47 results about "Gabor transform" patented technology

A signal analysis system and method for analyzing an input signal acquired from a mechanical system. The mechanical system may include at least one rotating apparatus. The signal analysis system may be configured to: (a) receive samples of the input signal, (b) perform an invertible joint time-frequency transform (e.g. a Gabor transform) on the samples of the input signal to produce a first array of coefficients which depend on time and frequency, (c) select first coefficients from the first array which correspond to a first subset of one or more order components in the input signal, (d) generate a time domain signal from the first coefficients, and (e) present the time domain signal to a user on a presentation device. The signal analysis system generate the time domain signal from the first coefficients by performing an inverse joint time-frequency transform on the first coefficients. The signal analysis system extract the one or more order components for presentation to the user by masking out all coefficients except those corresponding to the one or more components. Conversely, the signal analysis system may suppress the one or more order components, i.e. present to the user the input signal minus the one or more order components by masking out coefficients corresponding to the one or more components and keeping the remaining coefficients.

The invention discloses a deep learning face identification method based on multiple-characteristic fusion. The method comprises the steps of firstly performing 2D gabor transforming on a to-be-tested orl face database and extracting the face database with a gabor characteristic; because of overlarge dimension of the picture being 92*112, reducing the dimension of the image to 32*32 according to a bilinear interpolation method; then fusing the original orl face database with the face database with the gabor characteristic; and finally coding in a stack type self-coding manner in deep learning, and calculating a weight parameter in a sofmax regression manner, and predicting an identification accuracy. According to the deep learning face identification method, after multiple characteristics are fused under a precondition that the testing accuracy of the characteristic through singly utilizing the algorithm is not lower than 80%, higher accuracy and high algorithm stability are realized; and even after random initialization, identification accuracy is basically unchanged.

A signal analysis system and method for analyzing an input signal acquired from a mechanical system. The mechanical system may include at least one rotating apparatus. The signal analysis system may be configured to: (a) receive samples of the input signal, (b) perform an invertible joint time-frequency transform (e.g. a Gabor transform) on the samples of the input signal to produce a first array of coefficients which depend on time and frequency, (c) compute an instantaneous rotation frequency signal, (d) select first coefficients from the first array which correspond to one or more order components in the input signal, i.e. one or more multiples of the instantaneous rotation frequency signal, (e) generate a time domain signal from the first coefficients, and (f) present the time domain signal to a user on a presentation device. The signal analysis system may generate the time domain signal from the first coefficients by performing an inverse joint time-frequency transform on the first coefficients.

The invention provides a micro-expression recognition method based on a difference slice energy diagram and Gabor transformation. The micro-expression recognition method comprises the following steps: constructing a micro-expression sequence difference slice energy diagram, calculating micro-expression frames containing change regions in a micro-expression periodic sequence, graying the micro-expression sequence to extract micro-expression difference slices, and overlapping micro-expression difference slice frame sequences to obtain a micro-expression difference energy diagram; extracting the characteristics of the difference slice energy diagram, constructing a Gabor filter kernel function, performing Gabor characteristic extraction on the difference slice energy diagram, sampling the extracted characteristics, writing the characteristics of each sample in a column vector form, maximizing the quotient of inter-class divergence and intra-class divergence through linear discriminant analysis with supervisory information, and further extracting the characteristics of each sample; performing classification and identification, training a model according to training data, and predicting and classifying test samples according to the model. The method provided by the invention has general applicability. Compared with an existing method, the method provided by the invention has a higher identification rate.

A system for processing an image data volume acquired with a medical imaging acquisition device 100 from a body comprising a needle is provided. It has been realized it is advantageous to display 150 a plane intersecting the image data volume showing the needle to a user of the system. This allows better positioning of the needle. The system comprises a needle-plane determination 110 module for determining a plane being parallel to and intersecting a representation in the image data volume of the needle and being parallel to a viewing direction. The needle plane determination module may make use of pixel processing and/or spectral transformation, in particular the Gabor transform.

The invention discloses an SAR variant target identification method based on local textural features, mainly solving the problem of low SAR variant target identification rate with the existing identification method. The realization process is as follows: 1. improving the statistical distribution of each part of the SAR target by partial differential; 2. cutting a target part for the SAR target which is performed with partial differential transform by Otsu; 3. rotating the target to 90 degrees, selecting a sliding window with fixed size, and cutting in different directions according to different azimuth angles; 4. carrying out Gabor transform to the cut SAR target; 5. coding each image performed with Gabor transform by an LBP operator, and building a histogram; 6. matching a test sample with each SAR image of the training sample by the histogram, abandoning parts with small matching result, and only reserving parts with good matching result; and 7. judging an identification result with a nearest neighbour method. The invention can utilize the local textural features to improve the identification rate of the SAR target variant and is used for identifying a terrain object.

