39results about How to "Solve technical problems with low precision" patented technology

A commodity recommending method disclosed by the present invention comprises the steps of when a preset shopping website or a shopping APP is detected to be turned on, obtaining the identity information of a user who is browsing the shopping website or the shopping APP; determining whether the identity information exists in a background database; if the identity information exists in a backgrounddatabase, obtaining a commodity and the liking degree which are pre-saved and correspond to the identity information, wherein the liking degree is obtained by pre-acquiring the face expression information when the user browses the commodity; according to the commodity and the liking degree, obtaining a target recommended commodity, and displaying the target recommended commodity in a commodity recommending page of the shopping website or the shopping APP. The present invention also discloses a server and a computer readable storage medium. According to the present invention, the technical problem that the accuracy of the conventional commodity personalized recommendation is not high can be solved.

The invention discloses an image significance detection method and device based on depth information. The method comprises the steps: acquiring a to-be-detected image and a depth image of the to-be-detected image; acquiring a feature image of the to-be-detected image, obtaining a first feature image, and acquiring a feature image of the depth image, and obtaining a second feature image; obtaininga significance image of the to-be-detected image based on the first feature image and the second feature image. The method and device solve technical problems that a significance detection algorithm is not high in precision in the prior art.

The invention provides a positioning method and device, electronic equipment and a storage medium. The method comprises the following steps of by using a pre-set correction mode, according to the obtained MDT data in the preset duration and cell work parameter data, determining the cell work parameter data; according to correction work parameter data, MDT data and XDR data of a user terminal, determining amendment work parameter data, determining resident position information of the user terminal according to the amendment work parameter data, the MDT data and the XDR data, and finally determining positioning information of the user terminal according to the resident position information and the acquired real-time MR data of the user terminal by utilizing a mode matching algorithm. In theprior art, the positioning information of the MR data and the base station position in the cell work parameter data are not accurate; the technical problem that the judgment accuracy of the indoor resident position of the user is not high due to the fact that the indoor resident position of the user is not high in the prior art is solved, and the technical effect that the indoor and outdoor positions of the user are positioned with high accuracy by correcting cell work parameter data through MDT data and accordingly establishing accurate positioning fingerprint information of the user is achieved.

The invention discloses equipment and method for testing a Young modulus and a consumption factor of a loudspeaker cone. The testing equipment comprises a motivation acoustic wave generating device, a shock excitation device, a cone test sample clamp, a three-dimensional automatic guide rail system, a non-contact laser vibration pick-up device, and a processor. The testing method comprises the following steps that: a cone testing sample is clamped for searching a test point by means that a displacement sensor automatically searches a vertical distance position of the point by rotating a driving device; the motivation acoustic wave generating device generates acoustic pressure, so as to enable the cone testing sample to generate simple harmonic vibration through the shock excitation device; the displacement sensor continuously scans all frequencies of a given frequency band at the point, so as to form an analogue electric signal; and the analogue electric signal is subjected to amplification conversion, so that the Young modulus and the consumption factor can be calculated through the processor. With the adoption of the testing equipment and the testing method provided by the invention, the precision of testing the Young modulus and the consumption factor is high.

The method provides a method for acquiring conversation response contents. The method includes the steps: by collecting conversation language materials, obtaining conversation pairs in the conversation language materials, and according to preset scene labels, collecting scene label values corresponding to the obtained conversation pairs and the scene labels; conducting matching and combination on the conversation pairs, the scene labels and the scene label values corresponding to the scene labels to generate conversation corpora; conducting matching on the conversation corpora to obtain response contents matched with current conversation contents which are sent by an opposite communication side, and using the response contents as the conversation response contents corresponding to the current conversation contents. The technical problem is solved that conversation response contents obtained through current conversation corpora matching are not high in accuracy. By automatically establishing the conversation corpora on the basis of the conversation contents, the workload of the manual establishment of the conversation corpora is greatly reduced, and the intelligent degree is high.

The invention discloses a product information pushing method and device. The method comprises the following steps that: when a target user browses a webpage corresponding to a current product, determining the user category of the target user and the product category of the current product; obtaining the average resource transfer value of the product category, wherein the average resource transfer value is a resource transfer average value of the user under the product category; according to the average resource transfer value, determining a first target product under the product category, wherein a difference value of the resource transfer value required by the first target product and the average resource transfer value is smaller than a preset threshold value; and pushing the information of the first target product to the target user. By use of the method, the technical problem of low product pushing accuracy in the prior art is solved.

