15results about How to "Detailed description" patented technology

The present invention relates to the technical field of sleep diagnosis, and in particular to a sleep staging method based on deep learning, the method comprising: collecting a number of target physiological signals, preprocessing each target physiological signal to obtain a number of standard physiological signals; synchronously aligning and integrating each standard physiological signal in turn to obtain a standard sleep data set; determining the actual correlation between each physiological signal to be integrated in the standard sleep data set, and determining an actual signal combination strategy according to the actual correlation; generating an actual sleep data set according to the actual signal combination strategy, and extracting features from the actual sleep data set to generate actual feature parameters; inputting the actual feature parameters into a deep learning model to determine the actual sleep state; analyzing the actual sleep state, and performing corresponding sleep intervention according to the analysis results. The present invention realizes low-cost non-intrusive sleep monitoring, thereby improving people's physical health.

The present application relates to a kind of wide remote sensing image semantic description method and system based on adaptive region segmentation, belong to the cross-application technical field of natural language processing and computer vision.The present application first for wide remote sensing image, the semantic distribution characteristics of image are analyzed, semantic estimation point is generated, clustering is carried out to semantics, and image is segmented according to cluster center, to obtain target sub-region.For target sub-region, a small scale remote sensing image description model is used to generate the description of corresponding region, and multiple single sentence descriptions are merged according to their clustering relationship, to generate complete, clear final description.

The present application provides an asphalt pavement apparent disease evolution deduction method, system, device and medium, which belongs to the field of road engineering, transportation infrastructure operation and maintenance and intelligent detection technology, and comprises the following steps: constructing a unified space-time reference system, aligning and normalizing multi-source data in space-time; forming section-level disease quantification characteristics based on the recognition results of the inspection images and the structural indexes; event coding the maintenance measures of the road surface and pre-processing the section-level disease quantification characteristics corresponding to the time of the maintenance measures; constructing the correlation, time lag effect and spatial consistency between the disease types, disease characteristics and structural indexes to generate the collaborative evolution characteristics of the coupling relationship between diseases; constructing a deep time series prediction model to obtain the disease development trend at a specified time scale in the future, and forming the section-level disease evolution prediction result. The present application can depict the disease development law at different road sections and different time scales, and predict the future disease expansion.

The invention provides a continental facies shale oil lithofacies type identification method and device, and the method comprises the steps: carrying out the preprocessing and image enhancement processing of the electric imaging logging original data of a to-be-identified lithofacies, and obtaining a first electric imaging static image, carrying out the histogram processing of the first electric imaging static image, and obtaining a first electric imaging static image histogram; determining stratum structure indexes and lithofacies discrimination parameters of the to-be-identified lithofacies, which are used for reflecting stratum structure characteristics, on the first electric imaging static image histogram; and comparing the stratigraphic structure index and the lithofacies discrimination parameter of the to-be-identified lithofacies with preset stratigraphic structure indexes and lithofacies discrimination parameter limit standards of various shale oil and lithofacies types, and determining the shale oil and lithofacies type. According to the method, the electrical characteristics of the stratum can be objectively and truly reflected, the fine structure characteristics of the stratum can be described more finely and efficiently, and the continental facies shale oil facies type can be accurately identified.

The invention discloses a tooth grinding tooth surface roughness measuring method based on machine vision, and relates to the technical field of tooth grinding processing quality detection.The tooth grinding tooth surface roughness measuring method comprises the steps that graying processing is conducted on a multi-angle tooth surface image, a multi-angle tooth surface grayscale image is obtained, and each pixel point in the multi-angle tooth surface grayscale image is divided into a first feature point and a non-feature point; carrying out region clustering on each non-feature point to obtain a plurality of local regions, and carrying out adaptive Gaussian filtering operation on the multi-angle tooth surface grayscale image; feature point extraction is carried out on each angle tooth surface gray scale image in the multi-angle tooth surface gray scale image after noise removal, then feature point matching is carried out on each angle tooth surface gray scale image, each angle tooth surface gray scale image is spliced, and a complete tooth surface gray scale image is generated; the tooth surface roughness measurement model is constructed, texture feature extraction is performed on the complete tooth surface gray level image, the tooth surface roughness value of the complete tooth surface gray level image is obtained, and the accuracy and efficiency of tooth surface roughness value measurement are remarkably improved.

