223results about How to "Reduce the difficulty of identification" patented technology

The invention relates to an emotion recognition method and system and an intelligent nursing system using the method. The emotion recognition method comprises the steps that a tested face image and an electroencephalogram signal at the same moment or in the same time period are acquired; the tested face image and the electroencephalogram signal are processed to obtain an expression parameter value and an inner emotion parameter value respectively; data fusion processing is performed on the expression parameter value and the inner emotion parameter value to obtain an emotion fusion parameter value; and a current emotion of a tested face is recognized in a pre-stored emotion fusion database according to the emotion fusion parameter value to obtain emotion information. In this way, the information fusion technology is utilized to process a face and a body brain wave in a combined mode from face image processing and body brain wave processing, body emotion recognition is realized, the accuracy and reliability of emotion recognition are improved, recognition difficulty is lowered, and influences of individual differences on emotion recognition are avoided.

The invention relates to mobile phone voice control technology and discloses a voice control method of functions of a mobile phone, which is realized by voice recognition technology. The method comprises the following steps that: a user inputs selected voice key words into the mobile phone and setting the mapping relationship between the input voice key words and the functions of the mobile phone to form a voice control instruction collection; and in a voice control mode, the mobile phone automatically executes a function corresponding to a voice command when the voice command captured by the mobile phone is the same as a voice command preset by the user.

The invention provides an identification method for turnout faults. The method comprises steps as follows: a curve of each action of a turnout is collected; the obtained turnout action curves are divided into normal curves and fault curves; for the normal curves and the fault curves, one normal curve and one fault curve with the most representative characteristics are selected as representative curves of the normal curves and the fault curves respectively; similarity 1 of a to-be-identified curve and the representative normal curve and similarity 2 of the to-be-identified curve with the fault representative curve are calculated with a similarity algorithm; the similarity 1 and the similarity 2 obtained through calculation are compared, if the similarity 1 is larger than the similarity 2, the curve is a normal curve, and if the similarity 1 is smaller than the similarity 2, the curve is a fault curve. According to the identification method for the turnout faults, the problems of missing report and mistaken report caused by the fact that the turnout fault type is judged through artificial experience in the prior art are solved, so that the turnout fault category is automatically identified, repair efficiency and system reliability are improved, and traffic safety is guaranteed.

The invention relates to a generative adversarial network-based remote sensing image thin cloud removal method. The method includes the following steps that: step 1, a thin cloud removal system modelis established; step 2, a network model is designed; step 3, an identification criterion function is constructed; step 4, the thin cloud removal system performs identification; and step 5, thin cloudsin a remote sensing image are removed. According to the method of the invention, the generative adversarial network is adopted to construct the model of an image thin cloud removal problem and perform system identification on the model, thereby realizing end-to-end thin cloud removal; and the constructed criterion function combines the errors of data distribution and reconstruction accuracy, so that the system can better learn the characteristics of data, thus, achieving the removal of the thin clouds. The method provided by the invention has a capacity to adaptively remove uneven thin clouds. With the method adopted, a restored image has high color and texture consistency.

The invention discloses a method for analyzing amine substances in dansyl chloride derived-plasma based on liquid chromatography mass spectrometry. The method is characterized in that dansyl chloride is used as a derivatization reagent to perform derivatization on a metabolite containing primary amine and secondary amine, therefore the retention capability of the amine substances in reversion phase chromatography is increased, the ionization efficiency in mass spectrometric detection is increased, and the purpose of increasing the quantitative accuracy and detection sensitivity of the amine substances is achieved. The method has wide coverage for amine metabolites (comprising amino acid, methylation products, acetylation products, dipeptide and the like), can utilize 100 [mu]L of plasma samples to detect 113 amine substances, supplies the outline information of metabolome of the amine substances, and has the characteristics of good specificity, high accuracy and repeatability, less reagent consumption and the like.

