33 results about "Contourlet" patented technology

The invention relates to the technical field of geological disaster detection, in particular to a geological disaster hidden danger point detection method and system. Comprising the following steps: S1, synchronously acquiring vegetation canopy penetration data, multispectral soil humidity spatial distribution data and lithology distribution data by adopting an unmanned aerial vehicle carried image data acquisition device and a multi-band laser radar, and generating an orthoimage and a three-dimensional point cloud data set; s2, fusing the orthoimage and the three-dimensional point cloud data through an aerial triangulation encryption algorithm to establish a three-dimensional model, loading the multispectral soil humidity spatial distribution data into the three-dimensional model to generate a dynamic digital elevation model, and generating a contour line topological structure through a TIN triangulation network algorithm; s3, extracting a slope factor, topographic relief, vegetation coverage, fracture distance and a lithologic factor based on the three-dimensional model to construct a geological disaster hidden danger system; according to the system, the geological disaster occurrence time can be estimated, and the rescue and protection treatment period can be effectively planned based on the geological disaster occurrence time.

The invention discloses a device and a method for accurately acquiring working face abnormal body information by a multi-frequency ground penetrating radar. The device comprises a multi-frequency data acquisition module, a data alignment module, a data fusion module, a data management module, a model training optimization module and an abnormal body information identification module. The method comprises the following steps: acquiring ground penetrating radar data of different center frequencies of a target by using a multi-frequency ground penetrating radar; preprocessing the original multi-frequency ground penetrating radar data; performing spatial alignment on the preprocessed ground penetrating radar data; fusing the ground penetrating radar data through improved non-subsampled contourlet transform (NSCT) to obtain a multi-frequency ground penetrating radar fusion profile; fused images are collected to make a ground penetrating radar image data set; and importing an improved YOLOv11 model to perform underground anomalous body identification and marking on the ground penetrating radar image data set. Through multi-frequency data complementation and deep learning optimization, anomalous body detection precision and efficiency are significantly improved, and the method is especially suitable for geological exploration of complex underground environments such as a coal mine.

The invention discloses an underwater foresight sonar image segmentation method, system and device based on CNN and Transform, and relates to the technical field of digital image processing and mode recognition. The method comprises the following steps: S1, collecting image data collected by a foresight sonar, carrying out corresponding labeling, and obtaining a training set and a verification set after data enhancement through data processing; s2, performing feature extraction on the image through a cascade convolution module and a hybrid Transform model, wherein the convolution module pays attention to local semantic information of the image; in a data processing stage, data enhancement is realized by fusing technologies such as Contourlet transformation, in a network architecture, an encoder combines convolution and hybrid Transformer, learning of local and spatial information of an image is considered, deep feature fusion and data value mining are carried out by utilizing Swin Transformer and the like, and particularly, it is worth mentioning that an edge feature module can learn edge information of a target area, and the edge feature module can learn the edge information of the target area; the segmentation result is explicitly corrected, the network segmentation performance is comprehensively improved, and the accuracy of sonar image analysis is powerfully guaranteed.

The invention discloses a distribution cable multi-source data fusion method based on spectrogram reconstruction. The distribution cable multi-source data fusion method comprises the following steps: S1, constructing a distribution cable operation state multi-source database; s2, using contourlet transform to extract features from the cable partial discharge signal diagrams in the database, and performing hierarchical decomposition of multi-source data spectrograms on the extracted features; s3, graying processing is carried out on the multi-source data spectrogram obtained after grading decomposition, grayed images are decomposed through NSCT, and multi-layer high-frequency sub-band coefficients of all the images and discharge spectrogram contour feature information are obtained; and S4, adopting a low-frequency sub-image fusion rule of local entropy, selecting a fusion coefficient of a high-frequency sub-image based on an NSCT decomposition structure, and reconstructing an optimal fusion image through NSCT inverse transformation. According to the method, multi-scale decomposition and multi-directional analysis can be realized, the feature extraction capability of cable multi-source data is enhanced, the comprehensiveness and accuracy of data fusion are improved, and a distribution cable state spectrogram with heat, electricity and vibration fused is realized.