The invention discloses a label acquisition method and device, a storage medium and an electronic device. The method comprises the steps of obtaining a content vector of a to-be-classified text; inputting the content vector of the text to be classified into the classifier; obtaining the probability that the to-be-classified text output by the classifier belongs to n labels; wherein the classifiercomprises n nodes, wherein the n labels are labels corresponding to pre-acquired training data, each node represents one label in the n labels, the input vector of each node in the training process isdetermined according to the node vector of the node and the node vector of an adjacent node having an adjacent relationship with the node, and n is a natural number. According to the method and the device, the technical problems that different label categories are mutually independent in a text classification process and the precision of an identified classification result is not high are solved.

The invention discloses a target terminal positioning method and device, a storage medium and electronic equipment. The method comprises the steps of responding to a positioning request triggered by a target terminal, and acquiring target ephemeris information and target base station information, wherein the target ephemeris information comprises ephemeris information of a target satellite used for positioning the target terminal, and the target base station information comprises signal information generated by the target base station used for positioning the target terminal; obtaining a pseudo-range observation value based on the target ephemeris information and the approximate position information of the target terminal; and correcting the approximate position based on the pseudo-range observation value and the elevation information corresponding to the approximate position to obtain a target position where the target terminal is located. The technical problem that the positioning accuracy of the target terminal is low is solved, and the method can be applied to automatic driving, unmanned driving and other scenes in the fields of map navigation, high-precision maps, positioning and artificial intelligence.

The invention discloses an image retriever combined with significant features, a database and a retrieval method. The retrieval method comprises the steps: obtaining a three-dimensional model of a paleontology, and carrying out the mapping processing of the three-dimensional model of the paleontology, and obtaining a three-dimensional model view set and a mask image set of the paleontology; collecting a real fossil image set, and preprocessing the real fossil image set to obtain a simulated fossil image set GX; taking the simulated fossil image set GX and the mask image set as input, and training a saliency detection network to obtain a trained saliency detection network; according to the method, the saliency features extracted by the saliency detection network encoder are fused with the global features to serve as the description features of the final image, the description features serve as the final features of fossil image retrieval, feature description of the main body part in the fossil image is enhanced, the influence of noise such as rock on retrieval is reduced, and the retrieval efficiency is improved. The technical problem of low fossil image retrieval precision in the prior art is solved.

The embodiment of the invention discloses a multi-axis numerical control machine tool on-machine measurement data compensation method and system, and the method comprises the steps: measuring a reference coordinate point of a preset standard ball through a measuring head in a zero state of rotating shafts of all machine tools, and obtaining a first coordinate; rotating the rotating shaft to a rotating angle during on-machine measurement; measuring the reference coordinate point of the standard ball through a measuring head to obtain a second coordinate; and obtaining a compensation amount according to the first coordinate and the second coordinate so as to realize error compensation during on-machine measurement. The compensation amount is measured and calculated through the multiple rotating shafts and the preset standard ball, so that error compensation during on-machine measurement is achieved, and the technical problem that in the prior art, compensation is not provided for on-machine measurement of a workpiece, and consequently precision is not high is solved.

The invention discloses a control device for a separator. The device comprises a collection module, a processing module and a motor control module. The collection module is used for collecting image information of suspension liquid in the separator and operation parameters of the separator, the processing module is connected with the collection module and used for obtaining a target opening value range of a bottom valve of the separator according to the image information and the operation parameters, and the motor control module is connected with the processing module and used for controlling the bottom valve to move so that the opening value of the bottom valve can be in the target opening value range. The technical problem that the bottom valve opening is adjusted manually to keep stability of a fluidized bed and an interface according to separators in the prior art, and therefore accuracy is not high is solved.

The invention discloses a resource recommendation method, which relates to the technical field of electronic technologies, is applied to electronic equipment, wherein the electronic equipment is provided with a sensing device, and the method comprises the following steps: obtaining character attribute characteristics of a current user through the sensing device; determining a figure portrait of the current user according to the figure attribute characteristics of the current user, wherein the figure portrait is generated based on historical operation information of the current user; obtainingat least one piece of resource information based on the figure portrait of the current user; and outputting the resource information. The technical problem that in the prior art, when electronic equipment carries out resource putting, precision is not high is solved; recommendation of different resources is achieved according to preferences of different users; and then the effect of accurate recommendation is achieved.

