243results about How to "The solution accuracy is not high" patented technology

The invention relates to a satellite remote sensing image large-area seamless orthographic image manufacturing method. The method comprises the following steps: 1) constructing a remote sensing imageimaging geometric model; 2) reading a to-be-matched remote sensing image and DEM data, and matching the DEM data with the to-be-registered remote sensing image data; 3) carrying out remote sensing image ortho-rectification to obtain a corresponding DOM product; 4) according to the position relation between the orthographic images obtained in the step 3), determining an overlapping region between the images, performing matching in the overlapping region to obtain homonymous points, establishing an error equation based on a TPS model for each group of homonymous points, and performing iterativesolution to obtain a correction amount of geometric deviation between the images; And 5) carrying out dodging and inlaying on all the corrected ortho-images to obtain large-area ortho-image map products, and introducing a thin plate spline model in the ortho-image adjustment process to solve the problem of low precision of a traditional geometric correction model caused by a small overlapping range between the images.

The invention provides an identification method of characters of a licence plate, mainly solving the problems of low identification speed and complex realization in the prior art. The identification process of the identification method comprises the following step of: individually taking out the pretreated characters of the licence plate, uniformizing, and scanning in vertical and horizontal directions; recording horizontal and vertical characteristic vectors, and respectively calculating similarity with characteristic vectors already well stored in a standard character library; finding a certain character of the standard character library, which has maximum similarity, through similarity magnitude comparison, wherein the character is a first-time identification result of characters to be identified; dividing partial similar characters in the first-time identification result into an up and down part and a left and right part, and respectively scanning the characteristic vectors of the similar characters; and then carrying out similarity calculation with a characteristic vector library of the characters, and finding a certain character of the standard character library, which has the maximum similarity, as a final identification result. The invention has high identification speed and identification accuracy rate and can be used for the real-time detection of the licence plate of a high-speed running vehicle.

The invention discloses a question and answer method and device based on a knowledge graph, electronic equipment and a storage medium. The question and answer method based on knowledge graph comprisesthe steps of obtaining a question sentence in response to a triggered user input request; performing entity linking on the entities in the questions obtained through named entity identification in aconstructed knowledge graph to obtain candidate entities; determining a search subgraph containing the candidate entities in the knowledge graph, and generating a plurality of candidate paths pointingto candidate answers by the candidate entities according to the search subgraph; and according to the context information of the question sentence, screening from the plurality of candidate paths toobtain a target path, and pushing a candidate answer pointed by the candidate entity in the target path as an answer of the question sentence to the user. By adopting the knowledge graph-based question and answer method and device, the electronic equipment and the storage medium provided by the invention, the problem of low accuracy of knowledge graph-based question and answer in the prior art issolved.

The present invention provide a battery SOC value evaluation method and device, wherein the method comprises the steps of detecting whether a battery is located in a charging state or a discharging state; on the condition that the battery is located in the charging state or the discharging state, determining the SOC value of the battery via an ampere-hour integral method; when the battery is not in the charging state and the discharging state, and on the condition that a charging loop or a discharging loop of the battery is disconnected, determining the SOC value according to a correspondence relation of the battery open circuit voltage of the battery and the SOC value. According to the present invention, the problem that the accuracy of the SOC value valuated by the battery SOC evaluation method is not high is solved, and the SOC evaluation precision is improved.

The invention discloses an indoor map construction method for a mobile robot integrated with multiple sensors. The indoor map construction method comprises the following steps of respectively acquiring distance information of the mobile robot and an anchor point, gesture information of the mobile robot and environmental information by UWB, a speedometer and laser radar; building a top point-constraint map according to the distance information, the gesture information and the environmental information; optimizing the top point-constraint map by a map optimization algorithm, and acquiring optimized track data of the mobile robot; and building a raster map according to the optimized track data of the mobile robot and the environmental information. The speedometer is used for providing accurate gesture change within a short time of the robot, long-time accurate gesture change can be provided by integrating UWB positioning information, map construction is performed by combining the data ofthe laser radar, and the problem of poor accuracy during constructing indoor complicated environmental map by the laser radar is solved.

