56results about How to "For subsequent calculations" patented technology

The invention provides a vehicle lower boundary detection method based on multi-sensor fusion. Through measuring information of a millimeter-wave radar and a camera, space alignment points are obtained, an area of interest containing a target vehicle is selected in a camera image according to space alignment point information, by determining k peak values of the area of interest, k-1 threshold values of the area of interest are determined, a minimum threshold value is obtained, and accordingly a vehicle lower boundary shadow area corresponding to a gray level area marked out by the minimum threshold value, finally, a vehicle lower boundary shadow line is obtained in the shadow area, and vehicle lower boundary detection is achieved. By using a certain searching strategy, the target area of interest is determined, the target area of interest can contain the target vehicle, the size of the area can be proper, following computing is convenient, the k-1 threshold values are determined through a particle swarm optimization algorithm, operation steps are simplified, operation speed is improved, and meanwhile area dividing accuracy is improved.

The invention discloses a service data processing method and device, computer equipment and a storage medium. The method comprises the steps of obtaining a corresponding target service configuration file according to a service channel and a service role; Inquiring a service database based on the configuration file ID, and judging whether a service performance formula template corresponding to theconfiguration file ID exists or not; If not, generating response reminding information, and sending the response reminding information to the client; Obtaining a service adding request sent by the client based on the response reminding information, and entering a formula editing interface based on the service adding request; Configuring operation parameters and corresponding operation logic on a formula editing interface, obtaining a business performance formula template corresponding to the configuration file ID, and storing the business performance formula template in a business database; And calculating the operation parameter value in the to-be-imported data by adopting the operation logic to obtain target data. The business performance formula template can be conveniently adjusted atany time according to business requirements, and the maintenance cost of the business performance formula template is saved.

The invention discloses an audio similarity detecting method, an audio similarity detecting device and electronic equipment. The detecting method comprises the following steps that: an audio stream to be detected is obtained; a feature parameter value of the audio stream to be detected is determined; a previously known feature parameter value of a reference audio stream corresponding to the audio stream to be detected is compared with the feature parameter value of the audio stream to be detected; and according to a comparison result, the similarity between the audio stream to be detected and the reference audio stream is calculated. According to the technical scheme disclosed by the invention, the feature parameter values between the audio streams are compared, so that the judgment of the similarity is more accurate and quantitative, and the use experience of users is favorably improved.

The invention relates to an automobile engine fault judgment method based on sound recognition and a device thereof, belonging to the technical field of engine fault detection, which solves the problems of inaccurate self-judgment of faults in the prior art, and is unable to utilize engine running state information in judgment process and re-judgment of faults required in later maintenance. The invention discloses an automobile engine fault judging method which adopts a hybrid neural network, Fault identification is based on AlexNet, and LSTM is used to identify engine running time state to assist more accurate fault diagnosis. The number of layers of AlexNet and LSTM neural network is more than 9. The experiment results show that the identification accuracy is high and the judgment resultis very accurate. In practical application, the automobile engine fault judging method of the invention can simply, quickly and reliably judge the common faults of the automobile engine, timely remove the faults, and the later maintenance does not need to re-judge the faults.

The invention discloses a remote sensing image classification method based on partially random supervision discrete Hash. The method comprises the following steps that: according to calibrated real data, carrying out object segmentation on a remote sensing image, and carrying out characteristic extraction on a segmented object, wherein each object is represented by a characteristic vector, and thecharacteristic vectors of all objects are combined into a characteristic matrix; according to the same ratio, dividing each category of samples into training samples and testing samples; carrying outdiscrete Hash coding on all samples; carrying out partially random Hash coding on all samples; combining the discrete Hash coding with the partially random Hash coding, carrying out iterative optimization on parameters, and finally, obtaining more accurate Hash coding; and according to the generated Hash coding, calculating Hamming distance to finish classification. By use of the method, the problem of high calculation complexity caused by an overlarge data size in a remote sensing image processing process is solved, and the remote sensing image can be quickly and effectively classified.

The embodiment of the invention discloses a graphic processing method and device based on a writing track, equipment and a storage medium, and relates to the field of electronic whiteboards, and the method comprises the steps: displaying at least one writing track in response to a received handwriting control operation, at least one track point of the writing track being located in a closed area of a target graph, and the target graph being a displayed polygon; and adjusting the size data of the target graph to enable the target graph to surround and / or adapt to the distribution range of the writing track. By the adoption of the scheme, the technical problem that in the prior art, the integration efficiency of writing tracks and graphs is low can be solved.