An audio system is provided that employs time-frequency analysis and/or synthesis techniques for processing audio obtained from a microphone array. These time-frequency analysis/synthesis techniques can be more robust, provide better spatial resolution, and have less computational complexity than existing adaptive filter implementations. The time-frequency techniques can be implemented for dual microphone arrays or for microphone arrays having more than two microphones. Many different time-frequency techniques may be used in the audio system. As one example, the Gabor transform may be used to analyze time and frequency components of audio signals obtained from the microphone array.

The invention relates to a contour detection method on the basis of a primary visual pathway calculation model. A primary visual pathway computation model is constructed, and contour detection is realized by simulating the transmission and processing process of visual information flows. A classical receptive field direction selection model combining multi-scale features is provided, and a multi-scale feature fusion strategy is utilized to simulate the primary contour response of the retinal ganglion. In a visual pathway from the retinal ganglion to the lateral geniculate nucleus, a space-timecoding mechanism is adopted for reducing redundant features in the primary contour response. A computation model with non-classical receptive field isotropic suppression properties is provided, and bymeans of a synergistic effect of non-subsampled contourlet transform and Gabor transform, a processing effect of lateral suppression characteristics of a non-classical receptive field on texture background information is simulated; a feedforward mechanism of a primary visual pathway to the primary visual cortex is simulated, visual characteristics of multiple visual pathways are fused, and the contour response is finally obtained. According to the contour detection method, a main body contour can be effectively highlighted, and a texture background can be suppressed.

The invention discloses a method for realizing traveling wave protection of an ultra-high-voltage direct-current power transmission line. The method comprises the following steps: (1) the instantaneous voltage u(t) and instantaneous current i(t) of an ultra-high-voltage direct-current power transmission line are acquired in real time to obtain the polar wave PI(t) of the ultra-high-voltage direct-current power transmission line, wherein PI(t)=Zp*i(t)-u(t), and Zp refers to polar wave impedance, time-frequency analysis is made on the polar wave PI(t) through Gabor transform to obtain the modulus value Aset of an S transform matrix, and the three-dimensional time-frequency distribution map of the S transform matrix is drawn; and (2) when Aset<Krel*Amin, a protection outlet acts, wherein Krel is reliability coefficient, Amin=Min(A<Sigma>(t)), A<Sigma>(t)=(Sigma<f=0><fs12>A(t,f)), A(t,f) is the modulus value in a Gabor transform matrix, f is frequency, t is time, fs is sampling frequency, and Amin is the minimum value of external fault and disturbance noise. By adopting the method, transient information of a line fault can be accurately grasped so as to realize traveling wave protection of an ultra-high-voltage direct-current power transmission line.

The present invention discloses a human face identifying method and apparatus for solving the problems in the prior art that redundant information can be easily brought about in the characteristic extraction of a plurality of different characteristics through a parallel manner. The human face identifying method comprises: obtaining a human face image; conducting Gabor transform to the human face image to obtain a Gabor characteristic image; using the LBP algorithm to process the obtained Gabor characteristic image to obtain the first LBP histogram characteristics; using the LPQ algorithm to process the human face image to obtain the first LPQ histogram characteristics; and fusing the first LBP histogram characteristics and the first LPQ histogram characteristics according to a preset order as the obtained human face characteristics through identification. Compared with the prior art, in the human face identifying method featuring Gabor-LBP and LPQ serial-parallel fusion, complementary information is fully utilized and redundant information is reduced.

The invention discloses a face recognition method, comprising following steps: sample face images and face images to be recognized are obtained, wherein the sample face images contains at least two face images; circularly symmetrical Gabor transform, superposition, local energy extraction, and local binary pattern conversion are performed to the sample face images and the face images to be recognized to obtain sample histograms containing texture information of the sample face images and histograms to be recognized containing texture information of the face images to be recognized; the sample histograms are compared with the histograms to be recognized to obtain face images in the sample face images which are the same with the face images to be recognized. The invention also discloses a face recognition device. By use of the method and device, the calculation quantity in a face recognition process is reduced, the calculation time is shortened, and face recognition efficiency is increased.

