589 results about "Sample space" patented technology

The invention discloses a method for optimizing WLAN (Wireless Local Area Network) indoor ANN (Artificial Neural Network) positioning based on FCM (fuzzy C-means) and least-squares curve surface fitting methods, relating to an indoor positioning method used for indoor positioning and aiming to solve generalization capability reduction of an ANN system caused by singular reference points existing in a training sample space. The method comprises the following steps of carrying out clustering on pre-labeled reference points based on the FCM method to confirm corresponding clustering centers and membership degree of different reference points to clustering centers thereof; obtaining the space position of the singular reference points in a target positioning area on the basis of carrying out quantitative processing and similarity calculation on the membership degree of the reference points; updating positioning fingerprint database by utilizing the least-squares curve surface fitting method to reject abrupt change points in an intensity distribution chart; estimating the cluster of a terminal on the basis of calculating the Euclidean distance between signal intensity samples collected online and different clustering centers; and finally accurately estimating the terminal by utilizing corresponding ANN subsystems.

Optical cells are non-actively compensated to ensure that a sample gap of a sample space remains nearly constant upon a change in temperature. Fluids can be flowed through the sample space of the optical cells.

The present invention discloses a three-dimensional human face identifying method based on human face full-automatic positioning and belongs to the computer vision and mode identifying field. A human face virtual image generating method comprises the steps as follows: a two-dimensional human face shape model and a partial veins model are established; a two-dimensional human face image is positioned exactly; the two-dimensional human face image is processed for three-dimensional reconstruction according to the positioning result to obtain a three-dimensional human face image; the three-dimensional human face image is processed for illumination model treatment to obtain a virtual image with changeable gestures and illumination. The method comprises the steps as follows: characteristics are picked up from the human face image to be identified and compressed; the human face is identified according to the compressed and processed characteristics. The present invention embodiment generates the virtual image by the three-dimension reconstructing of the two-dimensional human face image and by processing the illumination model, thereby increasing the sample space of the gesture and the illumination change of the image; at the same time, the three-dimension reconstructing speed is improved greatly, thereby ensuring that human face image identification has higher efficiency and recognition rate.

The embodiment of the invention discloses a method for recognizing face images, which comprises the steps of: accurately positioning two-dimensional face images in a preset database; conducting three-dimensional reconstruction to the two-dimensional face images according to a preset three-dimensional face image model and the accurate positioning results of the two-dimensional face images to obtain three-dimensional face images; conducting illumination model treatment to the three-dimensional face images to obtain virtual images with changing postures and illumination; classifying the virtual images to obtain classification results and taking the classification results as preset classification results; and recognizing the two-dimensional face images to be recognized by using the preset classification results. The method increases the sample space of the posture and illumination change of images by the three-dimensional reconstruction and illumination model treatment of the two-dimensional face images to generate virtual images and accelerates the three-dimensional reconstruction to a great extent simultaneously, thus leading the recognition of face images to have higher efficiency and recognition rate.

A 2D or 3D velocity field is reconstructed from a cross-correlation analysis of image pairs of a sample, without first reconstructing images of the sample spatial structure. The method can be implemented via computer tomographic x-ray particle image velocimetry, using multiple projection angles, with phase contrast images forming dynamic speckle patterns. Estimated cross-correlations may be generated via convolution of a measured autocorrelation function with a velocity probability density function, and the velocity coefficients iteratively optimised to minimise the error between the estimated cross-correlations and the measured cross-correlations. The method may be applied to measure blood flow, and the motion of tissue and organs such as heart and lungs.

A method of constructing an enveloping mesh geometry (EMG) for a plurality of sample mesh geometries (MG). The enveloping mesh geometry is a geometric representation of the mean of the probability distribution underlying the sample space from which the enveloping mesh geometry (MG) have been drawn. The method is based on estimating probability densities f_j using the kernel density method. The method further comprises constructing, based on the enveloping mesh geometry (EMG) so constructed, further enveloping mesh geometries (EMG-α) related to the α-Quantiles of the probability distribution. The enveloping mesh geometries (EMG, MG-α) are suitable for development and test procedures in aircraft and automotive manufacturing processes.