The invention discloses a method for obtaining drainage basin underlying surface information based on remote sensing classification. The method includes a runoff grid extracting algorithm based on multi-source data and a confluence network constructing method. The runoff grid extracting algorithm based on multi-source data comprises data source selecting and processing and the runoff grid extracting algorithm. According to the method, due to the requirement for confluence calculation of a reservoir flood prediction model, the traditional confluence network constructing method with a traditional DEM as the unique data is improved, extractive ground covering information and the DEM are comprehensively utilized for constructing a confluence network, and the problem of a parallel river channel in a traditional algorithm is effectively solved.

The invention relates to a method for extracting roads from remote sensing map images in the field of remote sensing technique. The method comprises the following steps of: marking two edges of the roads; automatically generating road segments by using a projection method according to the marked edges; linking the road segments into the roads; adding road names; and saving every road in a data format by taking an intersection as the center. The method for extracting the roads from the remote sensing map image has the advantages of easy implementation, simple operation, high accuracy and the like, and is a fast and conveniently extracting mode for the roads of a remote sensing map.

The invention discloses a judgment method for a partial discharge development stage of oil paper insulation. A device capable of simultaneously testing partial discharge of oil paper insulation and amoisture content in oil is designed; an adopted test electrode is a column plate electrode; a pressurizing device is used for pressurizing the oil paper insulation; the partial discharge development process of oil paper insulation and the moisture content in oil are monitored through a partial discharge tester and an in-oil moisture tester respectively. The measured data are transmitted to a computer, and the oil paper insulation partial discharge development stage is determined according to the characteristic parameters measured by a PRPD spectrogram in the taken time period after the partialdischarge development stage is divided. The device involved in the method considers the influence of moisture in the insulating oil on the partial discharge development process, so that the method can more accurately judge the partial discharge development stage of the oil paper insulation, and the existence of serious defects of the oil paper insulation can be predicted in advance.

The invention discloses a retina image preprocessing method, and the method is executed in the computing equipment. The method comprises the steps: obtaining a retina image which comprises a retina area and the background; cutting the retina image, wherein the cut image comprises a circular retina area; adjusting the size of the cut image into a pre-determined size; and carrying out the contrast enhancing of the image in the pre-determined size. In addition, the invention also discloses a retina image preprocessing device which can implement the above method, and the computing equipment which comprises the retina image preprocessing device. According to the technical scheme of the invention, the method can achieve the preprocessing of the retina image, so as to improve and unify the quality of retina images.

The invention discloses a special-shaped part detection tool and method, and belongs to the technical field of workpiece detection. The device comprises a first detection block and a second detectionblock, the first detection block is vertically connected with the second detection block, the first detection block is provided with a first working surface, the second detection block is provided with a second working surface, and the first working surface and the second working surface are provided with arc members respectively. The tool is simple in structure, high in structural rigidity and convenient to manufacture and use; and according to the detection method, use of complex machine parts are not needed, existing detection equipment can be used, a dial indicator needs not to change theposition in the detection process, and the detection method is simple and convenient to use.

The invention discloses a car lamp lampshade stress measurement loading device. The car lamp lampshade stress measurement loading device comprises a frame structure and a loading structure arranged on the frame structure, the loading structure can slide on the frame structure, and the frame structure comprises a pedestal, a working plate and support columns arranged between the pedestal and the working plate. The loading structure comprises an upper adjusting slide block, a lower adjusting slide block fixedly connected with the upper adjusting slide block, an upper pressing plate arranged under the lower adjusting slide block and a lower pressing plate arranged under the upper pressing plate, a unit seat is fixedly arranged on the upper adjusting slide block, a pressure screw shaft is arranged in the unit seat in a matching manner, a worm gear is arranged on the upper end of the pressure screw shaft in a matching manner, a worm is arranged at one side of the worm gear in an engaged manner, and the lower end of the pressure screw shaft traverses the upper adjusting slide block and the lower adjusting slide block and is in threaded connection with the upper pressing plate. The car lamp lampshade stress measurement loading device is convenient for loading, and is easy to operate, good in visibility and simple in structure, and the problem that the previous car lamp lampshade stress measurement process is complex and troublesome is solved.

