534 results about "Sample number" patented technology

Methods and apparatus are provided for analysis and correlation of phenotypic and genotypic information for a high throughput sample on a cell by cell basis. Cells are isolated and sequentially analyzed for phenotypic information and genotypic information which is then correlated. Methods for correlating the phenotype-genotype information of a sample population can be performed on a continuous flow sample within a microfluidic channel network or alternatively on a sample preloaded into a nano-well array chip. The methods for performing the phenotype-genotype analysis and correlation are scalable for samples numbering in the hundreds of cells to thousands of cells up to the tens and hundreds of thousand cells.

Systems and methods for using sample numbers to pair timed metadata samples with media or hint samples is provided. A timed metadata sample can be paired with media or hint samples since a sample number contained in the time media sample is provided relative to the appropriate media or hint track. Additionally, an offset of sample numbers, applicable to scenarios where a plurality of timed metadata samples exist, may be added to the provided sample number to obtain the actual sample number within the media or hint track.

The invention discloses a target space knowledge and two-stage prediction learning-based target detection method. By utilizing various data conversion methods, a sample number is increased and samplediversity is improved; two deep neural networks including SSD and newly designed RefineNet are trained; for a prediction target with a relatively high probability in a primary prediction result of theSSD, the accuracy of judgment is further improved through the RefineNet; and by establishing peculiar spatial structure constraint rules of the target, the wrong prediction is reduced, thereby obtaining a final detection result. Compared with a few existing methods, the method provided by the invention has the advantages that visual and spatial characteristics of a remote sensing target are considered at the same time; and end-to-end target candidate selection, feature extraction and classified locating are realized by utilizing the deep networks with excellent feature extraction capabilities, so that the detection rate of the remote sensing target is remarkably increased and the false alarm rate is reduced.

The invention belongs to the technical field of image processing, discloses a method for identifying and positioning power transmission line insulators in unmanned aerial vehicle aerial images, and solves the problems in the prior art that the detection precision of an identification algorithm of the insulators is not high, the robustness is low, and the identification algorithm is easy to be affected by sample number. A group of Gabor wavelet basis with different sizes and different directions and training sample images are taken as convolutions so as to form a group of characteristic vectors which accurately describe sample image texture characteristics. A random forest machine learning algorithm with a semi-supervised learning mode is used to train sample data sets of the known category and the unknown category so as to obtain an insulator identification model. Through the mode from left to right and from top to bottom, a detection window with the same size as the training sample traverses the input images with different sizes. The detection window combining the identification model detects and positions the positions of the insulators in the input images with different sizes. And finally the accurate positions of the insulators in the input image with the original size are determined by using a non-maximum inhibition method.

A method for audio/video synchronization and an apparatus thereof are provided. At least one dynamic synchronization checkpoint is disposed on the playback sample number axis of a video. Whether or not the audio and the video are synchronized is checked when the playback sample number of the video is set the dynamic synchronization checkpoint. The video is paused playing if the playback speed of the video is too fast, and those delayed playback sample number of the video having not been played are skipped if the playback speed of the video is too slow. Furthermore, the location of the dynamic synchronization checkpoint and the frequency of performing synchronization detection are adjusted according to the playback time jitter between the audio and the video.

A fraud score calculating program primarily for use in determining the possibility of credit card fraud can calculate a score reflecting the reliability of a model created based on Bayesian theory. A model which is stored in a fraud detection model database 120 obtains new authorization data and continues learning as the number of data samples increases. Calculation of the score is performed by a calculation logic provided in a scoring subsystem 110. The sample number data for a case corresponding to the authorization data are obtained from the model, and the probability of the occurrence of fraudulent use is calculated. The reliability of the model is also calculated on the basis of, for example, the number of the registered samples, and a fraud score is calculated using both the calculated probability of the occurrence of fraud and the calculated reliability of the model.

The invention discloses an interference subspace tracking-based high-efficiency self-adaptive monopulse angle measurement method. According to the invention, for the massive computation and the independent and identically distributed training sample number for covariance matrix estimation of the self-adaptive monopulse technology in the high-dimensional system, the MPASTd method is applied to the self-adaptive monopulse to carry out recursive estimation of the interference subspace; a new self-adaptive sum and difference beam weight is calculated; and a signal source space angle is obtained by using an angle estimation formula. With the method, the computation load is obviously reduced and the convergence speed is accelerated; and the good angle measurement precision is realized under the circumstances of small samples.

The invention discloses a human body tumble detection method and a human body tumble detection system based on big data. The human body tumble detection method is characterized in that characteristic vectors can be established according to information acquired by cell phone built-in sensors; a machine learning model can be used to identify whether the tumble occurs; when the user is in the tumbling state, the acquired information data can be transmitted to a big data platform in a real-time manner, and can be stored according to individual similarities; the platform is used to analyze the uploaded data by adopting the similarity metric algorithm, and is used to determine whether the updated data exists; when the updated data exists, the new data sample can be generated by the platform, and at the same time, a new machine learning classification model can be generated according to the data sample. When the tumbling state of the human body is determined by the system, the cell phone is used to trigger an alarming system automatically, and therefore the tumbling user can be rescued timely, and at the same time, the accuracy of the system can be increased continuously with the increasing sample number. The human body tumble detection method and the human body tumble detection system can be used for monitoring the activity safety of the children, the elderly, and the patients.