The invention provides a vehicle sensor concurrent monitoring method facing road conditions / the method is finished through coordination of a vehicle node which carries a sensor and a road side base station. The vehicle node sends data collected by the sensor and observation data obtained by fusion of local data to a local road side base station, the road side base station receives and stores the observation data and decides whether concurrent monitoring is required to be started, the vehicle sensor data based concurrent event monitoring process can be started if conditions are met, vehicle sensor data of position correlation, time correlation and driving behavior correlation are mined, after the data correlation is obtained, a voting weighting method is used for processing, the road side base station judges whether a road condition event occur in a local sub-cell according to processing results, and if the event can be determined, the road side base station issues the event.

The invention discloses a method and a device for speech recognition. The method comprises the steps of acquiring speech recognition information of user current speech, and acquiring auxiliary recognition information of the speech recognition information based on a user current state corresponding to the user current speech; and determining a final recognition result of the user current speech according to the speech recognition information and the auxiliary recognition information. Through the method and the device disclosed by the invention, a problem in related technologies that the accuracy of speech recognition is caused to be low by acquiring the speech content of a user only through the sound of the user is solved, and thus the accuracy of speech recognition is improved.

The invention relates to an indoor positioning method and system, wherein the indoor positioning method comprises the steps of primary distance measurement, and precise positioning after positioning range and distance correction, wherein the wall body combination with the smallest signal loss estimation is used as parameter correction distance measurement information so as to obtain the precise indoor positioning coordinate through calculation. By using the indoor positioning method and system provided by the invention, the fixed and unique building structure information in indoor environment is sufficiently used; the signal obtained through measurement is subjected to fitting according to different signal attenuation of different building materials of building structures and wall bodies; the optimum wall body combination with the smallest signal loss estimation error corresponding to each position region is found; the time delay information of wireless signal spreading interference by the wall body is obtained; the distance measurement value is thus corrected; the problem of low precision of the existing positioning algorithm due to blocking between various wall bodies in the indoor environment is solved. The method has the advantages that the complexity degree is proper, and the more precise positioning result in the indoor complicated environment can be obtained.

The invention discloses an infant emotion detection method and device based on brain waves. According to the infant emotion detection method, brain wave signals are acquired at a plurality of positions of the brain of an infant; the acquired brain wave signals are processed, and a first energy ratio and a second energy ratio are calculated according to energy of the brain wave signals acquired at all the positions and energy of signals in preset frequency bands; the first energy ratio and the second energy ratio are matched in a prestored emotion sample library to obtain a final sample, so that corresponding infant emotion information is obtained. The infant emotion detection method and device based on brain waves realize real-time judgment of emotions of the infant through the brain waves, enable emotion judgment of the infant to be more accurate, and bring much convenience to a user.

The invention discloses a method for predicting the remaining service life of a rolling bearing integrated with the KELM (Kernel Extreme Learning Machine), and belongs to the technical field of the bearing service life prediction. The method is used to solve the problem that the prediction of the remaining service life of the rolling bearing has difficulty in prediction and low prediction accuracy. The method firstly extracts features of a vibration signal based on the variational mode decomposition, introduces a new similarity dimension reduction method for features dimension reduction, and further extracts the features-CEF (Cyclic Enhancement Features) with strong monotonicity, similarity, and stability. Multiple KELM models are constructed through that the CEF extracted by the multiplebearings is used as the input of the KELM, the ratio of the current service life to the whole life, p, that is, the life percentage is used as the output. A prediction model integrated with KELM is constructed by combining the random forest to obtain a current prediction result p value. The CEF of the test bearing is input into the prediction model, the current p value is predicted, and the secondorder exponential smoothing method is used for fitting to predict the RUL of the bearing. The experimental verification shows that the proposed prediction method has higher prediction accuracy than other literatures.

The invention provides a correlation sampling based laser multi-pulse ranging system, relating to a laser ranging system. The correlation sampling based laser multi-pulse ranging system is provided with a clock control circuit module, a laser pulse transmitting module, a laser pulse receiving module and a time information processing module, wherein the clock control circuit module is provided with two clock signal output ends and is provided with a local oscillator, a first frequency divider, a second frequency divider, a first mono-stability trigger and a second mono-stability trigger; the laser pulse transmitting module is provided with a pulse driving circuit, a semiconductor laser, an opto-isolator and a collimating lens; the laser pulse receiving module is provided with an avalanche photodiode, a pre-filtering amplifying circuit, an opto-isolator and a focusing lens; and the time information processing module is provided with a first sample holding circuit and a second sample holding circuit. A sampling signal is subjected to frequency doubling, the sampling time of transmitting pulse and receiving pulse is reduced, the ranging speed is increased, the measurement precision is improved, and the correlation sampling based laser multi-pulse ranging system can solve the problem of low precision in a short-range laser ranging mode.

