121results about How to "Proven validity" patented technology

The invention discloses a mechanical fault diagnosis method based on a migration relationship network. The method comprises the following steps: constructing source domain and target domain data of the migration relationship network; constructing a training set and a test set of migration relationship network samples; constructing a migration relation network capable of detecting the mechanical fault type; and training the migration relation network to obtain a mechanical fault diagnosis model, and performing test and performance evaluation on the final model. The invention provides a migration relationship network with a Siamese structure, which combines a relationship network in meta-learning and migration learning for the first time. A double-channel relation network is constructed by utilizing a Siamese structure, all data of a source domain and label-free data of a target domain are respectively input, information of the target domain is fully considered during additional training, and the accuracy of fault diagnosis is greatly improved. MK-MMD is fused into a network, so that the probability distribution distance between two different fields is effectively reduced, and the application of laboratory data to actual mechanical fault diagnosis becomes possible.

The invention relates to a method for assessing an insulation state of oil paper based on characteristic quantity of frequency domain dielectric spectroscopy. The method comprises following steps: utilizing a linear medium response equivalent circuit to simulate polarization characteristics of the insulation state of oil paper for a transformer and setting up the relationship between linear medium response equivalent circuit parameters, frequency-domain characterization complex capacitance and dielectric loss factors; then, utilizing complex capacitances with high frequency bands and a parametric equivalent circuit model for dielectric loss characteristic quantity and finally, integrating dielectric loss curves with low frequency bands at prediction sections, wherein the range of frequency bands in a dielectric loss frequency domain is equally divided into n frequency points (n larger than or equal to 1), n frequency points correspond to n dielectric loss integral value and n frequencies and corresponding dielectric loss integral value are fitted to a function curve. The method for assessing an insulation state of oil paper based on characteristic quantity of frequency domain dielectric spectroscopy has following beneficial effects: in order to obtain much more insulation information carried by dielectric loss factors within the limited measurement frequency domain, integrals of dielectric loss factors are utilized in the predicted low-frequency section as characteristic parameters for assessment of insulation state; and dielectric loss integral value within the low-frequency stage obtained by calculations is utilized for determining the insulation aging state of oil paper.

The invention belongs to the technical field of remote sensing image processing, in particular to a hyperspectral remote sensing image mixed pixel decomposition method based on independent component analysis. According to the physical characteristics of hyperspectral images, the invention introduces abundance sum-to-one constraint and abundance nonnegative constraint into the target function of independent component analysis, and provides a self-adapting abundance modeling method to describe probability distribution of data, thus presenting favorable applicability towards different remote data. The method can effectively solve the problems of high mixing degree and remote sensing data mixed pixel decomposition under various interferences, and has significant application value in high-precision ground feature classification and detection and recognition of ground targets based on multispectral and hyperspectral remote images.

The invention provides a method and a system for re-identifying a pedestrian based on view angle self-adaptive subspace algorithm. With the method, a rectangular image only comprising a single pedestrian or a target rectangular frame cut from an original video image is used as an input image; a feature vector is extracted from the input image; a data set is divided into a training data set and a testing data set; a transformation matrix is acquired on the training data set by learning the view angle self-adaptive subspace learning algorithm; and the transformation matrix acquired on the testing data set by learning the view angle self-adaptive subspace learning algorithm is utilized to calculate the distance and re-identify the pedestrian. According to the invention, the fact that different cameras have different properties is taken into account; and different transformations are used to compensate the transformation properties of different cameras, thus the optimum transformation relationship between each pair of the cameras can be acquired more flexibly through learning; and the features under different cameras after the cameras are transformed can be closer to the ideal feature distribution.

The invention discloses a method for deeply learning and predicting medical track based on medical records. The method comprises the following steps: S1, encoding diagnostic information and intervention information on admission through an encoding scheme and converting code into vector to acquire diagnostic information conversion vector (the formula is shown in the description) and intervention information conversion vector (the formula is shown in the description) separately, and converting the diagnostic information and intervention information on admission for one time into one 2M-dimensional vector [xt, pt]; S2, input the vector [xt, pt] into an LSTM model, and evaluating the current output value ht to obtain the current disease state; S3, predicting a diagnostic code dt+1 according tothe disease state ht and predicting the disease progression through the diagnostic code dt+1; S4, calculating an intervention code st of the time t, increasing a time structure in the LSTM model, collecting the historical disease states in multiple time ranges, collecting the state of each section of a horizontal time shaft, collecting all the diseases states, stacking into a vector (the formulais shown in the description), and feeding back the vector (the formula is shown in the description) into a nerve network to predict the future risk result Y.