The present invention discloses a classification model training method and a device, belonging to the technical field of data processing. The method comprise a step of carrying out model training according to a current round of positive example samples and a current round of negative example samples and obtaining a current round of classification model, a step of using the current round of classification model to classify all samples if the current round of classification model does not satisfies a specified condition and selecting a specified sample in all samples according to a classification result, a step of taking the current round of positive example samples and the specified samples as a next round of positive example samples and determining a next round of negative example samples according to the next round of positive example samples, and a step of continuously executing the above model training and sample processing process according to the next round of positive example samples and the next round of negative example samples until the classification model which satisfies the specified condition is obtained. With the increase of the positive example sample number, the potential positive example samples in the negative example samples decrease, the purity of the negative example samples can be effectively improved, the stability of model obtained through training according to superimposed number of positive example samples and negative example samples, and the classification accuracy is high.

A method for measuring particle diameter distribution includes placing particle samples in glass sample tank set on back light plate uniformly with particle sample number of 50 - 500, placing size calibration object in the tank, using digital camera to shoot image of said particle sample, guiding image to computer for processing the image and obtaining roundness and diameter of each particle, outputting particle sample distribution results of number, volume and weight according to division of particle diameter.

The invention provides a system and method for determining a state of a user at least through utilization of self-stated data and biological measurement data. The method comprises the steps of receiving first biological measurement data of the user; receiving first self-stated data from the user, wherein the self-stated data comprises the state of the user; and storing the data to a memory, thereby enabling the first biological measurement data to be linked with the first self-stated data on relationship and time. The data can be linked on the relationship through utilization of a single fileand/or at least one public ID number. The data can be linked on the time through utilization of at least a time stamp, offset, a sample rate and/or the sample number. The method also comprises the steps of receiving second biological measurement data, determining the state of the user through utilization of the second biological measurement data and the data stored in the memory device, and performing at least one action in response to the state.

The invention discloses a supervised snore source identifying method. The method includes data pretreatment, training and identification; the main steps include, firstly, performing Mel frequency conversion on actual measured snore data, and acquiring a data sample; secondly, arranging a structure of a convolution neutral network, quantity of output feature drawings of a convolution layer and convolution core size, pooling size, weight vector updating studying rate, batch training sample number, and training iterations; thirdly, using a snore time frequency spectrogram of a training set as the convolution neutral network to input; performing network initialization on the arranged network structure; completing the training process through forward process, direction bias transmission, weight value updating and offset until the appointed iterations; at last, delivering the testing set to the trained network model and acquiring the identifying result. The supervised snore source identifying method can effectively identify the snore source, and is exact in identifying result and good in performance.

The invention discloses a tire rubber carbon black dispersity evaluation method and a system based on neural network image processing. The method comprises the following steps: S1, rubber material images in a sample set are acquired, the rubber material images are pre-processed, and feature data are extracted; S2, feature data of a part of rubber material images in the sample set are acquired as training data, the training data are trained according to a BP network model and an RBF network model respectively, the optimal mixing coefficient is obtained through an adaptive genetic algorithm, the two network models are parallelly connected according to the optimal mixing coefficient, and a BP-RBF mixed neural network evaluation model is obtained; and S3, feature data of the other part of rubber material images in the sample set are acquired as evaluation data, the data are put in the BP-RBF mixed neural network evaluation model obtained through training, and the carbon black dispersity grade in the rubber material is acquired through output. The accuracy is high; parameters can be adjusted and optimized along with increasing of the sample number in use, and the robustness is strong.

A device for high accurate measurement of time intervals is based on the conversion of the time interval to a sequence of samples of the response of a surface acoustic wave filter excited at the beginning and at the end of the measured interval. In one of its configurations, the time interval measurement device includes the input of the pulse signal, the filter exciter, the surface acoustic wave filter, the amplifier, the sampler, the analog-to-digital converter, the sample registers, the register of sample numbers, the voltage comparator, the control circuits, the sample counter, the computer, the output of the reference clock signal source, and the output of the measured time intervals.

The invention provides a pipeline leakage weighted positioning method and device based on pressure waves and sound waves. The method comprises the following steps: obtaining a signal transmission time difference obtained through multiple instruments and multiple methods; and carrying out exponential weighting processing on the signal transmission signal difference by adopting an iterative method to obtain positioning results. The device comprises an obtaining module and a calculating module. According to the pipeline leakage weighted positioning method and device, comprehensive positioning precision for multiple signal sources can be remarkably improved by an exponential weighting algorithm established based on physical characteristics of a signal, and the pipeline leakage weighted positioning method is superior to a weighted positioning method based on probability statistics in requirements on sample number, positioning inaccuracy, calculated amount and the like.