The invention relates to the technical field of power cable intelligent monitoring, and aims to solve the technical problems that in an existing cable state monitoring system, multi-source heterogeneous data fusion is insufficient, the coupling relation between physical quantities is difficult to reveal, and insulation aging, current-carrying capacity and fault probability cannot be quantitatively predicted according to evaluation results. The invention discloses a cable state multi-source data fusion and prediction system. The system comprises a data acquisition module, a feature processing module, a prediction analysis module and a cloud management platform. The system collects partial discharge, surface temperature and vibration data of the cable through the distributed terminal; performing spectrogram decomposition and fusion of the multi-source data by adopting a non-subsampled contourlet transform algorithm to generate multi-scale fusion features; building a prediction model based on a particle swarm optimization least square support vector machine, and outputting an insulation aging state prediction value, a current-carrying capability evaluation value and a fault probability prediction value of the cable; and data management and full-life-cycle service are realized through the cloud platform.

The invention relates to the technical field of 3D printing, in particular to a three-dimensional model contour line drawing method and device, equipment and a medium, and the method comprises the following steps: obtaining a screen interaction position corresponding to a display screen for displaying a target model picture, and mapping the screen interaction position to a target model in a three-dimensional model space to obtain a target model position; determining contour line position information of a model contour line according to the target model position; according to the model color information corresponding to the target model position, determining contour line color information of a model contour line; according to the contour line position information and the contour line color information, drawing and displaying a target model picture comprising a model contour line; according to the method, by means of the height of the intersection point of the model contour line visualization mouse and the target model, transition processing is conducted on the color value of the contour line, the color of the contour line and color information corresponding to the target model are mixed, the edge is smoothed, the contour line anti-aliasing effect is achieved, and the display effect is improved.

The invention discloses an image processing method and device, a storage medium, electronic equipment and a program product. Relates to the financial science and technology field or other fields. The method comprises the following steps: acquiring a seal image of which the authenticity is to be verified; performing image decomposition processing on the seal image by adopting a non-subsampled contourlet transform method to obtain a plurality of sub-band images; performing feature extraction processing on the plurality of sub-band images by adopting a gray-level co-occurrence matrix method to obtain image features of the seal image; the absolute correlation degree between the image features of the seal image and the image features of the reference seal image is calculated through a grey correlation clustering analysis method, a verification result is determined based on the absolute correlation degree, and the verification result is used for representing whether the seal image is real or not. According to the invention, the method solves a problem that the recognition accuracy is low when the authenticity of the seal image is recognized in the prior art.

The invention relates to the technical field of optical interference imaging, in particular to a high-resolution image fusion method based on a one-dimensional sparse aperture rotating pupil. Comprising the following steps: inputting a target image, imaging the target image by adopting a one-dimensional sparse aperture rotation pupil, and obtaining one image when the target image is rotated to an angle; performing non-subsampled contourlet transform to obtain a low-frequency image and a high-frequency image corresponding to each image; performing equal-weight superposition and normalization on the plurality of low-frequency images and the plurality of high-frequency images; repeating the steps to construct a data set; establishing a double-branch residual fusion network; through a residual learning strategy, high-frequency branch detail information is extracted, and global linear transformation is carried out on the low-frequency image; and fusing the residual image of the high-frequency branch and the adjusted low-frequency image, and outputting a final high-resolution image. The method has the advantages that effective information is accurately extracted, a data set required by training is simplified, full-spectrum coverage is obtained through fewer sub-apertures, and an imaging effect equivalent to a larger aperture is achieved.

The invention discloses a tunnel water leakage detection method based on refrigeration type infrared thermal imaging. The method comprises the following steps: obtaining a temperature value two-dimensional matrix; traversing each part of original temperature field data, and extracting a feature threshold value; calculating a second-order gradient field by adopting an eight-direction Laplacian operator; extracting a second-order temperature gradient characteristic average value of the low-temperature domain; selecting a second-order temperature gradient upper limit and a second-order temperature gradient lower limit as leakage water quantitative screening criteria based on temperature field second-order gradient characteristics; selecting data meeting quantitative screening criteria, and preliminarily rejecting non-leakage data; for the data meeting the quantitative screening criterion, dividing dry and wet areas of the data by taking the dry and wet boundary temperature as a temperature contour line, and drawing; data which cannot be eliminated through a single temperature field identification mode and contains a regular-shape non-leakage area are eliminated; and outputting the data information containing the leakage water and a corresponding leakage water segmentation result image. The method has the advantage that the tunnel leakage water disease identification accuracy is improved.