The invention discloses a decision-making method for an intelligent type dynamic route guidance system and a guidance unit. The method comprises the following steps: firstly, the turning rate of the road section to the downstream road section during the period p is obtained; secondly, the predicted guidance turning rate corresponding to all guidance information in a knowledge base is updated; thirdly, the traffic flow in different directions during the period of p plus1 is predicted to form proportion; fourthly, the optimum guidance information of the traffic flow in the different directions is withdrawn from the knowledge base; fifthly, according to the state of the road net of the guidance unit, the composition of the traffic flow and the corresponding optimum guidance information thereof are comprehensively considered, and the comprehensive guidance information for the guidance unit during the period of p plus 1 is obtained; and sixthly, the guidance period is updated, i.e. the P is equal to p plus 1, and the guidance information acts on the traffic flow through a VMS. The invention adopts the intelligent method to handle the guidance compliance rate, solves the problem that the influence range of the data (such as guidance compliance rate, guidance information, and the like) to the traffic flow is not easy to be obtained, and obtains better effect. The implementation steps are simpler, and a great amount of matrix operation in the transitional method is not cited. Therefore, the implementation is easy.

The invention discloses a spectrum prediction method in a cognitive radio system based on a convolutional neural network. The method can obtain the higher prediction accuracy through the shorter-timetraining, and belongs to the field of information and communication. The method comprises the steps: 1, performing the uninterrupted spectrum sensing of a frequency band F through a period T, recording S<t0>, S<t1>, S<t2>,... of the spectrum sensing from the moment t0 and I<t0>, I<t1>, I<t2>,... corresponding to S<t0>, S<t1>, S<t2>,..., marking t0+kT as tk, wherein k=0, 1, ..., wherein S<tk> represents the occupation conditions of all channels of the frequency band F at the moment tk, I<tk> represents the number of idle sensing periods of each channel of S<tk> from the moment tk+T; 2, inputting S<t0>, S<t1>, S<t2>,... and I<t0>, I<t1>, I<t2>,... into the convolutional neural network, and performing the training of the convolutional neural network, wherein I<t0>, I<t1>, I<t2>,... are used for making labels in the training; 3, continuously predicting the channel C with the maximum future idle probability according to S<t0>, S<t1>, S<t2>,... through the trained convolutional neural network.

The invention belongs to the technical field of hyperspectral remote sensing image processing and in particular provides a method for selecting hyperspectral remote sensing image bands based on partial least squares. The method has the following beneficial effects: based on the characteristic that components extracted by partial least squares maintain hyperspectral image variation information andhas high degree of correlation with classified information, the energy of the product of the spectrum matrix and the membership matrix is regarded as the standard of band selection and the recursive residual of the selected band is obtained through iteration to select the next group of bands to realize the process of band selection; the following defects of the traditional methods for selecting multispectral image bands can be effectively overcome: the computation complexity is high and relevant bands need to be removed; the hyperspectral remote sensing image classification experiment result shows that the hyperspectral remote sensing image has good classification effect after the method is used for selecting the bands; and the method has important application value for efficiently utilizing the information resources of the hyperspectral images.

The present invention relates to a method of making video search of scene based on contents. It uses the fuzzy aggregation analysis method in the scene search. As compared with existent method it canobtain higher accuracy and quick searching speed.