The invention provides a traffic sign identifying method and a traffic sign identifying system based on cascading deep learning. By introducing a cascading convolutional neural network idea, expanding target sign sample space, and adding more samples having supervision functions, identification of traffic signs is additionally provided with more apriori information, and then sample space used for training of an identification device has the higher supervision function. The traffic sign identifying method is advantageous in that by fully using the various characteristic information of the signs, the deficiency of the conventional traffic sign identification based on the neural networks is remedied, and therefore the detection rate and the identification rate of the signs are improved.

The invention discloses a real-time robust far infrared vehicle-mounted pedestrian detection method. The method comprises the steps of catching a potential pedestrian pre-selection area in an input image through a pixel gradient vertical projection, searching an interest area in the pedestrian pre-selection area through a local threshold method and morphological post-processing techniques, extracting a multi-stage entropy weighing gradient direction histogram for feature description of the interest area, inputting the histogram to a support vector machine pedestrian classifier for online judgment of the interest area, achieving pedestrian detection through multi-frame verification and screening of judgment results of the pedestrian classifier, dividing training sample space according to sample height distribution, building a classification frame of a three-branch structure, and collecting difficult samples and a training pedestrian classifier in an iteration mode with combination of a bootstrap method and an advanced termination method. According to the real-time robust far infrared vehicle-mounted pedestrian detection method, not only is accuracy of pedestrian detection improved, but also a false alarm rate is reduced, input image processing speed and generalization capacity of the classifier are improved, and provided is an effective night vehicle-mounted pedestrian-assisted early warning method.

The invention discloses an improved eliminating method of tapered beam CT reconstruction image annular false shade, comprising the steps as follows: the invention changes the annular false shade of reading CT reconstruction image into a straight line through a method of coordinate change, then takes a one-dimension filter parallel with the false shade as a mask to search a middle image regularly line by line and acquire sample space, at the same time calculates out the variance about the pixel value of each sample space, acknowledges the false shade according to whether the variance is less than the preset valve value, substitutes the pixel value of each sample space with the average pixel value between the point and two upper and lower false shades in the ordinate axis direction in order to eliminate the annular false shade, and finally changes the middle image free from the false shade to be under the original frame of axes to gain a corrected image.

The invention discloses a multilevel inverter fault diagnosis strategy based on a principal component analysis and multi-classification related vector machines (PCA-mRVM). The multilevel inverter fault diagnosis strategy includes: subjecting primary samples to dimensionality reduction through principal component analysis, and extracting multiple principal components with fault features so as to form training samples; subjecting the training samples to fault diagnosis through the multi-classification related vector machine, outputting probabilities of fault classifications, and taking the fault classifications with the maximum probabilities as diagnosis results. The multilevel inverter fault diagnosis strategy has the advantages in the fault diagnosis with larger sample space and more classifications, is high in model sparseness, low in computation complexity and the like; most importantly, the probabilities of classification members can be outputted through the mRVM, probability and statistic significance is achieved, and uncertain problems can be conveniently analyzed.

The invention relates to a high reliable inrush current brake method for a transformer, which comprises the steps of sampling, calculating the three-phase differential current fundamental wave content and the second harmonic content of a current sampling point, setting and calculating a floating threshold, and carrying out differential starting judgment, the judgment of the change characteristics of the fundamental wave content and the judgment of the change characteristics of the ratio between second harmonic content and the fundamental wave content by a discriminant so as to determine reliable locking or immediate differential opening. On the basis of a traditional second harmonic braking principle, the method sets the floating threshold, tracks and masters the change characteristics of the fundamental wave and the second harmonic in a real-time manner, analyzes and judges the operation state of the transformer in time by utilizing the change characteristics of the fundamental wave and the second harmonic, solves the contradiction of the right and fast motion between the inrush current brake and differential protection in the current method and ensures the reliable brake of the transformer under the condition of the only existence of the inrush current and the right and fast motion of the transformer under the situation of any fault in the sample space. Meanwhile, the method reduces the value scope of the brake coefficient of the second harmonic and improves the sensitivity of inrush current brake.