The invention provides a hyper-spectral image classification method based on a Gaussian process classifier collaborative training algorithm-. The method belongs to the technical field of image processing, and mainly solves the problem of low classification precision when the number of marked samples of hyper-spectral data is small in the prior art. The method comprises the following implementation processes of: firstly, randomly dividing feature vectors of samples into two sub feature vectors which are used as two visual angles of the samples; performing Gaussian process classifier collaborative training by using the two visual angles of partial non-marked samples and existing marked samples to obtain two final Gaussian process classifiers; and marking the two visual angles of all the non-marked samples by using the two final Gaussian process classifiers respectively, wherein the class mark with higher probability is used as the classification result of the non-marked samples. The method applied to hyper-spectral image classification can be used for remarkably improving the classification accuracy under the condition of small marked sample number.

In an imaging device according to one aspect of the invention, an amplitude demand Ac is specified according to a wobbling amplitude Ai required for detection of a hill-climbing direction based on an F value and a wobbling amplitude Ag required for detection of the hill-climbing direction based on an AGC gain. An allowable amplitude Am set for preventing a potential image shake caused by a magnification change with a position shift of a compensator lens in wobbling is specified corresponding to a zoom position and an object distance. When Ac>Am, the control procedure restricts the wobbling amplitude to Am and increases a sampling number of AF evaluation values to be detected at respective positions in wobbling, in order to ensure the sufficient detection performance of the hill-climbing direction. When Ac□Am on the other hand, the control procedure sets Ac to the wobbling amplitude. This arrangement of the imaging device efficiently ensures the sufficient detection performance of the hill-climbing direction for autofocusing control, while effectively preventing deterioration of the quality of a captured image due to the magnification change in the wobbling motion.

The invention provides a method and a device for separating a single karyocyte, in particular a method and a device for separating single karyocyte of cord blood, marrow and peripheral blood. The single karyocyte can be separated without a hundred grades purification environment, and the method and the device are not limited by sample numbers and volumes, and has high efficiency, simple structure and no pollution, thereby being applied to laboratory research and human body treatment, and being easy to produce in batch.

The invention provides a CNN and transfer learning-based disease intelligent identification method and system, which can reduce interference of a picture background, can achieve high recognition accuracy under the condition of a limited sample number, and supports higher training sample multi-classification operation efficiency. The disease image identification method comprises the following steps: image preprocessing: normalizing the size of the image, quickly positioning a disease region by using Fast-RCNN, and eliminating background interference; image feature extraction: using a triplet similarity measurement model to extract image features, and then using SIFT features as compensation features to perform weighted fusion; and disease classification and recognition: learning a first image feature of a normal plant image by adopting a deep convolutional neural network, then learning a second image feature of the disease plant image by using transfer learning, and finally performing classification and recognition by combining the first image feature and the second image feature.

The invention discloses a sand-dust type aerosol inversion method based on support vector machine identification. The method comprises the steps of selecting a sand-dust aerosol layer with obvious characteristics, a thick cloud layer with obvious characteristics and a thin cloud layer with obvious characteristics as classification samples, training classifiers through various sample numbers and characteristic vectors, then confirming an optimal classifier, classifying data of a satellite borne laser radar sand-dust source region, obtaining a sand-dust aerosol identification result with high accuracy, and inverting level height and optical thickness of an aerosol. According to the sand-dust type aerosol inversion method based on the support vector machine identification, a support vector machine is used for confirming characteristics of a hyper-plane through support vectors, the requirements for the number of the samples can be reduced, interference on classification accuracy due to sample uncertainty is reduced, and 532nm polarization detection data and level height information of a satellite borne laser radar are effectively used for distinguishing non-spherical ice crystal cloud and sand-dust type aerosol particles. The sand-dust type aerosol inversion method based on the support vector machine identification is particularly suitable for processing satellite borne laser radar detection data of a sand-dust season in northwest region of our country.

The invention discloses a ground feature classifying method based on simplified polarization synthetic aperture radar data, which relates to the synthetic aperture radar remote sensing information processing technology. The ground feature classifying method includes steps: a) obtaining Stokes parameters after filtering; b) extracting characteristic parameters including polarizability m, relative phase difference delta among orthogonal channels and total polarization power span for showing scattering mechanisms; c) calculating power ingredients corresponding to the basic scattering mechanisms, and determining main scattering mechanisms of pixels and dividing data into three types of pixels with identical scattering characteristics; d) further dividing each type of pixels with identical scattering characteristics into a plurality of types of pixels with equal sample numbers by the aid of the span; and e) realizing iterative Wishart classification. The ground feature classifying method is simple in principle and convenient in calculation, and can be used for analyzing and classifying ground feature scattering characteristics.