The application discloses a GIS partial discharge pattern recognition method based on polar coordinate distribution entropy optimization, and belongs to the field of large power grid on-site power equipment defect diagnosis and recognition. Firstly, aiming at the signal missing phenomenon of photoelectric partial discharge signals, a non-subsampled contourlet transform is used to fuse and process photoelectric PRPD graphs, so that photoelectric fusion graphs are obtained. Then, feature points and scale vectors of the fusion PRPD graphs are extracted based on a KAZE algorithm. Next, the feature points are dispersed to polar coordinate expressions according to coordinates and scale vectors. Distribution entropy of each region divided on the polar coordinates is calculated to form a feature vector. Finally, the feature vector is loaded into a long short-term memory network optimized by an Adaboost algorithm to verify the accuracy of partial discharge pattern recognition. Compared with a statistical parameter method and a KAZE feature extraction method, the algorithm can better extract a PRPD graph feature vector, and the accuracy of partial discharge defect pattern recognition is significantly improved.

The present invention discloses an image segmentation method based on a multi-scale contourlet parallel reverse attention network, which relates to the technical field of image processing and includes: obtaining an image to be segmented; using a feature aggregation module in the multi-scale contourlet parallel reverse attention network to extract the spatial domain features, frequency domain features, and features after frequency domain enhancement of the image to be segmented, and aggregating the features after frequency domain enhancement through a parallel partial decoder in the multi-scale contourlet parallel reverse attention network to obtain a global mapping graph; according to the reverse attention module in the multi-scale contourlet parallel reverse attention network, combining the global mapping graph with the spatial domain features, features after frequency domain enhancement, and frequency domain features to layer by layer obtain reverse attention features, and processing the reverse attention features through an activation function to obtain a prediction graph. The present invention can increase the interpretability of the network model.

The invention discloses a communication line insulator fault detection method and system, and belongs to the technical field of communication line maintenance, and the method comprises the following steps: collecting an image of a communication line insulator; the method comprises the following steps of: preprocessing an acquired original image on the basis of a complex algorithm combination of non-subsampled Contourlet transformation and grey wolf optimization algorithm optimization guide filtering-Retinex; carrying out edge extraction on the preprocessed image; comprehensively extracting geometric and textural features of the insulator from an edge image obtained by edge detection, wherein the geometric and textural features comprise an edge feature, a textural feature and a contour feature; inputting the traditional RGB channel data and the extracted edge feature, texture feature and contour feature vectors into a DeepLabv3 + semantic segmentation model, and identifying an insulator fault; and outputting a fault identification result. According to the invention, rapid and accurate detection of the communication line insulator fault is realized, the maintenance efficiency and safety of the communication line are improved, and the maintenance cost is reduced.

The invention discloses a rapid magnetic resonance image reconstruction method based on a reweighted shrinkage threshold algorithm. The method comprises the following steps: step 1, data acquisition and preprocessing; step 2, sparse transformation and coding matrix construction; step 3, constructing an optimization model; step 4, an iterative reconstruction process; and 5, evaluating and analyzing a result. According to the method, the problem that the balance between the convergence speed and the reconstruction precision is affected due to the fact that regularization parameters are difficult to adjust in real time according to the convergence state in iteration of traditional ISTA is solved; the image detail reduction capability is further enhanced by combining the RISTAC algorithm of contourlet transform, effective improvement on traditional wavelet transform is proved, and beneficial reference is provided for rapid MRI imaging research and practical application.

The invention discloses a raft culture area extraction method based on a high-resolution optical image, and the method comprises the following steps: screening out a preprocessed high-resolution optical image of a target culture area, fusing R, G and B wave bands and a panchromatic wave band through discrete wavelet transform, and obtaining a high-resolution RGB fusion image; performing image enhancement processing on the RGB fusion image and the panchromatic image by using a non-subsampled contourlet transform mode to obtain an enhanced RGB fusion image; inputting the enhanced RGB fusion image and the initial breeding raft into an SAM model to carry out annotation guided prompt training, generating a breeding raft prompt mask and a candidate mask, and reserving a highly similar area of the prompt mask and the candidate mask as a pseudo breeding raft; evaluating the initial breeding raft and the pseudo breeding raft, making raft breeding area verification samples after the precision is met, and realizing automatic parameter adjustment of topology rules by using a genetic algorithm based on the verification samples so as to achieve parameter optimization.