The present invention relates to a simplified fastener information parsing method, comprising: S1. fastener information extraction, S2. fastener information merging, S3. fastener information structured expression, S4. fastener information parsing and reconstruction. In this way, the simplified fastener information in the R model can be extracted, parsed and reconstructed, so that the PDM system can quickly, completely and accurately identify and realize the automatic export of the fastener list. Designers no longer need to maintain manually, saving a lot of manpower costs; process personnel can describe the fastener process in detail and accurately when reconstructing PBOM, MBOM and AO design, simplifying the operation difficulty of workshop workers; suppliers can quickly and accurately obtain fastener material quota information, which is conducive to more accurate evaluation of material procurement needs.

The present application relates to a kind of frequency domain-space domain fusion perception methods for 2.5D integrated circuit fast high-precision thermal prediction, the network includes frequency space thermal encoder module, frequency domain cross-scale interaction module, frequency domain-space domain hybrid loss function (FSL) and model input and pre-processing network (PPNet).Frequency space thermal encoder module combines adaptive multi-frequency embedding and 3D convolutional neural network, extracts high-to-low frequency thermal dissipation gradient feature and detailed spatial feature;Frequency domain cross-scale interaction module realizes multi-scale feature interaction through cross attention mechanism, solves semantic gap problem;Frequency domain-space domain hybrid loss function loss function combines spatial mean square error and frequency domain loss, suppresses high-frequency thermal gradient noise.The present application first realizes the thermal prediction of frequency domain-space domain dual-domain perception, significantly improves prediction accuracy and speed, and has excellent generalization ability, and can be widely applied to the thermal management field of 2.5D integrated circuit.

The invention provides a sodium-water reaction accident simulation method and system for a sodium-cooled fast reactor, and the method comprises the steps: constructing a sodium-water reaction model which comprises a tiny sodium-water reaction model and a large sodium-water reaction model; a sodium-water reaction simulation instruction is obtained, a target sodium-water reaction model is determined based on a reaction type identifier in the sodium-water reaction simulation instruction, and the target sodium-water reaction model is the small sodium-water reaction model or the large sodium-water reaction model; simulating a sodium-water reaction accident based on the target sodium-water reaction model to obtain a sodium-water reaction simulation result; and inputting the sodium-water reaction simulation result into a full-reactor simulation model to obtain a full-reactor accident response result. According to the invention, the accuracy of the sodium-water reaction simulation result and the whole reactor accident response result is improved.

The invention discloses a three-dimensional multi-time scale load prediction method based on transverse and longitudinal multi-dimensional check, and relates to the technical field of power load prediction, and the method comprises the steps: collecting historical load data and environment data of a power grid, and dividing the historical load data into a short-term load data set, a middle-term load data set and a long-term load data set according to multi-time scale division; under each time scale, carrying out load characteristic analysis based on a transverse and longitudinal multi-dimensional checking method to obtain a load change rule under each time scale; and according to the load characteristic analysis result of each time scale, constructing a multi-dimensional load prediction model to predict the future load. Through hierarchical analysis and comprehensive modeling of load characteristics of multiple time scales, dynamic change characteristics of the load are accurately captured, prediction precision is improved, scientific support is provided for short-term scheduling and medium-and-long-term planning of a power system, power resource distribution is optimized, and the overall efficiency and stability of the system are improved.

The invention provides a thick layered bottom water common heavy oil reservoir well pattern structure adjusting method, which comprises the following steps of: according to the oil saturation of remaining oil in a reservoir and a well logging interpretation result of an oil well, judging a water content rising reason of the reservoir in a well area and the enrichment condition of the remaining oil among the oil wells; theoretical water cone radiuses of different depths in the longitudinal direction of each vertical well and theoretical water ridge radiuses of different depths in the longitudinal direction of each horizontal well in the well area are calculated according to the oil saturation of the remaining oil; making a longitudinal water logging dynamic profile map of the oil well in the well area by adopting an oil reservoir dynamic analysis method, and determining oil-water distribution in a remaining oil enrichment area by applying numerical simulation so as to determine the type of remaining oil; according to the type of the remaining oil, well pattern parameters are determined through numerical simulation, and well pattern structure adjustment is carried out according to the well pattern parameters; and horizontal well production parameters are adjusted according to the determined well pattern parameters. According to the method, the remaining oil distribution can be described more accurately and finely, the water content rising speed of the block is effectively slowed down, and the overall reserve utilization degree is improved.