The invention provides a knowledge point importance calculation method. The method comprises: firstly, obtaining all or part of knowledge points as a knowledge point set; secondly, determining directed relationship strength between any two knowledge points in the knowledge point set, and obtaining association information of the knowledge points; and finally, performing iteration by utilizing the association information, and determining importance information of each knowledge point in the knowledge point set. According to the scheme, the directed relationship strength between a current knowledge point and other knowledge point, and the directed relationship strength between the other knowledge point and the current knowledge point are considered, so that the relationship between the current knowledge point and the other knowledge point can be objectively represented through the two relationships, the importance of the current knowledge point can be objectively reflected, and an objective importance evaluation standard is determined for knowledge points in the fields.

An embodiment of the invention discloses a division operation method and device, equipment and a storage medium. The method includes: setting numerical values stored in exponent sign bytes in an original dividend and an original divisor to be identical to obtain a current standard dividend and a standard divisor; sequentially selecting numerator quotient estimation bytes and denominator quotient estimation bytes from the current standard dividend and the standard divisor; according to the numerator quotient estimation bytes and denominator quotient estimation bytes, dividing the current standard dividend by the standard divisor to obtain a quotient value, taking a first digit of the quotient value as a current digit quotient value, adding into a temporary quotient according to a sequence from high to low to serve as a last digit; according to numerical values stored in the exponent sign bytes in the original dividend and numerical values stored in the exponent sign bytes in the original divisor, calculating the temporary quotient to obtain a final quotient. By adoption of the method, division operation efficiency can be improved.

The invention provides a multi-feature fusion method and device, electronic equipment and a storage medium. The multi-feature fusion method comprises the steps of obtaining a plurality of nonlinear features in an image data pair of a target object; inputting each nonlinear feature into a preset multi-kernel function to obtain a feature weight value corresponding to each nonlinear feature; and obtaining the fusion feature of each nonlinear feature according to the weighted sum of each feature weight value and the kernel function corresponding to each nonlinear feature. According to the fusion features obtained by the invention, multiple features can be mapped to a nonlinear feature projection space, the optimal combination of the multiple features is realized, and the calculation complexityof big data is simplified.

The invention relates to a crime hot spot prediction method and a system, and relates to the field of public safety. This scheme solves the technical problem of how to find the collaborative crime indifferent regions, and is suitable for predicting the crime probability in different regions. The invention comprises the following steps: S1, acquiring source data, and generating a plurality of kdehot spots according to the source data; S2, creating a convolution neural network model and training that convolution neural network model accord to the kde hot spot graph; S3: using the trained convolution neural network model to extract the common features of the kde hot spots in different regions, and obtaining the cnn hot spots which contain the information of collaborative crime. The invention is suitable for predicting the crime probability of each region.

According to the multi-source heterogeneous data processing system and method based on the industrial system provided by the invention, a part of calculation tasks (data cleaning, screening, encryption processing and the like are performed on the preprocessed data) are completed by adopting the edge calculation module, so that the calculation pressure of a cloud data center can be effectively relieved; meanwhile, heterogeneous data are coded in a multi-path parallel mode to form a unified identifier, subsequent calculation is facilitated, the processing speed can be improved, and furthermore, data screening is performed based on unified coding by adopting an edge calculation module, so that the data storage expenditure of a cloud data center can be greatly saved. Meanwhile, data screening of the edge computing module is also an efficient data cleaning mode, the computing burden of the cloud data center can be relieved, and in addition, the timeliness requirement of abnormal alarm can be met by detecting vulnerability data in real time and directly uploading the vulnerability data to the cloud data center.

The invention discloses a semi-analytical method for predicting the performance of a deep-layer butt-joint buried heat exchanger. Based on basic rules of convection and heat conduction in heat transfer science, an energy conservation and iterative calculation method, and by utilizing structural parameters, operating parameters, rock-soil and aquifer physical property parameters and the like of theburied heat exchanger in actual engineering, rapid prediction of the performance of the buried heat exchanger in deep butt joint is achieved, so that data such as outlet water temperature, heat taking amount and surrounding rock-soil temperature distribution of the heat exchanger under different operating durations are obtained. According to the method, the heat exchange performance and the heatsupply capacity of the heat exchanger can be integrally evaluated according to geological conditions and heat exchanger parameters before an actual geothermal energy heat supply project is developed,so that a theoretical basis and data support are provided for development of the actual project. The method can be implemented before engineering development, and has the advantages of easiness in implementation and quickness in calculation.