The invention relates to a shale reservoir compressibility evaluation method based on a support vector machine technology. The method comprises the following steps: (1) mineral composition quantitative analysis and uniaxial compressive test are firstly carried out; (2) logging characteristic parameters for recognizing a crack are extracted; (3) maximum and minimum horizontal principal stress in different layers is calculated through a combined spring model, and a stress sensitivity degree is then calculated; (4) a compressibility evaluation model is built; and (5) discrete layer stratum compressibility evaluation sample space is then built, and a relationship between a compressibility evaluation result and different response logging curves is subjected to correlation analysis. The shale reservoir compressibility evaluation method based on the support vector machine technology has the beneficial effects that compressibility influencing factors are considered perfectly, the shale gas reservoir compressibility model is built from three aspects of a brittleness index, a natural crack development degree and the stress sensitivity, insufficient consideration on the factors by the existing evaluation method is remedied, and the comprehensive ability of the shale to form a complex crack network during volume cracking can be better reflected.

The invention discloses a fast sparse Bayesian learning based direction-of-arrival estimation method and mainly solves the problems of heavy computation and large location estimation error in the prior art. The method includes the implementation steps: (1) adopting antenna receivers to form a uniform linear array; (2) sampling space signals and computing an array covariance matrix R; (3) vectorizing R to obtain a sparse model vector y; (4) dividing space domain grids, and constructing an over-complete base phi(theta) according to the structure of the sparse model vector y; (5) establishing a sparse equation according to the sparse representation relation of the sparse model vector and the over-complete base; (6) defining a hyper-parameter vector alpha, and adopting a fast sparse Bayesian learning algorithm to solve the sparse equation; (7) drawing a magnitude spectrogram according to an optimal estimation value of alpha to obtain a direction-of-arrival value. By the method, estimation accuracy of target reconnaissance and passive location under the conditions of low signal to noise ratio and small snapshot number is improved, computational complexity is lowed, and the method can be applied to target reconnaissance and passive location.

Control systems, methods and power conversion systems are presented for controlling harmonic distortion, in which multi-sampling space vector modulation (SVM) is employed for controlling power converter switching devices, with a reference vector being sampled two or more times during each SVM period to update the SVM dwell times more than once during each SVM cycle.

A method for characterising texture or a texture-like region in an image includes the steps of obtaining saliency values (104) of an image or set of images and applying a threshold to the saliency values (108), to remove the less salient features. A three dimensional shape, for example a cuboid of a predefines size, is generated (210) and saliency space sampled by moving the cuboid across spatial dimensions of the saliency space. An estimation of z probablility density function of scales within that sample space is generated and texture or a texture-like region in the saliency space is characterised using the estimation. This provides a method by which texture can be classified within an image to aid image interpretation. In particular, the texture is classified independent of scale, orientation and illumination. The method is particularly useful for texture classification problems where: the scale is unknown, the scale may vary, or a general scene description is required.

A method of learning user query concept comprising: providing a multiplicity of respective samples; soliciting user feedback as to which of the multiple presented samples are close to the user's query concept; and wherein refining a user query concept sample space.

The invention discloses a human face detection and tracking method based on an unmanned aerial vehicle mobile platform and belongs to the field of computer vision. The human face detection and tracking method is mainly characterized by comprising the following steps: an offline studying step: taking a standard human face library as a sample space to carry out offline studying to obtain a human face detector, and taking a target human face library as a sample space to carry out offline studying to obtain a target human face detector; a human face detection steps: receiving a video frame photographed by an unmanned aerial vehicle; extracting a characteristic value of the video frame and conveying the characteristic value into the human face detector; judging whether a human face is detected or not by the human face detector according to the characteristic value; if the characteristic value of the video frame is conveyed into the target human face detector, judging whether a target human face is detected or not by the target human face detector according to the characteristic value; if so, marking the characteristic value of the target human face and adding the characteristic value into a tracking list; and a human face tracking step: predicating the position of the target human face in the next video frame according to a coordinate position in the current video frame according to the target human face.

The method includes following steps: obtaining online data of power network from energy management system and power market; determining probability of equipment fault in the time period based on arrangement of equipment repair and weather condition in each area; creating sample space from Monte Carlo simulation, each sample includes two parts: X (increment direction of power), and Y (set of fault equipment); calculating steady tolerance of voltage corresponding to each sample and obtaining probability density distribution of voltage of voltage in power system from layered computation method; determining value of steady tolerance system needed to reach, limiting values of system parameters, and precaution control value; adjusting working parameters of system to ensure steady voltage in power system under specified confidence level. Advantages are: getting rid of a great lot calculation in traditional 'point-by-point method' and avoiding CPF calculation again.