The invention discloses a fine-grained action detection method of a convolutional neural network based on multistage condition influence. The method comprises the steps of: establishing a convolutional neural network influenced by multistage conditions; fusing the explicit knowledge added in the visual scene with the multi-level visual features; enabling the multi-level conditional influence convolutional neural network MLCNet to take a conditional influence multi-branch convolutional neural network structure as a main trunk, generating multi-level visual features, encoding additional spatialsemantic information of human body structure and object context information as a condition, dynamically influencing feature extraction of a CNN through affine transformation and an attention mechanism, and finally fusing and modulating multi-mode features to distinguish various interactive actions; and carrying out model training on the convolutional neural network influenced by the multi-level condition, and outputting a fine-grained action detection result by the obtained model. According to the method, the proposed method is evaluated on the basis of two most common references, namely HICO-DET and V-COCO, and experimental results show that the method is superior to the existing method.

The invention discloses a coordination method for an intelligent type dynamic route guidance system and a guidance subarea. The method comprises the following steps: firstly, the same road junction guidance units in the guidance subarea are coordinated, the optimum guidance information is withdrawn from a knowledge base, and the saturatation degree of the road section of a connecting road junction is re-predicted; secondly, after the guidance information is added in, the states of the guidance units of the downstream relevant road junctions are readjusted from the most up-stream road junction in the guidance subarea according to the obtained saturatation degree (the step 1) of the road section of the connecting road junction by prediction; and thirdly, if the states of the guidance units for the downstream road junction are not changed, the coordination between the two road junctions is ended; otherwise, the states of the relevant guidance units for the downstream road junction are adjusted. The mutually independent guidance units in the guidance subarea are connected through the states of the guidance units under the intelligent type DRGS frame; and meanwhile, the influence of the traffic flow of the upstream guidance units as well as the traffic flow of other guidance units for the same road junction on the decision-making guidance units are reflected.

The invention discloses a maneuvering target parameter estimation method by combining correction RFT (Radon-Fourier Transform) and MDCFT (Modified Discrete Chirp-Fourier Transform). According to the method, a method for searching target echo data through the RFT is improved by using a previously known parameter range for movement of a maneuvering target, so the target echo data with a range curve can be taken out, corresponding matching processing is performed on the basis of the taken echo data, and further estimated values of initial speed and acceleration of the target are obtained. According to the method, a better maneuvering target parameter estimation result can be obtained under the conditions of limited radar pulse accumulation pulse number and low signal-noise ratio. According to the method, the target speed estimation result is compared with estimation results of MTD (Moving Target Detection), RFT, fractional fourier transform and Radon-fractional fourier transform methods; besides, the estimation result of the target acceleration by the method is compared with the estimation results of the fractional fourier transform and the Radon-fractional fourier transform methods, and the experimental result proves the effectiveness of the method.

The invention belongs to the technical field of remote sensing image processing and particularly relates to a hyper-spectral remote sensing image mixed pixel decomposition method based on a distance geometry theory. The invention provides an operational formula for calculating an areal coordinate of a high-dimensional data space by introducing the distance geometry theory into a hyper-spectral remote sensing image mixed pixel according to physical characteristics of a hyper-spectral image and geometric characteristics of a data set, and obtains a position estimation algorithm which can well keep a geometric structure of the data set according to a distance geometric constraint; and finally, a novel high-precision and low-complexity abundance estimation algorithm, namely the abundance estimation algorithm based on a distance geometry, is obtained. The algorithm has good applicability to various different hyper-spectral data (including emulated data and actual data sets). The hyper-spectral remote sensing image mixed pixel decomposition method based on the distance geometry theory, disclosed by the invention, has very important application value on aspects of high-precision ground feature classification of multispectral and hyper-spectral remote sensing images, and detection and identification of a ground target.

The invention relates to an intermediate infrared femtosecond mode-locked laser which comprises a collimating mirror, a focusing mirror, an input spherical surface mirror, a laser medium and a spherical surface high-reflection mirror which are sequentially arranged along a direction of a pumping light beam outputted by a laser diode, wherein lasers in a five-mirror laser resonance cavity formed by the input spherical surface mirror, the spherical surface high-reflection mirror, the spherical surface high-reflection focusing mirror, an output coupling mirror and a graphene mode-locking element is reflected onto the high-reflection focusing mirror through the input spherical surface mirror, is focused on the graphene mode-locking element, returns back along the original path by sequentially passing through the spherical surface high-reflection focusing mirror, the input spherical surface mirror, a laser crystal and the spherical surface high-reflection mirror, is deflected and reflected to a dispersion compensation prism pair by the spherical surface high-reflection mirror, and is output from the output coupling mirror through a slit. According to the intermediate infrared femtosecond mode-locked laser, graphene growing by adopting a CVD (Chemical Vapor Deposition) method is transferred to a laser wavelength high-reflection mirror, and is protected by using inert gas, and thus stable mode-locked laser pulse output is realized in an intermediate infrared band. The intermediate infrared femtosecond mode-locked laser has the advantages of being simple in regulation, low in manufacture cost, and easy to realize single layer (little non-saturated loss).