The present invention relates to the field of optical interference imaging technology, and in particular to a high-resolution image fusion method based on a one-dimensional sparse aperture rotating pupil. The method comprises: inputting a target image, imaging the target image using a one-dimensional sparse aperture rotating pupil, and obtaining an image for each rotation to an angle; performing a non-subsampled contourlet transform, and obtaining a low-frequency and high-frequency image for each image; performing equal-weighted superposition and normalization on multiple low-frequency images and high-frequency images; repeating the above steps to construct a data set; establishing a dual-branch residual fusion network; extracting high-frequency branch detail information through a residual learning strategy, and performing a global linear transformation on the low-frequency image; fusing the residual map of the high-frequency branch with the adjusted low-frequency image, and outputting a final high-resolution image. The advantages are: accurate extraction of effective information, streamlining of the data set required for training, obtaining full spectrum coverage through fewer sub-apertures, and equivalent imaging effects of a larger aperture.

The invention discloses a mineral target area delineation method based on electrical resistivity and polarizability image fusion, which comprises the following steps of: firstly, preprocessing actually measured electromagnetic data of an exploration area, including data review, gridding registration and image normalization, so as to respectively obtain a standardized electrical resistivity image and a standardized polarizability image; decomposing the two source images into low-frequency sub-band images and high-frequency sub-band images by using non-subsampled contourlet transform; fusing the low-frequency sub-band image and the high-frequency sub-band image by using a fusion rule based on weighted average and a new measurement parameter; and finally, a fusion image is obtained through NSCT inverse transformation, and the fusion image is compared with prior geological data to delineate a prospecting target area. Through a multi-scale image fusion technology, complementary information of resistivity and polarizability is effectively integrated, the accuracy and reliability of mineralization abnormal area identification are remarkably improved, and a new technical means is provided for mineral exploration.

The invention discloses a terrain geomagnetic fusion evolution synthesis feature construction method based on a gene expression programming method. According to the method, on the basis of the genetic algorithm iteration thought, the fixed chromosome length is the given adaptive feature number, multiple basic adaptive features of two types of maps including a geomagnetic map and a topographic map are combined, the expression evolution direction is associated with the map matching probability according to the training-testing thought, and the optimal terrain / geomagnetic evolution synthesis feature is obtained; the method has good generalization ability, adopts a genetic algorithm iterative optimization thought, combines data preprocessing, feature extraction and gene expression programming, automatically generates an optimal feature expression tree, and reduces the complexity and subjectivity of manual feature design. Through a fusion strategy of weight averaging and new measurement parameters, in combination with non-subsampled contourlet transform and inverse transform, low-frequency and high-frequency information of terrain and geomagnetic data is effectively integrated, so that a fused image has high resolution and anti-interference capability.

The application discloses a brain image fusion method based on convolution sparse representation, which takes an adaptive convolution sparse algorithm as a fusion framework, takes brain images as training samples to obtain a pre-training dictionary, and takes the brain images as test samples to participate in a fusion process; a non-subsampled contourlet transform (NSCT) is adopted to construct a new activity measure in combination with sub-band correlation; finally, a fused brain image is obtained through two-scale reconstruction; through the above method, the influence of local mismatch and texture structure loss on convolution sparse brain image fusion can be effectively improved, and the time consumption of the algorithm can be significantly reduced.

The application discloses a NSCT domain sonar image denoising method and device based on neutral set and bilateral filtering, and belongs to the technical field of sonar image denoising. In order to solve the problems of poor denoising effect, insufficient edge preservation ability and low operation efficiency of the existing image denoising method, the application first decomposes a noisy image based on a non-subsampled contourlet transform, then converts high-frequency subband images to a neutral set domain, uses a true subset, a false set and an uncertain set for representation, performs beta enhancement on the image of the true subset, and performs bilateral filtering processing on the enhanced true subset image; meanwhile, the low-frequency subband image is converted to the neutral set domain, the true subset image is extracted as a guide image of joint bilateral filtering, and the joint bilateral filtering is guided to perform smoothing processing on the low-frequency subband image; the NSCT inverse transform is used to reconstruct the high-frequency subband image after the bilateral filtering processing and the low-frequency subband image after the joint bilateral filtering processing, so that the finally denoised image is obtained.

