than time expenditure for matching a whole data stream, and transmission performance of an algorithm used in the message classifying method is higher than transmission performance of the algorithm before the preprocessing.

The invention belongs to the field of navigation and positioning technologies and discloses a high-precision outdoor positioning method integrating a Beidou satellite system and image features. The method includes the following steps that: reference images are collected; the reference images are processed; and a database is established according to the processed reference images; the Beidou satellite system is adopted to obtain a preliminary positioning result; and an image to be positioned is obtained, and the similarity of the image to be positioned and the reference images in the database is calculated on the basis of the preliminary positioning result, a reference image matched with the image to be positioned is obtained through search, and a visual aided positioning result is obtained. With the method of the invention adopted, the problem of low positioning accuracy caused by weak satellite signals under a poor visibility condition in the prior art can be solved.

The present application discloses a voice control method for an intelligent terminal and the intelligent terminal, wherein the voice control method for the intelligent terminal includes: determining the application scene where the cursor associated with the intelligent terminal is currently located, the The application scene has a corresponding voice command set; acquiring voice information; determining a voice command matching the voice information in the voice command set corresponding to the application scene; and executing the voice command. According to the technical solution provided by the embodiment of the present application, the range of voice command matching is narrowed by determining the application scenario, which can solve the problems of unprepared voice command matching and occupying a large amount of resources. Furthermore, according to some embodiments of the present application, by introducing cursor resources, the problem of inaccurate positioning of voice in position control can be solved, and the effect of accurate positioning of moving objects can be obtained.

The invention relates to the field of wireless communication, in particular to a method for rapidly and accurately positioning a mobile terminal. The method is characterized by comprising the following steps of: analyzing a geographical scene and a wireless scene of each cell in a network; sequentially establishing a covering and screening feature library of each cell; obtaining a primary positioning result; obtaining a TA (Terminal Adapter) value positioning result; determining a screening range of M meters from a primary positioning point, wherein the value of M is determined by the TA value positioning result; acquiring predicted data in a region which needs to be positioned; acquiring a fingerprint judgment feature value of each related cell; obtaining a matching value of a geographical position; recording fingerprint features which can be taken for reference; and obtaining a final positioning result. The method has the advantages that a report point is subjected to primary positioning by an improved CHAN method and the matching range of a fingerprint feature library is shortened; the rapid positioning is realized and the positioning cost is low; and the feature library is updated in time according to different metric drive tests and the positioning result is ensured to have high positioning accuracy.

The invention discloses a wireless access point connection method, a wireless access point connection device, a mobile terminal and a storage medium. The method comprises the steps of acquiring current location information of a user; judging whether the current location information is positioned in a coverage area of a target wireless access point; if yes, judging whether the user enters the coverage area of the target wireless access point for the first time; if no, acquiring at least one wireless access point to be scanned and positioned in the coverage area from a pre-built relation table of the coverage areas and optimal wireless access points, wherein the relation table comprises the plurality of coverage areas and the optimal wireless access point included in each coverage area; finding the wireless access point with the highest priority from the at least one wireless access point to be scanned; and controlling the mobile terminal to be connected with the wireless access point with the highest priority. According to the method, the scan matching time of the wireless AP can be saved, the purpose of quickly matching the optimal WiFi is achieved, and the user experience is enhanced.

The invention provides a rapid character string matching method based on a suffix array. The method includes two stages, the first stage includes that the appearing position of a pattern string in a text string is limited in a possible interval of the suffix array taking a first character of the pattern string as a beginning character by binary search, and the second stage includes that search conditions are further limited on the interval, suffixes with the length smaller than that of the pattern string and with last characters different from those of the pattern string are excluded, the comparison frequency of the characters is decreased, character string matching range is narrowed, and the appearing position of the pattern string in the text string is rapidly acquired.

A point target movement velocity detection method based on multiple linear moveout scanning, extending and sampling comprises the following steps: (1) establishing a multiple linear moveout scanning detection device; (2) using each linear detector to detect point targets in an extending and sampling manner; (3) processing Nt groups of image data acquired by each linear detector to obtain sub-pixel images; (4) performing non-uniformity correction and sub-pixel match on two sub-pixel images processed by the two adjacent processed linear detectors, then performing difference computation to complete the background subtraction; (5) performing threshold filtering on difference images processed by the two adjacent processed linear detectors, extracting positive and negative point pairs in the difference images by adopting the neighborhood constraint criterion, so as to complete the target detection and extraction of moving points in once scanning process; (6) marking regions of paired positive and negative points extracted in the step (5) respectively, calculating movement velocity and movement directions of the targets according to the position relation of the marked regions of the positive and negative points; (7) averaging the movement velocity and the movement directions obtained in the step (6) to obtain the velocity and the direction of a detected target.