The invention relates to a method for faults caused by foreign body pollution between electrified railway insulator plates based on affine invariant moment, which aims at solvating the problems of a complicated background of insulator problem detection and the like caused by much original image content. The method can be used quickly and accurately identifying and positioning insulators through overall image classification and by virtue of the affine invariant moment, and then can be used for carrying out fault detection on the insulators. The method comprises the following steps of: detecting, identifying and classifying shafts by virtue of a Hough transformation segment, thus obtaining an insulator type after setting the classified shafts as a background; then applying the affine invariant moment to extract the insulators in the insulator type; in order to solve the problem of different change of the angles of the insulators, carrying out straight-line fitting on the boundary curves of the insulators to obtain the rotation angles of the insulators; in order to avoid the loss of foreign body characteristics, amplifying the characteristics of the insulators by virtue of morphological dilation operation; in order to improve the local contrast of insulator images, dividing each piece of the image into a block to carry out local contrast control on each block; and then carrying out grey level value statistic on each row of the blocks, and finally identifying and positioning the faults through the width of each strip in a grey level statistical chart. The method based on prior classification disclosed by the invention simplifies the background of the insulators, efficiently improves the speed and accuracy of insulator identification and positioning, reduces the difficulty of fault detection, and provides an advisable reference for the insulating property detection of an electrified railway.

The invention relates to an automatic classification system of aerial photographing line patrol images based on depth learning. The method comprises the following steps of: establishing a line patrolimage classification image library and a label library thereof; establishing a deep learning model, including VGGNet, ResNet, DenseNet, NasNet and other high performance classification models; training a classification model, and randomly performing data enhancement operation on the input data of each iteration in the training process, including rotation, filling cut and grayscale; integrating thefour trained models and obtaining the integrated model; setting classification rules, inputting the images to be detected into the integrated model, taking the three image categories with the first three confidence levels as the classification results of the images to be detected, and copying and storing the results to the servers of the three categories for archiving; continuously optimizing theclassification model, and periodically upgrading the performance of the automatic classification system.

The invention provides a target object occlusion detection method and a target object occlusion detection device. The target object occlusion detection method comprises steps: input images are received; a well-trained neural network is used to detect a target image in the input images and output a thermodynamic chart of the target object; and based on the thermodynamic chart, whether occlusion exists in the target object is detected. According to the target object occlusion detection method and the target object occlusion detection device provided by the embodiment of the invention, the target object in the input images is converted to the thermodynamic chart based on the well-trained neural network, whether the occlusion exists in the target object is detected through the thermodynamic chart, the target object recognition difficulty can be effectively reduced, and the recognition precision and the stability can be enhanced.

The invention provides a cooker, a control method and device thereof and a storage medium. The method comprises the steps: obtaining image information and weight information of food materials; and determining the maturity corresponding to the food materials according to the image information and the weight information. According to the control method, size information and type information can be obtained by identifying the image information, volume information of the food material is further obtained according to the size information, density information is calculated according to the volume information and weight information, and the maturity corresponding to the food material is determined in a pretrained maturity model. Thus, volume and weight change, namely the density change, of the food materials can be monitored in real time, the maturity is identified automatically by means of the density, observation time of a user is greatly reduced, and the identification difficulty is effectively reduced.

The invention relates to a method for identifying a relationship between a transformer area phase sequence and line households. The method comprises the steps of: acquiring an active current value ofa low-voltage outlet line of each phase of the target transformer area, an active current value data and voltage time sequence data of each electric meter, and voltage time sequence data of a three-phase bus at a distribution transformer low-voltage side; classifying types of the electric meters according to the voltage time sequence data, and acquiring electric meter sets; determining the electric meter set to which the low-voltage outlet line of each phase is to be assigned according to the active current value of the low-voltage outlet line of each phase, the active current values and the voltage time sequence data of the electric meters, the voltage time sequence data of the three-phase bus at the distribution transformer low-voltage side and the electric meter set; and combining arbitrary electric meter sets in electric meter sets of each type, and matching the combined electric meter set to be an electric meter classification set Gamma of a phase low-voltage outgoing line i if asum of the active current values after combination and an active current value of the phase low-voltage outgoing line i satisfy a first preset condition. The method and the device have the advantagesof low identification difficulty, high precision, no need of additionally arranging other terminal equipment, high operability and small operation and maintenance pressure.

A capacitive touch keyboard includes an insulating substrate that defines a plurality of press key zones, and a first and a second conductive layers, both disposed on the insulating substrate. Each conductive layer includes a plurality of electrode lines arranged in two different directions. Each electrode line includes a plurality of electrode units connected in series. Each electrode unit is disposed at a position corresponding to a respective one of the press key zones. For each of the press key zones, the electrode units of the first the second conductive layers that are disposed therein are adjacent to each other and cooperate to form a touch sensing structure.

