32 results about "Slice sampling" patented technology

The invention discloses an engine fault prediction method. The method comprises the following steps that effective measurement data of a plurality of sensors of an engine are obtained; the effective measurement data are subjected to normalization processing; the effective measurement data obtained after normalization processing are sliced, and a plurality of slice samples are formed; the multiple slice samples are sent into a deep nerve network to be subjected to feature extraction, and after feature extraction, the slice samples are connected to a fully-connected network; and according to output of the fully-connected network, classification is conducted through a softmax classifier, and the left service life of the engine is determined. According to the provided engine fault prediction method, the process is small, operation is easy, and the engine fault prediction method can be widely applied to various engines of aircraft.

Mapping of myelin water content in white matter may provide important information for early diagnosis of multiple sclerosis and the detection of white matter abnormality in other diseases. It is disclosed here that free induction decay (FID) of each voxel at multiple slice locations is acquired in the brain using an echo-planar spectroscopic imaging (EPSI) pulse sequence. The multi-slice EPSI acquisition is designed to have a short first echo time (˜2 ms) and echo-spacing (˜1 ms) in order to acquire multiple sampling points during the fast decay of the myelin water signal. Multi-compartment analysis is then applied to the FID in each pixel using a 3-pool model of white matter to obtain quantitative maps of the myelin water fraction. Using this technique, the MR data for whole brain mapping of the myelin water can be acquired in less than 10 minutes, making this technique feasible for routine clinical applications.

The invention discloses a health condition monitoring method of complex equipment based on an attention mechanism and a neural network. The main steps comprise acquiring tge multi-sensor data of complex equipment; obtaining the effective measurement data by feature selection; obtaining a plurality of slice samples by preprocessing; establishing the neural network classification model which combines the attention mechanism and the depth neural network; inputting the slice samples and their corresponding labels into the neural network classification model to train the neural network classification model offline; inputting the slice samples of multi-sensor data to be predicted into the trained neural network classification model to obtain the health status of complex equipment. The method considers the data characteristics of the multi-sensor signal, fully excavates the local characteristics and the time sequence information in the data, has high prediction accuracy and wide applicability, and can be widely applied to various complex equipment.

A method for evaluation of cable aging degradation based on slice sampling is revealed. An insulation layer of a cable is aged or degraded from the outside in under influence of high temperature, liquid immersion, radiation exposure, or other factors. The present invention gets slices as samples of the cable by cutting from the outside in. Then related tests including dielectric withstand test, insulation resistance measurement, elongation test, etc. are carried out. Besides the state of cable aging degradation being of the slice being found out, different positions in the insulation layer such as an outer layer, a middle layer and an inner layer can also be evaluated.

The invention discloses a real-time human face attitude estimation method based on depth video streaming. The real-time human face attitude estimation method comprises two stages of sampling and training and estimating in real time. The sampling and training stage comprises the following steps of obtaining filed depth images of all human face angles; then conducting random sampling on the filed depth images of all angles and obtaining a training sample set; and using a supervised learning method to conduct training, and obtaining a classifier. The real-time estimation stage comprises the following steps of firstly extracting real-time human face filed depth images in the depth video streaming output by collection equipment and converting the real-time human face filed depth images into integral images; conducting random slice sampling on the integral images, utilizing the classifier obtained by training to conduct classification on samples, and obtaining a plurality of estimation results; and removing abnormal results for the results and conducting weighted average, and obtaining final human face attitude results. The real-time human face attitude estimation method based on the depth video streaming avoids influences of factors like illumination on the final results, and is good in instantaneity and precision.

