14results about How to "Improve division accuracy" patented technology

The invention belongs to the technical field of medical imaging, and discloses a tumor area rapid identification method and system based on cortical electroencephalogram signals, and the method comprises the steps: carrying out the preprocessing of collected multi-channel ECoG signals through an interpolation filling strategy; extracting a target frequency band ECoG signal of each channel; extracting a feature sequence of each channel in a given time period as a channel feature; performing clustering analysis on each channel feature to obtain a plurality of categories; and reordering according to the feature level of each category of region, and completing tumor region boundary identification by taking the category of which the feature level meets the requirement as a tumor region. According to the invention, rapid detection and effective identification of tumor boundaries can be realized.

The invention belongs to the field of soil pollution prevention and control planning, and particularly discloses a GIS-based soil pollution prevention and control zoning method, which specifically comprises the following steps: S1, collecting basic data and data; s2, establishing a GI database; s3, dividing survey units; s4, point distribution investigation; s5, soil pollution grading; s6, pollution source load risk assessment; s7, agricultural product sample detection; s8, adjusting the partitions; s9, visualizing the map; according to the method, soil pollution categories are preliminarily divided according to soil detection conditions, then edible agricultural products of the agricultural land are sampled and detected, pollution source load risks of the construction land are evaluated and graded to serve as auxiliary judgment data to carry out secondary partitioning on the soil, the dividing accuracy is improved, and the efficiency is improved. A distribution diagram can be made for each element through the GIS software, and a plurality of diagrams are superposed and synthesized by using the GIS software, so that the soil pollution prevention and treatment classification supervision level can be effectively improved.

This invention discloses a method and application for classifying different water invasion risk levels in carbonate bottom-water gas reservoirs, belonging to the field of oil and gas field development technology. The method includes the following steps: S1, obtaining the relative distribution of gas and water layers in each sublayer of the gas reservoir under study; S2, evaluating the activity of the bottom water body; S3, dividing the block into sealed and unsealed zones; S4, obtaining the distribution of faults, fractures, and high-quality reservoirs in the water invasion channel; S5, based on the bottom water activity obtained in step S2 and the distribution of faults and fractures in the water invasion channel obtained in step S4, classifying the gas reservoir in the study area into five categories: high-risk longitudinal intrusion, low-risk longitudinal intrusion, high-risk transverse intrusion, low-risk transverse intrusion, and low-risk zones. This method is beneficial for the efficient and rational development of bottom-water gas reservoirs, reducing the degree of water invasion damage to the gas reservoir, and improving the final recovery rate of the gas reservoir.

This application relates to the field of tracing the causes of cross-border returns, disclosing a method, apparatus, electronic device, and storage medium for tracing the causes of cross-border returns. The method includes: deeply fusing multilingual text comments and image evidence using joint semantic vector technology; dynamically associating front-end return data with back-end supply chain data through root cause mapping to form a preset root cause map, thereby tracing the causes of returns based on return comment information. The beneficial effects of this invention are: breaking down data silos, accurately identifying deep causal relationships, significantly reducing the root cause misjudgment rate, shortening the root cause location time for cross-border returns, and improving the accuracy of supply chain responsibility allocation, thus achieving precise tracing of the causes of cross-border returns.

The invention provides a training sample making method, a training sample making device and a computer storage medium. The training sample manufacturing method comprises the steps that according to a first video frame and a second video frame which are adjacent, a motion detection frame and an average video frame are generated, and the motion detection frame comprises an unknown area, a motion area and a static area; generating an initial unknown region pixel value according to the first unknown region pixel value of the first video frame and the second unknown region pixel value of the second video frame; establishing a three-dimensional array of pixel values of the initial unknown region according to the average pixel value of the average video frame and the motion pixel value of the motion detection frame; searching a pixel value of a target unknown area according to the three-dimensional array; updating the motion detection frame based on the target unknown region pixel value; and forming a training sample by using the updated motion detection frame, the first video frame and the second video frame. Through the training sample manufacturing method, high-quality training samples can be provided for the motion detection network.

The invention discloses a message control method, a product, equipment and a medium, and relates to the technical field of information propagation control. According to the scheme, the concept of the gravitation value is defined for each node of the social network; the gravitation value represents an information propagation association degree between the two nodes; according to the method, the corresponding clustering label is updated according to the gravitation value of each node, each node in the social network can be divided into a plurality of sub-networks according to the clustering label, and finally message control is performed on each sub-network, so that the calculation complexity is greatly reduced. Meanwhile, due to the fact that the split sub-networks are mutually independent, the different sub-networks can concurrently and synchronously execute related operations such as negative message control, concurrency of the system is improved, the execution period of the algorithm is shortened, expandability of the algorithm is enhanced, and the method is suitable for a large-scale social network.

The present invention discloses an optical property modeling method and device, which includes: dividing a microstructure in a periodic medium into N layers of thin slices in the z direction; performing the following processing for thin slices of the same layer of microstructure: obtaining each closed area of ​​the thin slice projected on the periodic space xy plane, and representing the boundaries of each closed area as each polygon; obtaining the x coordinate and y coordinate of each polygon vertex, and the line segment features of each polygon, dividing the periodic space according to the line segment features, and obtaining the integral sub-intervals divided in the x direction and the integral sub-intervals divided in the y direction; calculating the Fourier coefficients and Toeplitz matrix of the dielectric coefficient of the periodic medium based on the integral sub-intervals in the x and y directions, and performing strict coupled wave analysis to achieve optical property modeling of the periodic medium. The method optimizes the division method of the integral interval, and efficiently and accurately obtains the Fourier coefficients and Toeplitz matrix of the dielectric coefficient of the periodic medium.