The invention provides a method for judging aging of insulator room temperature vulcanization (RTV) coatings. The method includes: taking samples of the insulator RTV coatings to obtain a test piece; performing space charge test on the test piece in a pulsed electro-acoustic (PEA) method to obtain a space charge density distribution curve of the test piece; calculating the total space charge amount of the test piece according to the space charge density distribution curve of the test piece; and judging the aging degree of the RTV coatings according to the total space charge amount of the test piece. By adopting the technical scheme, evaluation of the aging state of the insulator RTV coatings is carried out in view of micro electric characteristics, and the aging degree is quantized with the total space charge amount as the characteristic quantity, so that the method provides data basis for evaluation of antipollution performance of insulators of a power system, provides basis for further making time and schemes of insulation re-coated RTV materials, and solves the problem that the method in the prior art for judging aging of insulator RTV coatings is not high in accuracy.

The invention discloses a schizophrenia sMRI image classification method based on a multi-input convolutional neural network. The method comprises the steps of collecting structural magnetic resonanceimaging image data of schizophrenia patient and a normal person; preprocessing sMRI image data to obtain grey matter density images which are not subjected to smoothing processing and spatial smoothing respectively, and constructing an original data set; and sending the original data set into a multi-input convolutional neural network model for training, finally inputting sMRI image data into themulti-input convolutional neural network model, and finally outputting a classification result. According to the invention, the images which are not subjected to the spatial smoothing processing andimages which are subjected to spatial smoothing are input into a multi-input convolutional neural network, high-frequency information in an sMRI image lost in spatial smoothing can be compensated, andmeanwhile, the generalization ability of the convolutional neural network can be enhanced due to the noise contained in the images which are not subjected to smoothing processing, so that the problemof low accuracy in schizophrenia sMRI image classification can be effectively solved.

The invention requests to protect a multi-image regularization depth hash multi-modal medical image retrieval method. The method specifically comprises the following steps of: simultaneously extracting features of a multi-modal medical image group through a multi-channel depth model; Correspondingly constructing a plurality of graph regularization matrixes according to the characteristics of the multi-modal medical image group; fusing Multiple graph regularization matrixes, and obtaining Hash codes of the multi-mode medical image set through modal self-adaptive restricted Boltzmann machine learning; solving The distance between a single modal data hash code and a multi-modal medical image group hash code through Hamming distance measurement, carrying out sorting according to an ascending order, and selecting and returning n groups of multi-modal medical images with the minimum distance to a user, so that multi-modal medical image retrieval is realized. According to the method, a doctorcan be helped to quickly find data of other multiple modes through data of a certain mode in multi-mode medical images such as ultrasonic images, dispute end texts and nuclear magnetic resonance images, medical diagnosis of the doctor is facilitated, the workload of the doctor is reduced, and the working efficiency is improved.

The invention relates to a man-machine interaction control method of a robot system in an unknown environment. A referent track is improved by reference self adaption of a robot mechanical arm in man-machine interaction, so that the man-machine interaction performance needed by a user can be realized by the robot even in an unknown environment. On the basis of the reference self adaption based on a track parameter and repeated learning, minimization is realized; and the track tracing and force minimization are described by using a minimized cost function, wherein the cost function has been parameterized and the track parameter has been minimized. And a robot established based on a target impedance model is controlled. The provided method being a repeated learning method can be applied to remote operation of a robot in an unknown environment and has advantages of good man-machine interaction, good real-time performance and high accuracy.

The invention discloses a service handling method and device based on face recognition, equipment and a medium, and the method comprises the steps: obtaining a service operation request, and obtaininga video stream recorded in real time if the service type is a specific type, wherein the video stream to comprise a video frame image; performing face detection on the video frame image to obtain a current face image; querying a user database based on the user identifier to obtain a query result; if the query result is not empty, obtaining a target comparison image; calling a face recognition interface to process the current face image and the target comparison image to obtain a first score value; and if the first score value is smaller than the preset score value, inputting the current faceimage, the target comparison image and the third-party reticulate pattern image into a face recognition interface for processing to obtain a service operation result, thereby solving the problem of low accuracy of face recognition identity verification only according to the current face image and the third-party reticulate pattern image at present.