The invention relates to a SAR image classification method based on a multi-feature and a non-negative automatic encoder, belonging to the technical field of image processing. The method comprises steps: extracting Spatial Features of SAR Image Blocks Based on Gray Level Gradient Co-occurrence Matrix; feature Extraction of SAR Image Block Transform Domain Based on Two-Dimensional Gabor Transform;combining the spatial domain features and the transform domain features of an image block; the training sample set and test sample set of SAR image block being selected. training multi-layer non-negative automatic encoder and softmax classifier with training sample set; classification being based on the trained non-negative automatic encoder network, so that a classification result diagram is obtain. The invention combines the spatial information of the SAR image and the information of the transform domain, obtains the multi-dimensional features of the SAR image, optimizes the features by using a non-negative automatic encoder, improves the distinctiveness of the features, further effectively improves the accuracy of classification, and can be used for the classification of the ground objects of the high-resolution SAR image and the target recognition, and the like.

The invention provides a Gabor transformation based method and device for extracting and analyzing breathing and heartbeat signals. The method comprises following steps: S1, radar echo signals are subjected to Gabor transformation, and time-frequency power distribution signals of the transformed signals are calculated; S2, the time-frequency power distribution signals are subjected to frequency spectrum weighted average operation, and a frequency spectrum centroid changing curve is obtained; S3, the frequency spectrum centroid changing curve is subjected to filtering operation, and breathing signals and heartbeat signals are obtained; S4, the breathing signals and the heartbeat signals are subjected to the Gabor transformation, and breathing frequency changing with time and heartbeat frequency changing with the time are obtained. With the adoption of the method and the device, the breathing frequency and heartbeat frequency changing with time, the signals and the frequency change of the signals can be effectively extracted.

The invention discloses a semi-reference image quality evaluation method based on weighted dimension of Gabor difference box. Firstly, the input distorted image and the reference image are grayed andthen two-dimensional Gabor transform is performed, Gabor sinusoidal coefficient and cosine coefficient of that distorted gray image and the reference gray image are obtained, then the fractal dimension of Gabor sine coefficient and cosine coefficient of distorted gray image and reference gray image is calculated, difference box method is used to obtain the weighted dimension of difference box, Gabor entropy, sinusoidal weighted coefficient and cosine weighted coefficient of Gabor coefficients are calculated, and finally the semi-reference image quality evaluation score is calculated accordingto the weighted dimension and weighted coefficient of difference box. The image quality evaluation method of the invention overcomes the shortcomings that the dimension of the extracted image featureis too large to be transmitted conveniently, and the absolute difference value of the weighted dimension feature of the difference box is weighted by Gabor entropy to obtain the final objective imagequality evaluation score value, and the semi-reference image quality evaluation precision is improved.

The invention relates to the technical field of fingerprint identification, in particular to a Gabor transformation-based fingerprint feature extraction and compression coding method, which comprisesthe following steps of: (1) reading an original fingerprint image; (2) extracting fingerprint features based on Gabor transformation to obtain a Gabor amplitude response feature map of the fingerprintimage; (3) dividing the Gabor amplitude response feature map of each fingerprint image into Wx W small blocks, and carrying out block processing on the Gabor amplitude response feature maps; (4) in each small block, performing coding operation on the Gabor amplitude response value of each pixel point to obtain a total eigenvector F = [F1F2... FW x W] of the whole image; (5) taking F total = [F1 F2... FW * W] as a final feature of one fingerprint image, storing the final feature into a database, and using the final feature for data matching in fingerprint identification; According to the method, a Gabor amplitude response coding method is adopted, effective features are reserved, meanwhile, the data scale of fingerprint features can be reduced to a large extent, and data storage and matching operation are facilitated.

The invention relates to a lane detection method for improving Gabor transform and updating vanishing points. The lane detection method for improving the Gabor transform and updating the vanishing points includes the steps that 1, an original lane image is input; 2, graying pretreatment is conducted on the original image; 3, an improved two-dimensional Gabor filter is adopted to extract gray levelimage texture features in multiple scales and multiple directions; 4, principal component analysis is conducted on the image texture features to obtain texture features which have large influences onthe vanishing points; 5, according to the texture features determined in the step 4, a soft voting method is adopted to detect an initial vanishing point; 6, according to the initial vanishing pointdetected in the step 5, a histogram based on angle deviation and chromatic aberration is adopted to update the initial vanishing point and determine a lane border line; 7, according to a new vanishingpoint and the lane border line in the step 6, a lane range is determined. According to the lane detection method for improving the Gabor transform and updating the vanishing points, by adopting the improved Gabor filter, the accuracy of texture feature extraction is improved, and the problem is solved that traditional Gabor filters are low in real-time performance; the mode of adopting the histogram to update the vanishing points and detect the lane border line is convenient and high in accuracy.