The invention provides a license plate recognition method, a device, a device and a storage medium, belonging to the technical field of image processing. The method comprises the following steps of: acquiring a license plate recognition result recognized by a bit-by-bit recognition method; determining a hypothetical license plate recognition result one-to-one corresponding to the plurality of preset license plate types according to a processing rule of the plurality of preset license plate types and a license plate recognition result; determining a confidence level corresponding to a pluralityof hypothetical license plate recognition results; The hypothetical license plate recognition result corresponding to the maximum confidence level is selected as the target license plate recognitionresult from a plurality of confidence levels. By generating a hypothetical license plate recognition result one-to-one corresponding to a plurality of preset license plate types, thus, the heterogeneous license plate will retain a higher confidence level, and the confidence level of the wrong license plate will be greatly reduced after the above processing. Furthermore, by comparing the confidencelevel of the license plate recognition result after passing the processing rules of different preset license plate types, the accuracy of the license plate recognition is effectively improved.

The invention discloses a mobile terminal reading recommendation method. The method comprises: when detecting that an e-book application installed on a mobile terminal is used, obtaining identity information of a user using the e-book application; determining whether the identity information exists in a background database of the e-book application; if yes, acquiring a pre-stored e-book book corresponding to the identity information and favorite degree based on the e-book, wherein the favorite degree based on the e-book is obtained by pre-collecting facial expression information when the userreads the e-book; obtaining a recommended target e-book according to the e-book and the favorite degree of the e-book, and displaying the target e-book on a recommendation page of the e-book application. The invention also discloses a mobile terminal and a computer readable storage medium. The method can solve a technical problem that accuracy of personalized recommendation of an existing e-book is not high.

The invention discloses an advanced detection method and system for nuclear magnetic resonance, and electronic equipment. The method comprises the steps: obtaining a current detection signal of a nuclear magnetic resonance device; calculating and obtaining the kernel function distribution information of a receiving coil of the nuclear magnetic resonance device according to the forward modeling of nuclear magnetic resonance; and obtaining the moisture content distribution of a front region of a tunnel face according to the current detection signal, the kernel function distribution information, an advanced detection signal and the pre-obtained prior information of an aquiclude. The invention solves a technical problem that the detection precision of water distribution is not high in the prior art.

The embodiment of the invention provides a regional wind power plant group forecasting and intermittent quantitative depicting method and device. The intermittent parameter of the wind speed can be described based on the abrupt change duty ratio of the wind speed; then, a Pearson cross-correlation analysis method is used for carrying out clustering division on the wind speed abrupt change duty ratio time sequences of different wind turbine generators; wind turbine generators with strong correlation are gathered into a virtual unified scheduling unit; determining the optimal forecast length with high predictability in the virtual unified scheduling unit by utilizing a Pearson autocorrelation analysis method; Finally, the BP neural network is used for forecasting the forecast wind speed abrupt change duty ratio of the wind power plant group in the future area, intermittent quantitative description of the wind power plant group in the future area is completed, and the technical problem that in the prior art, a mode of quantitatively describing the intermittence of the wind power plant group through specific parameters does not exist, and the accuracy is not high is solved.

The embodiment of the invention discloses an information push method. The method comprises the steps of after sending attribute information of an expected terminal to at least one user terminal, receiving user preference information from the at least one user terminal, wherein the user preference information is used for representing the preference attribute information of a user generated by the attribute information based on the expected terminal, and the expected terminal is a terminal expected by the user and determined by use of a matching algorithm based on the attribute information of the user; when the user preference information satisfies a preset condition, correcting the matching algorithm based on the user preference information to obtain a corrected matching algorithm; and determining a new expected terminal by use of the corrected matching algorithm, and pushing the attribute information of the new expected terminal to at least one user terminal. The embodiment of the invention also discloses an information push device.

The invention discloses a media file delivery method and device, a storage medium and an electronic device. The method comprises the steps: acquiring first position information of first target user equipment in a first target user equipment set at a first moment, wherein the target user equipment set represents a set of user equipment using a target network protocol IP in a first target time period; acquiring second position information of the first target user equipment at a second moment; under the condition that the time interval between the second moment and the first time is greater thanor equal to a first preset time threshold, dividing the first target user equipment into a second target user equipment set; delivering the target media file to the user equipment in the second targetuser equipment through the display terminal corresponding to the second position information, wherein the technical problem of relatively low media file delivery accuracy caused by incapability of accurately dividing the crowd in the target area in the prior art is solved.

The invention relates to the technical field of power dispatching, and discloses a virtual power plant day-ahead accurate peak clipping method, system and device and a storage medium. When the day-ahead peak clipping demand is received, the real-time power supply path of each user is obtained according to the user account data, each real-time power supply path is inquired according to the identifier of the heavy overload device in the day-ahead peak clipping demand, and if the heavy overload device is inquired in the power supply device set corresponding to the real-time power supply path, the real-time power supply path of each user is inquired in the power supply device set corresponding to the real-time power supply path. Taking the user corresponding to the real-time power supply path as a target user, constructing a target user set, initiating an invitation to each target user in the target user set, determining a bid-winning user according to the feedback of each target user for the invitation, and issuing a day-ahead scheduling plan curve to each bid-winning user, and each bid-winning user carries out peak clipping feedback according to the day-ahead scheduling plan curve. According to the method, the accurate peak clipping requirement on a certain main transformer, a distribution transformer or a feeder line can be met, and the peak clipping accuracy is improved.