The invention discloses a ginseng slicing device with high safety. The device comprises a support, a machine case, a worktable, a spring clamping device, a sliding mechanism, a guide column, a top plate, a movable plate, an air cylinder, a connection rod device, a cutter, a gas conveying pipe, gas tank, a gas pump, a control panel, a power line, a pushing mechanism, a sliding mechanism, a supporting bottom frame and a collecting device; the periphery of the machine case is arranged in the middle of the inner side of the support through screws, the periphery of the worktable is arranged on thetop end of the support through screws, the pushing mechanism is arranged, the control panel starts a first electric push rod to positively rotate, after work, a left end push rod carries out leftwardpushing, a sliding plate slides on the upper ends of a first sliding rail and a second sliding rail through a first sliding block on the bottom, ginseng is driven to move leftwards, through mechanicalwork, the ginseng is fed into the bottom of a cutter, and the problems that according to an existing ginseng slicing device, the ginseng needs to be manually pushed to the bottom of the cutter to besliced, the human hand is prone to being hurt, and safety is poor are solved.

The invention discloses a comprehensive bituminous-mixture low-temperature-performance detection device, and relates to a comprehensive detection device. A low-temperature shrinkage test, a low-temperature stress relaxation test and a freezing breaking test need to be carried out through different devices, the problems that as the universality of the devices is poor, test cost is high, and meanwhile the accuracy of the detection result of a bituminous-mixture testing piece is low exist. The comprehensive bituminous-mixture low-temperature-performance detection device comprises an environmental box, a counterforce frame, a displacement measuring system, a data collecting control system, a micro-data collecting system and a heating system. The environmental box is arranged in the counterforce frame, a digital speckle camera faces to a visible window of the environmental box, a bituminous-mixture testing piece is arranged in the environmental box, one end of the bituminous-mixture testing piece is connected with a motor sequentially through a dowel steel and a transmission device, the other end of the bituminous-mixture testing piece is connected with a stress sensor through the other dowel steel, and the displacement measuring system is arranged outside the environmental box and connected with the counterforce frame. The comprehensive bituminous-mixture low-temperature-performance detection device is used for detecting the low-temperature performance of bitumen.

The invention discloses and specifically relates to a washing method and a washing machine automatically adding a washing agent; the washing machine comprises a water inlet valve, a washing agent automatic input system and a washing tub; a water inlet path with a venturi tube is arranged between the water inlet valve and the washing tub; when water flows into the washing machine, a negative pressure zone can be formed in the venturi tube throat under the venturi effect principle; the negative pressure zone is connected in a constant volume valve so as to form negative pressure in the constant volume valve, thus drawing the washing agent in a washing machine distributor connected with the constant volume valve to a constant volume unit; then water enters the constant volume valve through a negative pressure channel of the venturi tube, and flushes the washing agent from the constant volume valve to the washing tub. The washing method and machine can effectively solve the problems that an existing washing agent automatic adding device is complex in structure, poor in automatic input precision, and long in water inflow time.

The invention discloses a method for calibration of an optical centering device fixed to a base. The method comprises the following steps of 1, horizontally putting an apparatus needing to be detected on a stand-by base, and keeping a state that a base of the apparatus is hung at an outer side of a table edge and faces vertically to a wall surface, 2, observing an eye lens of an optical centering device of the apparatus needing to be detected so that a wall position corresponding to a crosshair is found and pasting a crosshair scoring board to the wall position, and 3, lightly turning the base of the apparatus for 180 degrees, and observing a relative position relationship of two crosshair cross points, wherein if the reticle cross point of the apparatus is in a circle drawn on the crosshair scoring board, the apparatus is qualified and if the cross point is outside the circle, a centering error exists and a correction process must be carried out. The method is fast and convenient and greatly improves a correction precision.