This invention discloses a contamination judgment method, a contamination judgment system, and a cleaning method for a vehicle-mounted camera. The contamination judgment method includes the following steps: allowing a vehicle-mounted camera to capture a poster containing scene information through contaminated specimens of different gray levels to obtain a series of specimen images; converting the original image data of the series of specimen images into HSV format image data; extracting the mean and variance of the HSV color space of the HSV format image data of the series of specimen images as features and inputting them into a support vector machine algorithm model to train a classification algorithm model; inputting the image data of a detection image captured by the vehicle-mounted camera as input features into the classification algorithm model to allow the classification algorithm model to classify the detection image captured by the vehicle-mounted camera; and determining whether the lens surface of the vehicle-mounted camera is contaminated based on the classification result of the classification algorithm model.

The invention relates to an emotion evaluation method based on a 12-dimensional emotion model. The emotion evaluation method comprises the following steps of: acquiring vibration image data of head reflection movement of a tested person by utilizing a vibration image acquisition device; preprocessing the vibration image data, wherein the preprocessing comprises but is not limited to noise removal and feature information extraction; 12 emotion dimension parameters are calculated based on the preprocessed vibration image data; based on the parameters of the 12 emotion dimensions, an emotion evaluation report is generated, and the emotion evaluation report at least comprises the score of each emotion dimension and the description of the overall emotion state. According to the emotion evaluation method based on the 12-dimensional emotion model, the emotion state of the testee can be objectively and accurately recognized and quantified, the limitation that a traditional emotion evaluation method is high in subjectivity and difficult to quantify is overcome, the emotion state can be described more comprehensively and meticulously, and richer information is provided for emotion recognition and intervention.

The invention relates to a locking mechanism provided and adapted to interact with at least one locking element to establish a lock between the locking mechanism and the locking element, wherein the locking mechanism comprises a first locking pawl and a second locking pawl, which are independently rotatable from each other about a first rotation axis, wherein the first locking pawl is arranged and adapted to interact with a first locking element, and wherein the second locking pawl is arranged and adapted to interact with a second locking element, wherein a secure lock exists when the first locking element is locked to the first locking pawl and the second locking element is locked to the second locking pawl.

The invention provides an asphalt pavement apparent disease evolution deduction method, system, equipment and medium, and belongs to the technical field of road engineering, traffic infrastructure operation and maintenance and intelligent detection, and the method comprises the steps: constructing a unified space-time reference system, and carrying out the space-time alignment and normalization of multi-source data; based on an identification result of the inspection image, combining the structural indexes to form section-level disease quantitative features; carrying out event coding on the maintenance measures of the pavement and carrying out feature preprocessing on the section-level disease quantitative features at the moments corresponding to the maintenance measures; constructing correlation among disease types, disease characteristics and structural indexes, and generating co-evolution characteristics of a coupling relationship among the diseases according to a time-lag effect and space consistency; and the constructed depth time sequence prediction model obtains a disease development trend of a specified time scale in the future, and forms a section-level disease evolution prediction result. According to the method, the disease development rule can be described in different road sections and different time scales, and future disease extension is predicted.

The invention belongs to the technical field of transformer winding defect detection, and particularly relates to a transformer winding thermal defect detection method based on Hypergrid-XGBoost. Comprising the following steps: S1, building a simulation model of the transformer and a winding on a Comsol platform according to physical parameters of a target oil immersed transformer, and verifying the effectiveness of the simulation model through a temperature rise test; s2, arranging optical fiber sensor nodes, performing abnormal value test on data transmitted by a sensor by using a Hypergrid strategy, and constructing a winding actual measurement temperature rise matrix; and S3, establishing a temperature rise matrix of the winding in a normal state, and determining the position of the thermal defect of the winding. S4, establishing an overheat point temperature rise model by using an XGBoost algorithm based on the temperature data, and determining the severity of the winding thermal defect; according to the invention, Hypergrid abnormal value detection and an XGBoost model learning data processing technology are applied to an actual scene of winding thermal defect detection, high-precision positioning of a transformer overheat point is realized, the severity of temperature rise of the overheat point is identified, and a direct basis is provided for formulation and implementation of transformer protection measures.