The invention provides an energy distribution method in a two-hop energy collection relay network. In a two-hop half-duplex relay network, in an energy collection process, a source node S and a destination node D transfer a part of energy to an unreliable relay R, in a first time slot of an information transmission process, the S sends useful information to the R, meanwhile the FD sends an interference signal to the R, and in a second time slot of the information transmission process, the R forwards the received signal to the D. The total secure rate of the system is defined, and an optimal energy distribution factor is solved. By adoption of the energy distribution method provided by the invention, the situation of insufficient energy of the relay node can be effectively relieved, the energy distribution factor of the node is iteratively optimized on the premise of ensuring the own energy of the node, and the maximization of the secure rate of the system is achieved.

The invention provides a monitoring method for corrosive damage of the hole edge of an aluminum alloy porous structure based on an iterative algebraic reconstruction algorithm. The monitoring method comprises the following steps: firstly, selecting a porous aluminum alloy, and arranging piezoelectric sensors by a square array layout; secondly, adopting a sine wave exciting signal; thirdly, imaging the damage by applying a tomography scanning algorithm of ART according to the network layout of the piezoelectric sensors and selected signal characteristic quantity; fourthly, carrying out location imaging on the damage by analyzing a correlation coefficient of Lamb waves; fifthly, adopting mean filtering processing, and solving a value of a final pixel by using a mean of a template to replace a value of an original pixel. According to the monitoring method disclosed by the invention, by performing the steps, different piezoelectric sensors are acquired by layout optimization of the piezoelectric sensors and signal monitoring; characteristic parameters are extracted by signal processing, the corrosive damage degree of a porous aluminum alloy member can be quantitatively characterized, and the function of monitoring the corrosive damage of the aluminum alloy structure in real time can be realized.

The invention provides a cascade rolling-ring amplification strategy mediated with a dumbbell molecular probe under protection of a target object and applied to sensitiveness detection of DNA transmethylase activity. Based on the cascade rolling-ring amplification strategy mediated with the dumbbell molecular probe under protection of the target object, a novel fluorescence detection system for DNA transmethylase activity analysis is developed. The detection system has excellent specificity and sensitivity, the detection limit is as low as 0.0024 U/mL, and potential application value is exhibited in quantitative monitoring of transmethylase activity and screening of anticarcinogen.

The invention provides a method for identifying a lysine single methylated modification substrate. The method comprises the following steps: 1) an external acylated derivative reaction is used for deriving a single methylated epsilon-amino group of a substrate protein lysine residue to a propionyl methylated epsilon-amino group; 2) a affinitive enrichment is carried out for peptide fragments with propionyl methylated modification with a pan anti-propionyl methylated lysine antibody prepared by specificity; and 3) a mass spectrum identification is carried out for sites, polypeptide sequences and substrate proteins with propionyl methylated modification after affinitive enrichment by the antibody.

The invention relates to an RST attack resistance stereo image zero watermark method based on FFST and Hessenberg decomposition. The method comprises a zero watermark generation part and a zero watermark detection part. An authentication zero watermark of a three-dimensional image is constructed by utilizing the double-viewpoint characteristic of the three-dimensional image and utilizing the sizerelation between absolute values of elements at the upper left corners in respective coefficient sub-blocks after paired random image blocks are decomposed in low-frequency sub-bands after fast finiteshear wave transformation (FFST) of the left viewpoint image and the right viewpoint image, and the method is simple, novel and high in robustness. Meanwhile, a fractional order Arnodo chaotic systemis adopted to generate a random sequence, and encryption of an original binary watermark image is achieved. Before watermark extraction, geometric attack correction is carried out on a to-be-authenticated three-dimensional image by using an image matching method based on Fourier-Merlin transform. Experimental results show that the method provided by the invention has good robustness for resistingvarious common image processing attacks such as noise addition, filtering, JPEG compression, shearing, rotation, zooming, translation (RST) attacks and the like.