The invention discloses a method for estimating the remaining power of an electric vehicle power battery. The method comprises steps that based on the volume Kalman algorithm, a model for estimating the remaining power of the power battery is established; polarization internal resistance Rp(k), polarization capacitance Cp(k), equivalent ohmic internal resistance R0(k), the remaining power SOC (k)and an end voltage Ut(k) of the power battery at the time k are obtained; an open circuit voltage UOC(k) of the power battery at the time k is calculated; a state estimation error delta e(k) and a noise error V(k) are calculated; the BP neural network based on width learning is constructed, the delta e(k) and the V(k) are inputted into the BP neural network, and a variance compensation value deltaQk of process noise distribution at the time k and a variance compensation value delta Rk of observed noise distribution are outputted by the BP neural network; the delta Qk and the delta Rk are utilized to compensate Qk-1 and Rk-1 at the time k-1, and Qk and Rk at the time k are generated; a value of x(k+1) of the model for estimating the remaining power of the power battery is calculated through the volume Kalman algorithm, and the remaining power SOC (k+1) at the final (k+1) time is obtained. The method is advantaged in that the remaining power of the power battery can be rapidly and accurately estimated.

A valuable file identification method includes step 1: selecting a characteristic area of the valuable file, and extracting a valuable file characteristic for last classification; step 2: an input valuable file is fast classified according to the extracted valuable file characteristic in step 1 to gain the banknote kind, denomination, direction and image quality information of the valuable file, and the banknote with better image quality and bad image quality are selected; step 3: an image restoration technique is utilized based on a partial differential equation to restore the old banknote image; step 4: the new banknote is directly identified and the old banknote is identified via the restored image to judge the authenticity of the current banknote; step 5: a result is output. The method enables eliminating restoration treatment for images comprising good quality and uninterested area, and saving time and improving system processing efficiency. A valuable file identification system and a valuable file identification device are also disclosed.

The invention provides an antenna controller for an advanced broadcasting system-satellite (ABS-S) satellite signal mobile-satellite, and belongs to the technical field of antenna control of the mobile-satellite. The antenna controller solves the problems that the antenna controller cannot accurately identify an ABS-S standard satellite, is inconvenient for system upgrading, and complex in control and operation and the like in the prior art. The antenna controller is independent of a mobile-satellite antenna, is connected with the mobile-satellite antenna and a satellite receiver respectively, and comprises a main control circuit, a power supply management circuit, and a liquid crystal display (LCD) liquid crystal screen and a key circuit which are connected with the main control circuit respectively, wherein the power supply management circuit is connected with the main control circuit, supplies power to the main control circuit, and can convert input direct current voltage into required voltage; and the main control circuit can distinguish the voltage of the satellite receiver, and simultaneously the main control circuit normally communicates with the mobile satellite antenna and controls the mobile satellite antenna until a target satellite is locked. The antenna controller can accurately identify the ABS-S standard satellite, and is convenient for system upgrading and simple and convenient for control and operation.

The invention provides a turnout action reference curve selection method and an application thereof. The method comprises the steps of obtaining N action curves of a same turnout through a turnout microcomputer monitoring system, obtaining action time of the N curves, selecting the action time with a highest repetition number in the N action time, and marking the action time with the highest repetition number as Tx; assuming that M curves with action time lengths of Tx exist, and renumbering the M curves; for the M curves, dividing each curve into P equally divided parts according to the action time, and for a current value of a jth moment of the ith curve, obtaining current value matrixes I of the M curves in sequence for all the M curves; for M current values of any moment tj, calculating out a clustering center by utilizing a clustering algorithm, and marking a current value of the clustering center as C(j), wherein C(j) is the current value of the clustering center; and calculatingclustering centers of P moments in sequence, and by analyzing the clustering centers, selecting out a curve with a highest clustering frequency as a reference curve. Based on the reference curve, anabnormal curve diagnosis method is obtained by utilizing a similarity method.