The invention provides an earthquake side slope permanent displacement calculation method, belongs to the technical field of earthquake side slope calculation, and an improved Newmark model is adoptedto perform mechanical analysis, select proper earthquake motion time history data, acquire sliding surface shear strength parameters through comparative analysis, and perform data processing. Contribution percentages of different factors such as a sliding surface mechanical parameter, an inertial reference system and vertical seismic oscillation to the permanent displacement of the slope and a friction coefficient dynamic equation determined by different water content conditions of a slope body soft zone are considered to jointly establish an acceleration function of the slope along the downward inclination direction of the sliding surface; and the acceleration is subjected to quadratic integration in the whole gliding process time domain to obtain the permanent displacement, so that thecalculation of the slope displacement is more accurate and faster, and a foundation is laid for establishing a permanent displacement prediction model to carry out regional risk quantitative evaluation in slope engineering.

The invention relates to an obstacle detection method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring point cloud data of a detection area, and processing the point cloud data to obtain preprocessed data; establishing a coordinate system, and converting the preprocessed data into coordinate data corresponding to the coordinate system; dividing the detection area into a plurality of grids, judging whether the grids are occupied by an object, if the grids are occupied by the object, judging that the grids are object occupied grids, adding all the grids to obtain a multi-dimensional grid, traversing the multi-dimensional grid, and obtaining object information of the object; constructing a prediction model according to the object information, and predicting the object to obtain state information of the object; and constructing a threat field according to the state information, calculating a threat coefficient, judging whether the object is about to threaten, if so, judging the object as an obstacle, and giving out an early warning. The method can detect obstacles in multiple aspects, and compared with a traditional detection technology using a laser radar, the method has the advantage that the detection capability is remarkably improved.

The invention discloses a toilet flushing method, device and a system. The method comprises the following steps: S1, wakening up a controller by a human body sensing sensor after a human body leaves ;S2, controlling the image sensor to collect the urinal image; S3, detecting that largest dirt spot from the urinal image, and extract the coordinate value of the center point of the largest dirt spotas the target coordinate; S4, calculating that rotation angle of the steering gear which drive the spray nozzle to rotate in the plane constructed by the virtual two-dimensional coordinate system andthe rotation angle of the stepping motor which drive the spray nozzle and the steering gear to rotate around the horizontal axis of the virtual two-dimensional coordinate system according to the target coordinates, so that the falling point of the spray nozzle water column is located in the target coordinates; S5, record that accumulative flushing time; and S6, flushing is over. In the process offlushing water, the nozzle follows the largest dirt spot in the urinal image to carry out fixed-point flushing, and the falling point of the nozzle water column is located at the midpoint of the largest dirt spot through a stepping motor and a steering gear, so that the decontamination effect is good, the water consumption is low, the power consumption of the instrument is low, and the operationcost is low, which can be popularized on a large scale.

The invention relates to a sound recognition-based automobile engine fault judgment method and its device, which belong to the technical field of engine fault detection and solve the problems of inaccurate fault autonomous judgment in the prior art, inability to use engine operating state information in the judgment process, and re-faults required for subsequent maintenance. issue of judgment. The automotive engine fault judgment method disclosed in the present invention adopts a hybrid neural network, performs fault recognition according to AlexNet, and uses LSTM to identify engine running time status to assist more accurate fault judgment, and the layers of the AlexNet and LSTM neural networks used are more than 9 layers , it has been verified by experiments that its recognition accuracy is high, and the judgment result is very accurate. In actual application, the automobile engine fault judging method of the present invention can simply, quickly and reliably judge the common faults that occur in the automobile engine, eliminate them in time, and do not need to re-judgment the fault for subsequent maintenance.

The multi-source heterogeneous data processing system and method based on an industrial system provided by the present invention can effectively alleviate cloud data by using an edge computing module to complete a part of computing tasks (data cleaning, screening and encryption processing for preprocessed data, etc.). At the same time, the heterogeneous data is encoded in a multi-channel parallel way to form a unified identification, which is convenient for subsequent calculation and can improve the processing speed. Furthermore, the edge computing module is used to screen data based on unified encoding, which can greatly save cloud data. The data storage expenses of the center, and the data screening of the edge computing module is also an efficient data cleaning method, which can reduce the computing burden of the cloud data center. In addition, real-time detection of vulnerability data and uploading directly to the cloud data center can also meet the requirements of abnormal alarms. timeliness requirements.

The present application provides a multi-feature fusion method, device, electronic equipment, and storage medium, including: acquiring a plurality of nonlinear features of image data of a target object; inputting each of the nonlinear features into a preset multi-kernel function, respectively A feature weight value corresponding to each of the nonlinear features is obtained; a fusion feature of each of the nonlinear features is obtained according to a weighted sum of each of the feature weight values ​​and a kernel function corresponding to each of the nonlinear features. The fusion features obtained in this application can realize the mapping of multiple features to the nonlinear feature projection space, realize the optimal combination of multiple features, and simplify the computational complexity of big data.