The application discloses a radar signal recognition method based on time-frequency image feature fusion, and mainly solves the problem of low radar signal recognition rate in the prior art under a low signal-to-noise ratio environment.The steps of the application are as follows: (1) time-frequency transformation is performed on a radar signal to obtain a time-frequency image; (2) time-frequency image pre-processing; (3) texture feature and shape feature extraction of the image; (4) support vector machine is used to recognize the radar signal.The application extracts the texture feature and shape feature of the radar signal based on the Choi-Williams time-frequency image and the contour map of the fuzzy function, and fuses the two to form a new radar signal recognition characteristic value, thereby solving the problem of low radar signal recognition rate under a low signal-to-noise ratio.

Taking advantage of the strong robustness of the frequency-domain digital watermarking algorithm, the present invention discloses a blind watermarking method for color digital images based on Contourlet transform and Schur decomposition. According to the characteristics of multi-scale, multi-directionality, translational invariance of the Contourlet transform and low complexity of Schur decomposition, the present invention first performs hierarchical processing on the host image and conducts Contourlet transform on the hierarchical image. Then, the selected coefficient blocks are subjected to Schur decomposition to obtain an upper triangular matrix, and the embedding and blind extraction of the color digital watermark are completed by quantifying the first main diagonal element of the upper triangular matrix. The present invention belongs to the technical field of cyberspace security, not only has good watermark concealment, but also has the characteristics of strong robustness and high security, and is suitable for occasions where color image copyright protection is carried out quickly and efficiently.

The invention discloses a ground penetrating radar shallow target detection method and device and a medium, and the method comprises the steps of data collection, zero offset correction processing, median taking from data and the like. The contour line algorithm can completely and clearly describe the forms of the target and the weaker target so as to extract the complete edge curve of the target. A traditional lower opening scanning algorithm needs to distinguish lambda-shaped, X-shaped and V-shaped intersection, but the method can avoid the difficult points, and only needs to find point segments conforming to normal lower opening characteristics.

The invention discloses a conveyor belt longitudinal tear detection system based on binocular image fusion. The system comprises an original image acquisition module, an original image processing module, an image decomposition module, an image fusion module and a detection module. The binocular camera is used for collecting visible light and infrared images of the conveying belt, clear images are obtained through denoising, filtering, registering and other processing, and the images are decomposed into high-frequency sub-band signals and low-frequency sub-band signals through non-subsampled contourlet transformation. And the image fusion module fuses the sub-band signals of different frequency bands through a maximum entropy threshold segmentation method and a weighted average method to generate a fused image. The improved YOLOv5 model is used for accurately detecting tearing in the detection module, and the model improves the detection precision of tearing features by adding depth separable convolution and an SE attention mechanism. The system can effectively cope with the influence of factors such as illumination change, noise interference and material conveying, and the stability and robustness of a detection result are improved.

The invention discloses a sound source position and frequency joint characterization method in acoustic imaging, and belongs to the field of microphone array signal processing, and the method comprises the steps: collecting time domain data through a microphone array, converting the time domain data into a plurality of sub-band data through short-time Fourier transform, and constructing an observation model; establishing a sparse Bayesian learning model based on the observation model, performing probability modeling on noise and signals, and setting prior distribution; utilizing array data to iteratively update model parameters until convergence; acquiring a position-frequency energy matrix according to the convergence parameter; a sound source position is determined according to a non-zero line in the position-frequency energy matrix, a frequency sequence is generated for the sound source position, a joint visual graph in a concentric circle or contour line form is constructed, each geometric layer corresponds to a sub-band, and a visual attribute represents an energy amplitude; and finally, superposing the visual graph and the video image. According to the invention, joint visual display of the sound source position and the frequency is realized, and frequency characteristic monitoring and sound source distinguishing are supported.