The invention provides a method for estimating the open-circuit voltage of a battery based on a gas-liquid dynamic model and a device. The method for estimating the open-circuit voltage of the batterybased on the gas-liquid dynamic model comprises the following steps that 1, an undetermined open-circuit voltage estimation equation is deduced according to the gas-liquid dynamic model; 2, equationparameters are identified and estimated according to experimental data; and 3, an open-circuit voltage estimation method is designed according to the complete open-circuit voltage estimation equation,and calculating is carried out to obtain an open-circuit voltage estimation value. The open-circuit voltage estimation equation comprises the battery temperature, does not need to be coupled with thebattery temperature again, and simplifies the open-circuit voltage estimation process. The principle of gas dissolution / precipitation in the estimation equation is equivalent to the polarization effect of the battery, the phenomenon that the open-circuit voltage lags the terminal voltage of the battery cab be better fitted. An estimation result of the estimation equation does not depend on the precision of an input initial value, and excellent estimation robustness is achieved. According to the method and device for estimating the open-circuit voltage of the battery based on the gas-liquid dynamic model, the non-linear process of battery charging and discharging can be more accurately depicted, a simple analytical formula, easy parameter identification and small calculation amount are achieved, the influence of battery temperature characteristics on the open-circuit voltage is reflected, and the method and device for estimating the open-circuit voltage of the battery based on the gas-liquid dynamic model is convenient to realize in engineering.

The present invention discloses a radar networking identification target method, belonging to the field of the radar technology. The method comprises: s radars are configured to respectively measure n attributes of a detection target, forming s groups of radar measurement data, and belonging the detection target to one of the possible targets, wherein each attribute of each possible target has one specific value range. According to the invention, the anti-interference capability of an identification system is improved, the identification difficulty is reduced, and the identification accuracy is improved; the identification classification process of the method is performed from top to bottom, is more simple and reliable relative to the current technology, and has high popularization and application values.

The invention discloses an instrument pointer reading automatic identification method and system based on vision. The method comprises the steps of firstly collecting a watch image; segmenting and extracting a pointer and each region of the watch by using a neural network; using edge detection to screen out the scale centroid, fitting the inner circle of the dial plate, determining the center of the dial plate, detecting the center line of the pointer, calculating the rotation angle of the center line relative to the reference line by taking the center point of the dial plate as the referencepoint and the centroid of the reference point and the 12-point position as the reference line, converting the calculated angle into watch reading, and storing the current watch reading into a database.

The invention provides a bridge stay cable force online detection method and system based on fiber sensing. The system comprises a fiber grating vibration sensor group, a fiber grating vibration demodulation instrument and a stay cable vibration signal processing unit. The fiber grating vibration sensor group comprises a plurality of fiber grating vibration sensors connected in series with the same fiber, each fiber grating vibration sensor is designed based on a cantilever structure and a double-raster coupling method, and chirp rasters are used to replace Bragg rasters, such that the detection scope of the sensors is enlarged, the stability and the anti-interference capability of the sensors are improved, and at the same time, the manufacturing difficulty of the sensors is reduced. The fiber grating vibration demodulation instrument is based on an edge filtering demodulation principle and employs a wavelength division multiplexing technology to realize distributed measurement of a stay cable force, and multiple fiber grating vibration sensors can be multiplexed in one fiber signal, such that the detection capability of the system is improved, the distributed measurement of the stay cable force is made possible, and a more reliable technical means is provided for online detection of a wide-span bridge stay cable.

The invention discloses an internet-of-things terminal intelligent configuration method and system based on appearance information recognition. The method comprises the following steps of: carrying out photographing on controlled equipment by a camera and obtaining appearance information of the controlled equipment; transferring the appearance information of the controlled equipment to a home control center by an internet of things; and carrying out matched processing on the appearance information of the controlled equipment and internal digital codes of the same controlled equipment as complete identification information of a controlled object, and returning a remote controller for operation of the controlled equipment. According to the invention, due to adoption of matching of the appearance information of the controlled equipment and the internal digital codes of the same controlled equipment as the complete identification information of the controlled object, the external characteristics of different household appliances are different completely, so that the recognition difficulty for different household appliances is greatly reduced, the recognition accuracy is improved, further recognition software can be optimized and used hardware equipment also can be simplified.