The present application relates to a power distribution network fault detection apparatus, system and method. The power distribution network fault detection apparatus comprises at least two voltage acquisition devices, at least two current acquisition devices, and a distribution transformer terminal. The distribution transformer terminal can determine a fault between an outlet terminal of a high-voltage end device and an inlet terminal of a low-voltage end device according to a first voltage and a first current, and determine a fault between an outlet terminal of the low-voltage end device andan inlet terminal of a user side device according to a second voltage and a second current. When an abnormality occurs in any parameter of the first voltage, the first current, the second voltage andthe second current, a specific section of the power distribution network can be determined according to the abnormal parameter. Therefore, the staff can quickly locate the fault, and further investigate and repair the fault. The apparatus, system and method solve the technical problem of low precision of power distribution network fault detection in the prior art, and achieve the technical effectof improving the precision of power distribution network fault detection.

The invention discloses a voice noise reduction method and device, a computer readable storage medium and an electronic device. The method comprises the following steps of performing voice separationon voice data to be denoised to obtain first voice data and second voice data of the voice data, and determining a first probability function corresponding to the first voice data and a second probability function corresponding to the second voice data; determining target noise reduction data of the voice data through the first probability function and the second probability function; performing noise reduction processing on the voice data through the noise reduction data; obtaining target voice data. According to the method, the voice data subjected to noise reduction is separated into two branches, namely the first voice data and the second voice data, and the noise data separated from the two branches is utilized to reduce the noise of the voice data mixed with the noise, so a technicalproblem of low accuracy of voice noise reduction in the prior art is solved.

The invention discloses a method and device for determining an optimal verbal skill sequence and a storage medium. The method comprises the following steps: determining user feature information and dialing information of a target user; by utilizing a pre-trained prediction model, according to the user feature information and the dialing information, respectively determining a predicted hang-up rate and a predicted hang-up conversion rate of each verbal skill in the plurality of verbal skill sequences and interaction process information of interaction with the target user, wherein the interaction process information comprises a current verbal skill sequence interacting with the target user, a verbal skill subsequence interacting with the target user and a subsequent verbal skill sequence interacting with the target user; forming a plurality of verbal skill sequences by verbal skill corresponding to each verbal skill node; and determining an optimal verbal skill sequence interacting with the target user according to the predicted hanging-up rate and the predicted hanging-up conversion rate of each verbal skill in the plurality of verbal skill sequences.

The invention discloses an abnormal cervical blood cell screening method based on TCT film preparation. The abnormal cervical blood cell screening method comprises the following steps: constructing an abnormal cell screening model; training an automatic coding instrument; training a single-class SVM (Support Vector Machine) classifier; acquiring a cervical cell image of a patient, and cutting the cervical cell image to obtain a plurality of cervical image packets; inputting the plurality of cervical image packets into the abnormal cell screening model to obtain prediction results corresponding to the plurality of cervical image packets; screening out the corresponding cervical image packet according to the obtained prediction result, and recording the cervical image packet as a first cervical sub-image set; inputting the plurality of cervix uteri image packets into the automatic coding instrument to obtain all the positive areas in the plurality of cervix uteri image packets; substituting all the obtained positive regions into the single-class SVM classifier, extracting the cervical image packet belonging to positive features, and recording the cervical image packet as a second cervical sub-image set; and setting a comparison model so as to quickly and accurately screen out the abnormal cervical blood cells.

The invention discloses a personality testing method and system. The method comprises the following steps: acquiring a brain wave signal; preprocessing the brain wave signal; transmitting the preprocessed brain wave signal to a diagnosis module; and analyzing the preprocessed brain waves by using a diagnosis terminal to obtain personality information. The personality testing method and system solve the technical problem of low precision of a personality evaluation method in the prior art.

The invention discloses a control method and device for a separator. The method comprises the steps that image information of suspension liquid in the separator and operation parameters of the separator are collected; a target opening value range of a bottom valve of the separator is obtained according to the image information and the operation parameters, and the bottom valve is controlled to move so that the opening value of the bottom valve can be in the target opening value range. The technical problem that the bottom valve opening is adjusted manually to keep stability of a fluidized bed and an interface according to separators in the prior art, and therefore accuracy is not high is solved.