The invention provides a method for enhancing a moving target through multi-linear-array time difference scanning expansion sampling. The method includes the steps that (1) a multi-linear-array time difference scanning expansion sampling detection device is constructed, and Nt detection arrays in each linear array detector in the multi-linear-array time difference scanning expansion sampling detection device are made to conduct imaging at the same time to obtain Nt groups of image data; (2) the Nt groups of image data of each linear array detector are processed to form a frame detection image; (3) after the preset sampling frequency is completed in the scanning direction, the frame detection images obtained by the linear array detectors correspondingly are spliced according to time to obtain two sub-pixel images; (4) non-uniformity correction is conducted on the two sub-pixel images respectively; (5) the sub-pixel image corresponding to the linear array detector which conducts imaging firstly is used as a standard, the sub-pixel image corresponding to the linear array detector which conducts imaging later is moved forwards in the scanning direction; (6) after the two sub-pixel images are matched, difference calculation is conducted to obtain a residual image; (7) enhancement processing is conducted on the residual image after filtering is conducted on the residual image.

The invention relates to a traffic sign detection method and apparatus, a computer device and a storage medium. The method comprises the steps of obtaining a to-be-detected road image containing a target traffic sign; obtaining a matched reference road image containing the target traffic sign from the historical road image set based on the position information of the to-be-detected road image; performing feature extraction on the to-be-detected road image and the reference road image to obtain a target feature set corresponding to the to-be-detected road image and a reference feature set corresponding to the reference road image; calculating the matching degree of the target feature set and the reference feature set, and determining the consistency detection result of the target traffic sign in the to-be-detected road image and the reference road image based on the matching degree; and determining a target detection state of the target traffic sign in the to-be-detected road image based on the consistency detection result. By adopting the method, the traffic sign detection accuracy and efficiency can be improved, and the to-be-detected road image with the changed target detection state of the target traffic sign can be used for updating a map.

The invention discloses an SLAM method based on an RGB-D image. The method comprises the following steps: 1, acquiring the RGBD image; 2, extracting and matching feature points of the RGB-D image; 3,calculating a rotation matrix and a translation matrix; 4, acquiring a key frame in the RGBD image; 5, performing closed-loop judgment on the key frame; 6, carrying out global optimization on the RGB-D image; and 7, constructing a grid map. The method is simple in step, reasonable in design and convenient to implement, the violent matching problem is solved, errors, caused by mass difference of matching points, of the pose of the depth camera are reduced, the precision and real-time performance of SLAM are improved, and therefore the requirement for real-time SLAM of a robot can be met, and extremely high precision is achieved.

The invention relates to a speech enquiry method of Chinese large vocabulary in an embedded type environment, belonging to the speech enquiry field in the information technology. The method comprises the following steps that: a tone-free syllabary used in a speech recognition engine is generated; a syllable similarity table is generated; realtime monitoring of an input speech signal is carried out, and continuous vocabulary speech signals through which a user expresses vocabulary content are picked up; the continuous vocabulary speech signals are divided into a plurality of monosyllabic vocabulary speech; all vocabulary with the length more than or equal to the syllable number of the vocabulary speech are picked up from a Chinese large vocabulary font library to form an equilong font library; the matching degrees of all vocabulary in the vocabulary speech and the equilong font library are calculated; and the calculated matching degrees are sized down to choose a final enquiry result. On the premise of ensuring higher accuracy, the speech enquiry method ensures that an enquiry function breaks the limitation on the font library number of a speech recognition engine and can complete a Chinese large vocabulary speech enquiry function which the speech recognition engine does not have.

The invention belongs to the field of image processing and computer vision, and relates to a multispectral stereo camera dynamic registration algorithm based on Fourier transform registration. The method comprises the following steps: 1, performing distortion removal and binocular correction on an original image according to respective internal parameters and original external parameters of an infrared camera and a visible light camera; 2, performing Fourier transform registration; 3, respectively extracting feature points from the registered infrared image and visible light image; 4, matchingthe feature points extracted in the previous step; 5, calculating the feature points of the registered infrared image feature points corresponding to the infrared original image according to the results of the step 2 and the step 3; 6, judging a feature point coverage area; 7, correcting a calibration result. According to the invention, matching is carried out according to the result after registration, the position relationship between the visible light image and the infrared image can be effectively utilized, so combined self-calibration is carried out on the infrared camera and the visiblelight camera more effectively, the operation is simple and convenient, and the result is accurate.

The application provides a method used for identifying an individual. The method comprises: determining that candidates which are in a database and have matching degrees which relate to a first biological feature and are higher than a first critical threshold value A2 are a first candidate collection; determining that the candidates which are in the first candidate collection and have matching degrees which relate to a second biological feature and are higher than a second critical threshold value B2 are a second candidate collection; and identifying the individual on the basis of the second candidate collection. According to the method, the number of matching in two matching processes is effectively reduced while the recognition accuracy is ensured, and the recognition efficiency is improved; and at the same time, the method has the characteristics of the high pass rate and the high precision which fingerprint identification has and the anti-counterfeiting ability which finger blood-vessel identification has, and effectively solves the defect of homogeneous biological information identification in the prior art.