The invention discloses a robust zero watermarking algorithm based on NSCT and UPEMD-PHT. The method comprises the following steps: step 1, carrying out Tent chaotic encryption on a watermark image; step 2, embedding a watermark; step 3, extracting a watermark; and 4, decrypting the watermark image. According to the method, before a carrier image is embedded, encryption of a watermark image is carried out based on Tent chaotic mapping so as to ensure the security of the watermark, non-subsampled contourlet transform (NSCT) is carried out on a host image, low-frequency information is extracted, and then uniform phase empirical mode decomposition (UPEMD) and polar harmonic transform (PHT) are carried out on the low-frequency information so as to complete watermark embedding. According to the invention, a chaotic mapping algorithm and a novel watermarking technology are organically combined, and the method for embedding the digital zero watermark in the digital image is realized.

This invention discloses a device and method for accurately acquiring information on anomalies at a working face using multi-frequency ground-penetrating radar (GPR). The device includes a multi-frequency data acquisition module, a data alignment module, a data fusion module, a data management module, a model training and optimization module, and an anomaly information identification module. The method involves: acquiring GPR data at different center frequencies of the target using multi-frequency GPR; preprocessing the raw multi-frequency GPR data; spatially aligning the preprocessed GPR data; fusing the GPR data using an improved non-subsampled contourlet transform (NSCT) to obtain a multi-frequency GPR fused profile; collecting the fused images to create a GPR image dataset; and importing an improved YOLOv11 model to identify and label underground anomalies in the GPR image dataset. This invention significantly improves the accuracy and efficiency of anomaly detection through multi-frequency data complementarity and deep learning optimization, making it particularly suitable for geological exploration in complex underground environments such as coal mines.

The invention discloses a whole-body SPECT lung cancer bone imaging data processing method and system, and relates to the technical field of image processing, and the method comprises the steps: inputting whole-body SPECT lung cancer bone imaging data into a trained improved UNet network, extracting local texture features through a similar contourlet decomposition module, fusing the local texture features with an original image, and obtaining a whole-body SPECT lung cancer bone imaging data fusion model; inputting the fusion features into a UNet encoder to extract image features; and acquiring a lung cancer bone metastasis focus area by using a UNet decoder added with a contourlet feature splicing and reconstruction strategy for the extracted image features. According to the invention, the feature extraction capability of the bone imaging lesion segmentation method under the conditions of low resolution and fuzzy boundary can be improved.

The invention provides a seismic data denoising method based on Contourlet transformation, and the method is characterized in that the method comprises the following steps: 1, carrying out the loading and preprocessing of seismic data; step 2, denoising is carried out by adopting multi-scale cyclic translation Contourlet transformation; step 3, adopting improved non-local mean filtering; and 4, processing and outputting data. According to the method, a cyclic translation algorithm and non-local mean filtering (FNLM) are combined, so that the denoising capability and weak signal identification capability of seismic signals are effectively improved.

The present invention discloses an interactive image segmentation method based on multi-scale geodesics, comprising the following steps: step S100: inputting a grayscale image I: Ω→R, wherein Ω is the image domain of the grayscale image I and R is a real number space; adding noise to the grayscale image I to obtain a noisy grayscale image I0; step S200: performing pyramid decomposition on the noisy grayscale image I0 using a contourlet transform; step S300: performing threshold processing on the contourlet coefficients to reconstruct a denoised image I n (n≥2, n∈Z); Step S400: Calculate the denoised image I n The segmentation results of geodesics at different scales are shown in Figure 2. The multi-scale strategy and contourlet threshold are used in the minimum path model to reduce the influence of noise on image segmentation; and the computational complexity of the minimum path model is reduced by finding geodesics from coarse to fine.

The invention relates to the technical field of image sharpening, in particular to an archive digital image processing method and system.The method comprises the following steps that according to archive page scanning data, a pyramid filter bank and a direction filter bank are called to execute non-subsampled contourlet decomposition, and a multi-scale direction sub-band coefficient set is generated; and dividing the multi-scale direction sub-band coefficient set into overlapped sliding processing units according to preset step length and size parameters, and generating a coefficient sliding window block set. According to the invention, while the sharpness and contrast of the stroke edge are enhanced, the background region coefficient is maintained or suppressed, and synchronous amplification of noise is avoided. And inverse transformation reconstruction is performed on the direction coefficient set subjected to texture adaptive enhancement processing and the low-frequency component, and stroke details and background intensity in the multi-direction feature data stream are fused, so that a restored image with high definition and historical texture can be output, and the visual readability and subsequent character recognition accuracy of a digital archive finished product are improved.