The invention discloses a vehicle identification method based on fusion of radar and vision information. The vehicle identification method based on fusion of radar and vision information includes thefollowing steps: S100, entering all the information of a vehicle to be identified; S200, acquiring the current positional information and the motion state information of the vehicle; S300, acquiring the appearance information of the vehicle to be identified; S400, comparing the acquired appearance information of the vehicle to be identified with the entered vehicle information; and S500, recordingand uploading the information of the illegal vehicle and the vehicle driver. The vehicle identification method based on fusion of radar and vision information monitors the vehicles on the road in real time through the method of integrating the radar with vision information, determines the best position of shooting through the radar, can capture the clear image of the vehicle to be identified, canobtain the information of the vehicle to be identified through vision information processing, can determine whether the driver has illegal behavior through information comparison, and can greatly improve the accuracy and efficiency of vehicle identification through the method of combining the radar and the vision information.

The invention discloses a method for automatically acquiring a fetal key section ultrasonic video center dynamic period video, which specifically comprises the following steps of: firstly, positioninga heart part which is continuously deformed by using a target tracking technology, and then primarily identifying diastolic and systolic end stages by using a classification method, so as to obtain arough heart period boundary frame; and finally, using a cross gating bilinear matching model for checking and finely positioning a fragment matched with the standard video in the long video, so thata complete cardiac cycle video under a related standard section can be more accurately matched from two aspects of obvious boundary frames in the cardiac cycle video and dynamic matching in the middle. The invention aims to intelligently and automatically intercept a fetal cardiac cycle video from an ultrasonic video by using deep learning, and provide the most direct and effective basis for subsequent intelligent diagnosis.

The invention discloses a semantic similarity-based industry type recommendation method and a device, equipment and a medium, and the method comprises the steps: generating an enterprise operation description text and a plurality of industry type description texts corresponding to a target enterprise based on a target enterprise name when the target enterprise name is received; preprocessing the enterprise operation description text to obtain a first keyword set, and preprocessing the plurality of industry type description texts to obtain a plurality of second keyword sets; calculating semantic similarity between the first keyword set and each second keyword set, and generating a semantic similarity set; wherein the semantic similarity is obtained on the basis of a pre-trained Word2Vec language model; and recommending an industry type corresponding to the target enterprise name based on the semantic similarity set. According to the invention, the identification difficulty of the industry type to which the enterprise belongs can be reduced, and the monitoring efficiency of the supervision department is improved.

The invention relates to a method and a device for recognizing a sequence in an image, electronic equipment and a storage medium, and the method comprises the steps: carrying out the feature extraction of a to-be-processed image, obtaining a first feature map of the to-be-processed image, and enabling the to-be-processed image to comprise a sequence formed by the stacking of at least one object ina stacking direction; determining regional features of each block region in the to-be-processed image according to the first feature map, each block region being obtained by dividing the to-be-processed image into k regions along a stacking direction, k being a set number of objects stacked along the stacking direction, and k being an integer greater than 1; and determining the category of each object in the sequence according to the regional characteristics of each block region. According to the method of the invention, the object in the stacking sequence can be recognized.

The invention discloses an object pose detection method and device. The object pose detection method comprises receiving an original image acquired by a camera, extracting and obtaining a marked code image from the original image, wherein the marked code image is an image of a marked code attached to the object to be detected; obtaining the pose of the marked code through calculation in dependence on the marked code image and a pre-generated pose calculating model, wherein the pose of the marked code includes the rotation amount and the displacement amount of the marked code relative to the reference pose; obtaining the current pose of the object to be detected through calculation in dependence on the pose of the marked code. According to the invention, the object pose detection process can be simplified, the efficiency of the detection process can be improved, and the accuracy of the detection result can be improved.

The invention relates to a method for simplifying an electric power system synthetic load model and identifying model parameters, and belongs to the technical field of electric power system load stability monitoring. According to the method, the internal relationship between parameters of an induction motor is deeply analyzed, and on this basis, by converting the input-output relationship of the model, a simplified electric power system synthetic load model is provided. According to the simplified electric power system synthetic load model, possible steady-state operation states of multiple sets of simplified electric power system synthetic load models are preset, parameter identification is carried out through the least square method on the basis of a state equation of the part of the induction motor of the simplified electric power system synthetic load model, corresponding indexes are provided to reflect identification precision, and the set model with the highest precision is selected from all preset steady state conditions, and an identification result corresponding to the set of model is used as the final identification result. The method has high generalization performance and feasibility, and can accurately and quickly identify the parameters of the electric power system synthetic load model.