The invention discloses an apparatus and method for detecting objects in a three-dimensional volumetric image. The apparatus includes a slice sampling unit which is configured to sample the three-dimensional volumetric image so as to obtain a plurality of two-dimensional images, a detection unit which is configured to detect two-dimensional regions of interest from each two-dimensional image in the plurality of two-dimensional images which are obtained by the slice sampling unit through sampling, a combination unit which is configured to combine three-dimensional regions of interest through utilizing the two-dimensional regions of interest which are detected by the detection unit, and a segmentation unit which is configured to perform three-dimensional segmentation on the three-dimensional regions of interest so as to extract targeted objects.

The invention provides an electroplating quality detection module, a printed circuit board motherboard and an electroplating quality detection method, wherein the electroplating quality detection module is located on the edge of the printed circuit board motherboard, including : detection hole, metal plating is attached on the hole wall of the detection hole, and the aperture of the detection hole gradually decreases from top to bottom, so that the detection hole is in the shape of a circular platform with a large top and a small bottom, so that the detection hole can pass through the The metal coating on the hole wall is inspected for electroplating quality. Through this technical solution, the electroplating quality of the printed circuit board can be directly judged by detecting the detection hole, and whether there is no copper in the hole can be quickly and accurately judged, which reduces the quality risk of the product and improves the efficiency of electroplating quality inspection , to avoid product scrapping caused by destructive slice sampling, and to reduce production costs.

The invention belongs to the technical field of power distribution network operation control, and particularly relates to a power distribution network voltage stability probability evaluation method based on large-scale wind power and electric automobile grid connection. The method comprises the following steps: (1), inputting relevant data such as actually measured data and the like of a wind power plant; (2), building a probability model of an electric automobile charge load and a probability model of a wind power plant output; (3) sampling the probability models of the electric automobile charging load and the wind power plant output in the step (2) by adopting a slice sampling algorithm, and constructing sample spaces of input random variables; (4) calculating corresponding values in the sample spaces of the input random variables at the moment t by adopting a continuous power flow algorithm to obtain output random variables of a system voltage stability margin at the moment; and (5) obtaining probability distribution characteristics of the output random variables according to a probability statistics method. The method has the effects that the accuracy and effectiveness of voltage stability probability evaluation are greatly improved from the perspective of the sampling algorithm.

The invention discloses a seismic signal regularization processing method based on slice sampling. The method is applied to the field of seismic exploration, and for the problem of surface missing inVSP data, a Low-tubal-rank and Sparse Sensor Complementation (LRSTC) algorithm is provided to complete signal regularization processing; an LRSTC model is solved by using an enhanced Lagrange method;through experimental simulation and analysis on synthetic seismic data and actual seismic data, the excellent performance of the LRSTC is verified.

The invention discloses a lateral erosion, crack and cavity measurement slice manufacturing method. A current fuzzy measurement slice graph boundary problem is effectively solved. The method comprisesthe following steps of (1) slice sampling; (2) rough grinding; (3) polyester resin filling and fine grinding; and (4) metallographic microscope measurement, and also comprises the step of defect areacolor filling after the step (1) or the step (2), wherein the step of defect area color filling includes the following step of coating a defect area with a filler color to distinguish from other material colors so as to enhance a slice defect contrast ratio; and the other material colors include substrate, copper, and solder resist colors, and a surface processing color.

The invention provides a biological microscope structure for plant quarantine, which comprises a top fence and a supporting pipe; the back end face of a main body handle frame is equipped with a square structured connecting base, one side face of the connecting base is equipped with a metal corrugated connecting pipe; the other end of the metal corrugated connecting pipe is equipped with a circular plate shaped slice operation platform; the side face of the connecting base is connected with a spray pipe in a penetrating manner through a matched rotating head, and the side face of the slice operation platform is provided with a first placing slot and a second placing slot from outside to inside; the outer wall at the top side of the main body handle frame is connected with a circular sliceoperation platform by the metal corrugated connecting pipe, and the side face of the slice operation platform is provided with two sample culture glass carriers for mounting cutters and quarantining,thus the original operation mode of performing slice sampling on a workbench is replaced, the space is saved, and the operation space of the original operation platform is saved.