The invention discloses a real-time cough sound counting method, and relates to the technical field of audio recognition, and the method comprises the steps: determining a user side, collecting and processing the audio of the user side, and obtaining a plurality of target audios; extracting the plurality of target audios based on a neural network algorithm to obtain cough audios of the user side, and performing quantity statistics on the cough audios; and transmitting the quantity of the cough audios to a server as use data of the user. According to the system, the accuracy of cough sound recognition and extraction is improved, analysis data of the physical state of a user side person is better provided, the difficulty of collecting cough data of a patient suffering from chronic respiratory diseases is reduced by collecting the cough information of the patient through the system, and the accuracy and timeliness of collecting the cough data are guaranteed.

The invention relates to a similarity community clustering method and system capable of defining targets, and belongs to the field of computer networks. The clustering method comprises the steps that a user customizes a similarity measurement method and a threshold value, constructs a fully connected graph, calculates the number of expected communities, aggregates attributes through a neural network, iteratively updates a loss function, and outputs a division result. The system framework comprises a data input module, a user definition module, a community trial calculation module, a community clustering module and a community output module. According to the method, the target is customized by the user, the number of communities does not need to be specified in advance, the privacy disclosure risk can be reduced, and the community clustering accuracy can be improved.

The present invention relates to the technical field of electronic digital data processing, and specifically discloses a parallel tensor filling method based on a GPU. The method comprises: initial matrix generation, tensor division accuracy determination, sub-tensor parallel processing and parallel tensor filling, wherein the initial matrix is ​​generated according to a matrix method, sub-tensors are obtained by a preset tensor division method based on the data amount of a data set to be filled, and the division accuracy is verified to optimize the division method; sub-tensor binary coordinates are obtained by a predefined coordinate positioning method, a parallel sub-tensor sequence is obtained by combining a parallel tensor search method, and the sub-tensor sequence is transmitted to a GPU memory space for parallel processing, the GPU parallel processing performance is evaluated, the weight of each sub-tensor sequence is obtained by parallel processing, the initial matrix is ​​updated to an update matrix, and finally the missing data is filled in a predefined method, the parallel tensor filling of the data is completed, and the accuracy and integrity of the data are improved.

The invention discloses a fully mechanized coal mining face goaf coal spontaneous combustion three-zone detection method, a fully mechanized coal mining face comprises a goaf, a working face, an open-off cut, an air inlet way and an air return way, and the detection method comprises the steps that S1, a plurality of drilling channels are drilled from the ground to the goaf, and a plurality of drilling terminals are mutually spaced in the inclined length direction of the fully mechanized coal mining face; s2, a plugging assembly and an exhaust pipe are arranged in the drilling channel, the plugging assembly plugs the opening of the starting end of the drilling hole to form a detection cavity, one end of the exhaust pipe penetrates through the plugging assembly and extends into the detection cavity, and the other end of the exhaust pipe is connected with a gas sampling device; s3, regularly sampling the gas in each detection cavity by using the sampling device to obtain an oxygen concentration value in the detection cavity, and stopping sampling until the detected oxygen concentration value is lower than a preset oxygen concentration threshold value; and S4, determining the distribution range of the spontaneous combustion three zones of the goaf according to a corresponding change rule between the detected oxygen concentration value in the detection cavity and the distance of the working face.

The present invention relates to a roadside hazard level classification method and classification system based on slope stability and roadside environment. The purpose of the present invention is to solve the problem that the existing roadside hazard level classification has a single classification basis and does not consider the influence of factors such as the geological conditions of the roadside slope, resulting in low accuracy of roadside hazard level classification. The process is: 1: obtain basic information on the roadside; 2: obtain the strength reduction coefficient; 3: construct the weight of each factor; 4: construct an evaluation matrix for each factor; 5: calculate the fuzzy comprehensive evaluation set based on the weight of each factor and the evaluation matrix of each factor, calculate the comprehensive evaluation value based on the fuzzy comprehensive evaluation set, and determine the roadside hazard level based on the comprehensive evaluation value. The roadside hazard level classification system based on slope stability and roadside environment is used to execute the roadside hazard level classification method based on slope stability and roadside environment. The present invention belongs to the field of road traffic safety design.

The invention relates to the technical field of radar, in particular to a data fusion method and device, and the method comprises the following steps: S1, completing lane division through employing a lane calibration automatic division method, and obtaining a lane division result; and S2, according to the lane division result, fusing the collected vehicle target data by using a split target fusion method to obtain fused vehicle target information, and completing the statistics of the traffic flow and the vehicle speed, thereby effectively solving the problem of large truck target splitting, and improving the vehicle counting accuracy.

The invention provides an urban gas logic pipeline division method and device, a medium and equipment. The method comprises the following steps: acquiring an urban gas physical pipeline distribution map based on a geographic information system; providing a node selection page to enable a user to select a starting point and an ending point on the urban gas physical pipeline distribution map on the node selection page; taking a physical pipeline communicated between the starting point and the ending point selected by the user as a logic main line; acquiring a preset urban gas logic pipeline division rule; automatically generating a logic branch line connected with the logic main line according to the urban gas logic pipeline division rule and each equipment node on the physical pipeline corresponding to the logic main line in the urban gas physical pipeline distribution diagram; and according to the logic main line and each logic branch line connected with the logic main line, constructing the logic pipeline network of the urban gas. According to the invention, the accuracy and efficiency of logic pipeline division can be improved, and the labor cost is saved.