The embodiment of the invention discloses an orbit error information generation method and device, electronic equipment and a readable medium. According to one specific embodiment, the method comprises the steps: acquiring a single-view complex image being an image generated after echoes are acquired by a synthetic aperture radar; generating a defocus complex image based on the single-view complex image; performing aperture division on the defocused complex image along the azimuth to generate a sub-image set; generating frequency modulation slope error information corresponding to each sub-image in the sub-image set to obtain a frequency modulation slope error information set; determining satellite acceleration information corresponding to each piece of frequency modulation slope error information in the frequency modulation slope error information set to obtain a satellite acceleration information set; and generating satellite orbit error information according to the satellite acceleration information set. According to the implementation mode, the method does not need to depend on satellite positioning data acquired by a plurality of observation stations to determine the satellite trajectory error. And the data processing amount is greatly reduced.

The invention discloses a semi-analytical method for predicting the performance of a deep-layer butt-joint buried heat exchanger. Based on basic rules of convection and heat conduction in heat transfer science, an energy conservation and iterative calculation method, and by utilizing structural parameters, operating parameters, rock-soil and aquifer physical property parameters and the like of theburied heat exchanger in actual engineering, rapid prediction of the performance of the buried heat exchanger in deep butt joint is achieved, so that data such as outlet water temperature, heat taking amount and surrounding rock-soil temperature distribution of the heat exchanger under different operating durations are obtained. According to the method, the heat exchange performance and the heatsupply capacity of the heat exchanger can be integrally evaluated according to geological conditions and heat exchanger parameters before an actual geothermal energy heat supply project is developed,so that a theoretical basis and data support are provided for development of the actual project. The method can be implemented before engineering development, and has the advantages of easiness in implementation and quickness in calculation.

An embodiment of the invention discloses a division operation method and device, equipment and a storage medium. The method includes: setting numerical values stored in exponent sign bytes in an original dividend and an original divisor to be identical to obtain a current standard dividend and a standard divisor; sequentially selecting numerator quotient estimation bytes and denominator quotient estimation bytes from the current standard dividend and the standard divisor; according to the numerator quotient estimation bytes and denominator quotient estimation bytes, dividing the current standard dividend by the standard divisor to obtain a quotient value, taking a first digit of the quotient value as a current digit quotient value, adding into a temporary quotient according to a sequence from high to low to serve as a last digit; according to numerical values stored in the exponent sign bytes in the original dividend and numerical values stored in the exponent sign bytes in the original divisor, calculating the temporary quotient to obtain a final quotient. By adoption of the method, division operation efficiency can be improved.

The invention relates to the technical field of transportation management, in particular to an intelligent transportation management system and method based on image recognition. The system comprises:a cargo analysis subsystem which is used for obtaining an image A of a vehicle contour with a reference mark, obtaining a preset reference length of the reference mark, and obtaining the mark lengthof the reference mark according to the image A and the reference length, and is also used for acquiring theoretical cargo carrying information and calculating the actual cargo carrying volume according to the marked length and the theoretical cargo carrying information; and an auxiliary analysis subsystem which is used for acquiring the standard auxiliary distance, the actual auxiliary distance and the loading material, and generating an auxiliary analysis conclusion according to the standard auxiliary distance, the actual auxiliary distance, the loading material and the actual loading volume.By the adoption of the scheme, effective statistics can be conducted on the actual loading volume of the vehicle at a time according to the reference marks, the actual loading volume is checked through downward movement of the carriage, and therefore the situation that another building material is hidden under the building material is avoided, and the transportation situation is effectively supervised.

The invention provides a large-scale network streaming data cache-write method including that a client side packages collected data and sent the same to a server side; the server side analyzes the received data to determine data resources and unifies data formats; the data with uniform formats are written into cache regions of different inner memories according the data resources; different cache strategies are made according to the data which are written into the cache regions of the different inner memories, and the data meeting strategy trigger conditions are written into a local file system from the inner memories; the data written into the local file system are loaded to a distributed file system or distributed database according to the upper application demand; and to the data which are loaded to the distributed file system or distributed database, small data blocks are combined into large data blocks at fixed time. The large-scale network streaming data cache-write method is applicable to write of large-scale network data that are of different resources and different inflow velocity by the aid of a multi-stage cache mechanism.