The invention discloses a two-level-subspace partitioning method and device in the technical scheme. The two-level-subspace partitioning method and device are used for achieving the aims that a secondary matching method based on model subspaces is adopted; the processing process includes the two steps that the subspaces are positioned through a coarse screen, and a range of a to-be-detected audiofrequency module space is determined; fine matching, wherein a target model is hit inside the subspaces through a high-accuracy traditional algorithm, and four processes of feature extraction, clustering calculation, subspace partitioning and central point calculation are conducted. The two-level-subspace partitioning method and device has the advantages that achievement is simple, less time consumption is wasted, and for model features with the large quantity, the real-time performance of the device enough meets the business requirements, values of the cluster quantity can be adjusted at random, the subspaces can be effectively partitioned, the values can be evenly partitioned into different subspaces, and meanwhile the model matching range is narrowed; the quantity of models can be effectively reduced through single voice features, and the efficiency is improved.

The invention discloses a method of matching and invoking of an antenna thermal deformation compensation database based on orientation adjustment. The method comprises the following steps: based on the acquired temperature data and antenna orientation adjustment amount, obtaining a thermal deformation compensation database; a sensor acquiring temperature data, calculating a root mean square error RMS value of a temperature data acquiring group; the RMS of the temperature data acquiring group matching the RMS of a storage temperature data group in the database, calculating a similarity matching area and performing matching, calculating a critical matching area and performing matching, comparing similarity matching rates of data groups that have successful matching, finding the only matching data group, obtaining antenna orientation adjustment amount of a plane of reflection, determining whether the orientation adjustment amount is used for orientation adjustment, adding the newly acquired temperature distribution to the database. According to the invention, based on the real-time acquired temperature data, the method performs matching and invoking on the database through the RMS value for matching data, similarity area, and critical area, reduces matching scope gradually, thus completing orientation adjustment compensation of the plane of reflection and guaranteeing compensation accuracy and effect.

The invention belongs to the field of image processing and computer vision, and discloses a visible light and infrared camera dynamic pose calculation method based on a bidirectional homography matrix. The method comprises the steps that a transformation matrix between an infrared image and a visible light image is calculated by extracting and matching linear features on the infrared image and thevisible light image; Harris angular points are determined on the infrared image, initial correspondence of the detected Harris angular points on the visible light image is calculated according to thetransformation matrix, and the Harris angular points are detected around the initial correspondence to serve as final correspondence; and the same operation is performed on the visible light image. The matching number and matching efficiency of the infrared feature points and the visible light feature points are improved, so that the infrared camera and the visible light camera are subjected to joint self-calibration more effectively.

Provided is a point object detection method based on multiple-linear time difference scanning and expansion sampling. The point object detection method comprises the steps of (1) constructing a double-linear-detector imaging device; (2) performing point object detection through each linear detector in an expansion sampling method; (3) respectively processing Nt groups of image data collected by each the linear detectors to obtain sub-pixel images, performing heterogeneity correction on two sub-pixels processed by the two linear detectors, performing sub pixel matching, and performing calculus of differences to complete background removal; (4) performing threshold filtering on residual images obtained through difference in the step (3), extracting positive and negative point pairs in the residual images through a neighborhood constraint criterion to complete detection and extraction of a moving point object in a scanning process, and forming neighborhood constraint criterion (as specified in the specification) according to imaging time differences of different linear detectors and the target movement speed, wherein delta D represents a distance between a positive area and a negative area, vmin represents the minimum movement speed of the detected object, vmax represents the maximum movement speed of the detected object, and GDS represents ground sampling distances of the linear detectors.

The invention discloses a rule-based extraction method of doctor-patient dispute case keywords. The method includes: (1) establishing generation templates of extraction rules of the doctor-patient dispute keywords according to keyword categories; (2) modularizing a doctor-patient dispute case document according to mark word containing quantity; (3) defining a keyword set for each module of the doctor-patient dispute case document, and extracting keywords in the corresponding keyword set from each module; (4) writing each keyword, which is extracted in the step (3), as a pseudo-expression according to the generation templates defined in step (1), and automatically mapping the same to a machine-recognizable expression; (5) discriminating whether each extracted keyword is correct, if yes, storing the machine-recognizable expression corresponding to the keyword and document and paragraph attributes to which the same belongs, and otherwise, storing the same into a set of to-be-perfected rules; and (6) adjusting and perfecting the extraction rules according to the set of the to-be-perfected rules. The method increases an accuracy rate and a recall rate of keyword extraction.