The invention discloses an inverter output voltage state stabilizing method based on sliding mode control. The method comprises the following steps: step 1, a reference voltage vref is set, an operation model is established according to an inverter circuit, and the difference between an output voltage V and the reference voltage vref is defined as the error signal value s of a sliding mode variable; step 2, a dynamic parameter value Beta is set, wherein s<=Beta sM, s>=Beta sm, and a controller structure is composed of four states S<M><+>, S<m><+>, S<M><-> and S<m><-> and an initial state; step 3, the error signal value s and the derivative thereof S(.) are converged to a sliding mode surface s=0 and S(.)=0 through a control algorithm according to real-time update of the dynamic parameter value; and step 4, the output voltage V=vref+s is output.

The invention discloses a hub site selection and distribution optimization method for a multi-distribution express shaft spoke network. The node set is shown in the description, at least one pivot butno more than two pivots is arranged on the express delivery path between any two points; only 2, 3 and 4 nodes are possibly arranged on the path, express services between any two nodes shown in the description are transmitted along the pth in the network G, the whole express delivery path can be represented by a matrix (please see the formula in the specification), and pij represents a second node on the path from the node i to the node j. According to the invention, comprehensive optimization is carried out as a unified whole; Under the constraint of transportation operation time budget, a combined simulated annealing algorithm based on conditional shortest circuit is designed with the goal of minimizing express network transportation cost, transit cost and sorting cost, calculation andanalysis are carried out on examples, and the effectiveness of the model and the algorithm is verified, the enterprise operation efficiency is improved, and the express enterprise operation cost is reduced.

The invention belongs to the field of remote sensing image processing technology, and particularly relates to a simplex triangular decomposition-based method for decomposing mixed pixels of hyperspectral remote sensing images. The method adopts a linear mixture model, and comprises the two steps: end member extraction and abundance estimation. The method not only is a geometric method based on simplex, and meanwhile is based on the algebraic principle of triangular decomposition. The triangular decomposition adopts either Cholesky decomposition or QR decomposition, and can improve the search efficiency of an end member during the process of end member extraction through recursive operation. The method can effectively extract the end member from hyperspectral remote sensing data, thereby solving the problem in decomposing mixed pixels. The method has a specifically important application value in the hyperspectral remote sensing image-based high-precision surface feature classification as well as the inspection and identification of ground targets.

The invention provides an application of a virtual construction equipment technology based on BIM in steel bridge engineering. Research is conducted from aspects such as a fundamental theory, a BIM technology, a virtual technology and technology applications, a theoretical framework of a virtual construction technology is provided, a simulation application model of the virtual construction technology is built, and based on an AECOsim-Building-Designer platform in Bentley series software, simulation function application is conducted. According to the application, through test and application offunction models built in the application in a practical engineering project, the effectiveness and the feasibility of those simulation models are verified, so that the construction process is effectively simulated, construction equipment resources are effectively simulated, utilization of the resources are effectively analyzed, optimization of construction duration-cost is effectively conducted,and an expected goal of the application is achieved.

The invention discloses a posture-spanning colored image facial expression recognition method of a direct push type migration group sparse discriminant analysis (TTGSLDA). Training and testing human face images are photographed under two different face visual angles; one group of auxiliary unmarked human face images are selected from target human face postures and are integrated into a marked training image set in source human face postures; and a label of an auxiliary image is a parameter to be optimized by a direct push type linear discriminant analysis (TTLDA) method. After an obtained class label of an auxiliary image set is learnt, a support vector machine (SVM) is trained based on the parameter so that the classification of the testing human face images is finished. In order to sufficiently utilize face information of a colored image to improve the accuracy of the facial expression recognition, characteristics of the human face images are represented by colored scale invariant feature transform (SIFT). The posture-spanning colored image facial expression recognition method has a facial expression recognition effect with relatively high recognition rate and relatively good robustness.