The invention discloses a slice sampling method used for determining a recycled concrete carbonization zone microstructure. The method includes the following steps: (1) molding a recycled concrete test piece, and performing one-dimensional carbonization after demolding and maintenance; (2) performing vertically slicing from an intersection point of a front carbonized surface and an upper non-carbonized surface to the center of the test piece at intervals along the one-dimensional carbonization direction of the test piece so that test piece slices with different carbonization depth can be obtained; and (3) performing sampling, knocking out new and old mortar particles in each test piece slice, and performing carbonized zone microscopic test after performing further crushing and vacuum drying. The method can control slice intervals according to actual carbonization depth, so that new and old hardened cement slurry samples of recycled concrete in different carbonization depth sections canbe obtained; and the method can separately analyze the carbonization depth, the division of carbonization areas and microstructure of the new and old mortar in the recycled concrete, and therefore, reference can be provided for accurate carbonation life prediction of the recycled concrete and other sampling methods for durability studies.

The invention relates to a general anesthesia invasive systolic pressure prediction method and system based on machine learning fusion. The method comprises the following steps: S1, collecting the body feature data and vital sign data of a patient, carrying out the preprocessing of all data, and dividing the preprocessed data into a training set, a test set and a verification set; respectively performing slice sampling on the training set, the test set and the verification set to collect n training subsets, test subsets and verification subsets; s2, constructing a prediction model based on machine learning fusion, and predicting the general anesthesia invasive systolic pressure by using the prediction model; the prediction model comprises n primary learners and a secondary learner, and the output values of all the primary learners are used as the input of the secondary learner. The system comprises a database module, a data collection module, a prediction module and an interaction module. A prediction program is stored in the prediction module. According to the method, the defect of low prediction precision of a single algorithm is overcome, the input of the secondary learner is optimized, and the prediction precision is further improved.

The invention discloses a resolution power test method and system based on ISO12233. The test method comprises the steps of obtaining Raw format sample images of ISO12233 test images; selecting or inputting test block coordinates from the Raw format sample images, determining test blocks, and extracting two-dimensional image data in a coordinate range as image data in the test blocks; carrying outseparation recombination on image data channels in the test blocks and carrying out slicing sampling on the recombined image data channels; and calculating and judging whether all sample slices passor not. According to the method and the system, the rapid and batch test can be carried out on resolution power of camera modules.

The invention relates to a seed locating device. The locating device comprises a push mechanism and a clamping mechanism, wherein the push mechanism is arranged to be used for pushing seeds to a preset slice sampling station along a first moving direction and extending out of the slice sampling station by a preset thickness; the first moving direction is a linear moving direction perpendicular tothe edge of the slice sampling station in a plane on which the slice sampling station is positioned; and the clamping mechanism is arranged to be used for clamping the seeds pushed onto the slice sampling station by the push mechanism from the upper side. The invention further relates to automatic seed slice sampling equipment. In addition to the seed locating device, the automatic seed slice sampling equipment further comprises a sample collection device and a seed collection device.

The invention provides a device and a method for obtaining micro tissue blocks at high throughput, and relates to the technical field of slice sampling. The device comprises a sample table, sampling needles distributed in an array, a flexible assembly for driving the sampling needles to advance or retract, and a three-dimensional translation table. The flexible assembly comprises an air cylinder and a positioning part used for limiting rotation of a piston rod of the air cylinder. The sampling needles are distributed in an array manner, and high-flux sampling can be completed at a time. The flexible assembly drives the sampling needle to flexibly advance for sampling, so that the acting force between the sampling needle and the tissue can be well adjusted. The positioning part ensures thatthe sampling needle does not rotate along with the piston rod of the air cylinder in the whole sampling process, so that the relative position between the sampling needle array and the sample is prevented from deflecting, the success of repeated sampling of similar samples can be ensured, and the purpose of protecting the sampling needle can be achieved.