The present invention discloses a dynamic interval-based face age estimation method. The method comprises the steps of extracting the features of a face; finding a central face at each age based on the clustering algorithm; according to the similarity degree between a to-be-estimated face and the central face, selecting top K ages corresponding to a most similar face as prediction ages and defining a correlation function between the confidence interval and the confidence coefficient; according to the correlation function between the confidence interval and the confidence coefficient, estimating the bar graph, the normal distribution curve, the normal distribution expected value and the standard deviation of a face age according to the square estimation method by utilizing the top K ages of all test images at each age; calculating different confidence coefficients and the corresponding confidence intervals thereof according to the cumulative distribution function of the normal distribution. According to the technical scheme of the invention, the problem in the prior art that the traditional single-age estimation method is not high in accuracy can be effectively solved.

The invention discloses a method and a device for measuring the position and temperature of a combustion zone in a sinter bed. The method comprises the following steps of: determining the position of the combustion zone by tracking the highest temperature of the combustion zone; and determining the instantaneous speed at the moment and the average speed of the whole process by recording the actual temperature, displacement and time, and controlling the sintering process by taking the instantaneous speed and the average speed as production control guidance. In order to implement the measuring method, the invention discloses the measuring device matched with the measuring method, wherein sensors arranged inside a sintering cup can be used for measuring the real-time temperature and displacement parameters and transmitting the real-time temperature and displacement parameters back to a computer, and the computer is used for doing corresponding calculations and sending control signals.

The invention discloses a space filter value-based logistic regressive landslide hazard analysis method. Aiming at analytical investigations of landslide hazards, a space filter value thought is imported into a common logistic regression model so as to design a regressive landslide analysis algorithm which comprises the steps of non-landslide site selection, disaster-inducing factor obtaining andgrading, adjacent matrix construction, feature value and feature vector calculation, stepwise regression feature vector selection and regressive modeling. According to the method, the problem that themodel precision is not high as logistic regression models are influenced by spatial autocorrelation between variables can be solved. By adding filter value operators which are constructed by utilizing selected feature vectors into logistic regression models, autocorrelation influences of residuals can be effectively filtered, the fitting goodness and prediction correctness of the regression models can be improved and correct simulation and prediction for landslide hazards can be realized.

The invention discloses a rapid prototyping system and method for a high-precision ester cured water glass sand mold / core, and belongs to the technical field of sand mold / core prototyping. The method comprises the following steps: firstly establishing a CAD geometry solid model of the sand mold / core, performing layered discretization, generating a moving trajectory file of each discrete level dripping device and guiding the moving trajectory file of each discrete level dripping device into a controller, and recording the number n of discrete layers; evenly mixing the mold / sand core with water glass by using a mixing device; flatly spreading the mold / core sand wrapped by the water glass on a lifting working platform by a spreading device; and dripping composite organic ester according to the extracted moving trajectory of a first discrete level by the dripping device, after the first discrete level is completed, enabling the lifting working platform to move down, sequentially completing the dripping working procedures of the follow-up discrete levels, curing for 1h to 8h, and taking the mold core out. The rapid prototyping system and method provided by the invention have the advantages of being capable of shortening the manufacturing cycle of the sand mold / core, improving the prototyping precision and performance, increasing the repeated utilization rate of used sand, and reducing the production cost.

Embodiments of the invention disclose a communication state detection method and a terminal device, relating to the technical field of communication. The method of determining whether the communication state is normal according to signal strength indicated by a signal icon has low accuracy. The invention solves the above problem. The method is applied to a first terminal device and comprises the steps of sending a first message to a first target terminal device; and within a first preset time period, if a second message corresponding to the first message sent by the first target terminal device is not received, determining that the communication state of the first terminal device is abnormal, and showing a prompt message, wherein the prompt message is used for indicating that the communication state of the first terminal device is abnormal. Specifically, the method is applied to a scene where the communication state of the first terminal device is detected.

The embodiment of the invention belongs to the field of artificial intelligence, and relates to an intervertebral disc section image construction method, which comprises the steps of obtaining three-dimensional image data of a lumbar vertebra; extracting intervertebral disc region data in the three-dimensional image data; performing principal component analysis on the intervertebral disc region data to obtain an orthogonal coordinate system of the intervertebral disc region data; constructing a reconstruction plane according to the orthogonal coordinate system; calculating a rotation matrix between the reconstruction plane and the horizontal plane; constructing plane point cloud data of a horizontal plane, and multiplying the plane point cloud data by the rotation matrix to obtain projection data of the plane point cloud data on a reconstruction plane; obtaining a gray value corresponding to the projection data in the three-dimensional image data; and constructing a section image of the intervertebral disc according to the projection data and the gray value. The invention further provides an intervertebral disc section image construction device, computer equipment and a storage medium. According to the method, principal component analysis is carried out on the intervertebral disc region data, so that the accuracy of the constructed intervertebral disc section image can be improved.

The present invention belongs to the field of navigation, and provides a navigation remaining time acquiring method and a navigation remaining time acquiring apparatus. The navigation remaining time acquiring method comprises: receiving a navigation remaining time acquiring request; according to the current position and the end point position of a vehicle, determining a navigation remaining route of the vehicle; acquiring real-time vehicle speed information of the navigation remaining route; and according to the real-time vehicle speed information, calculating the navigation remaining time corresponding to the navigation remaining route. According to the embodiments of the present invention, the navigation remaining time is calculated according to the real-time vehicle speed information of the navigation route, such that the accuracy of the acquired navigation remaining time is high.

The invention discloses a joint training method and device for entity recognition and entity disambiguation and a medium, and the method comprises the steps: carrying out the named entity recognitionof a text based on an entity recognition model, and obtaining a to-be-disambiguated entity; in the constructed knowledge base, performing entity mapping of the entities to be disambiguated, and obtaining candidate entities used for entity disambiguation; inputting the entity to be disambiguated and the candidate entity into an entity disambiguation model for entity prediction to obtain a disambiguation entity, and adjusting a first parameter of the entity disambiguation model according to the disambiguation entity; feeding back the disambiguation entity to the entity recognition model throughan entity attention mechanism layer additionally arranged in the entity recognition model, and adjusting a second parameter of the entity recognition model according to the disambiguation entity; andwhen the first parameter and the second parameter enable the sum of the loss values of the first loss function and the second loss function to meet the minimization condition, completing the joint training. By adopting the method and the device, the problem of low accuracy of entity identification and entity disambiguation in single-model learning in the prior art is solved.

The invention provides a stereo space monitoring and rapid distance measurement method of a powder transmission line channel. The stereo space monitoring and rapid distance measurement method comprises the steps that a binocular vision system is calibrated to obtain binocular vision system internal and external parameters, powder transmission line feature points are matched, three-dimensional reconstruction is conducted on the matched powder transmission line feature points by utilizing the calibrated binocular vision system internal and external parameters, three-dimensional coordinates of the reconstructed feature points are utilized to establish a powder transmission line space equation, then a powder transmission line stereo monitoring area model is established through the powder transmission line space equation and a planer monitoring area, the minimum distances from monitoring area space points to the powder transmission line space equation are calculated, accordingly a monitoring space point-minimum distance database is established, finally whether foreign matter is involved or not can be judged by conducting three-dimensional reconstruction on foreign matter feature points and directly utilizing the database, and the distance between the foreign matter and a powder transmission line is obtained. By adopting the method, the real-time operating amount of later three-dimensional reconstruction and spatial distance measurement can be greatly decreased through earlier-stage establishment of the three-dimensional space point database and the model of the corresponding space distance relation, and an effective guarantee is provided for real-time monitoring of the powder transmission line.

The embodiment of the invention provides a three-dimensional inversion initial model construction method and system based on multi-detection-mode resistivity logging. The method comprises the steps offirstly acquiring a logging curve, and layering a target layer based on the logging curve; then on the basis of a layering result, correcting apparent resistivity in the logging curve so as to obtaintrue resistivity; and finally, carrying out kriging interpolation on the true resistivity, and carrying out resistivity assignment on a finite element mesh required for forward modeling on the basisa result obtained by kriging interpolation. Through correcting the apparent resistivity in the logging curve, the true resistivity closer to a true stratum can be acquired and is used as the resistivity of a three-dimensional inversion initial model based on multi-detection-mode resistivity logging. Through the method provided by the embodiment of the invention, the problem of low initial model precision in the prior art can be solved, the number of iterations of inversion can be effectively reduced, a whole three-dimensional inversion workflow is accelerated, and a technical support can be provided for reservoir assessment and thin oil layer distribution.