365 results about "Discrete wavelet transform" patented technology

The invention relates to the field of image data processing, and discloses a texture preserving type image denoising and enhancing method based on a generative adversarial network, which comprises the following steps: acquiring an original image signal, and calculating low-frequency sub-band data and high-frequency sub-band data by using discrete wavelet transform; calculating the gradient magnitude of the low-frequency sub-band data to generate a structural significance gradient map; establishing a reverse mapping relation based on the structure saliency gradient map, and generating a spatial self-adaptive dynamic gating threshold; performing statistical gating on the high-frequency sub-band data by using the dynamic gating threshold to generate a high-pass gain coefficient and a low-pass suppression coefficient; according to the method, cross-band modulation logic of the structure flow to the texture flow is established, so that the problem that weak texture signals are easy to lose under non-uniform illumination is solved, and non-structured noise filtering and structured micro texture restoration are realized on the premise of not depending on semantic tags.

The invention relates to the technical field of mode recognition and data recognition, and discloses an intelligent visual detection method for non-standard precision part surface microdefects, which comprises the following steps: acquiring surface gray level image data of a to-be-detected part, physically abandoning low-frequency components through discrete wavelet transform, and reserving high-frequency detail components to construct a frequency domain input tensor; constructing a double-flow reconstruction model containing a space domain coding network and a frequency domain coding network, and minimizing the distribution difference of the same feature between double-domain characterization through potential feature space consistency constraint joint optimization; the method comprises the following steps of: calculating a spatial domain residual image and a frequency domain residual image, combining a texture topological residual image extracted by structural tensor characteristic decomposition, and generating a comprehensive abnormal response image through weighted fusion to judge the defect, and effectively inhibiting macroscopic geometric contour interference through frequency domain decoupling and a topological check mechanism on the premise of not needing a standard geometric template. And sensitive perception and accurate identification of weak texture defects on the surface of the non-standard part are realized.

The invention provides a dense overlapping target detection method based on a wavelet enhancement sparse hybrid expert model. The method comprises the following steps: firstly, extracting multi-layer features through a backbone network to capture multi-scale spatial representation; secondly, discrete wavelet transform is introduced to each level of features, spatial features are decomposed into a frequency domain, collaborative modeling of frequency domain and spatial domain features is realized, the reservation capability of detail and texture information is improved, a lightweight dynamic hypergraph aggregation module is introduced into the deepest layer of features, a hyperedge structure is adaptively learned, and the feature fusion is realized; modeling a high-order incidence relation in a local area in an explicit manner; and thirdly, in the decoding process, candidate queries are screened and reweighted through an IoU perception query selection mechanism, and a dynamic routing mechanism of sparse hybrid experts is introduced, so that query self-adaptive specialized representation learning is realized, and the target detection precision and reliability in a complex scene are effectively improved.

The invention belongs to the technical field of wastewater treatment, and provides a wastewater treatment equipment remote control system based on the Internet of Things, which aims at solving the problems that the traditional fixed DO concentration control cannot adapt to water inlet load fluctuation, the effluent is easy to be substandard or the aeration energy consumption is high, the monitoring data noise is large, and the time sequence is misplaced. According to the scheme, a multi-dimensional real-time sensing network is constructed, flow, COD, BOD, ammonia nitrogen, water temperature, distributed DO and sludge activity sensors are deployed, discrete wavelet transform is adopted for noise reduction, and a data time sequence is aligned; a three-layer load-DO-energy consumption dynamic correlation model is designed, a basic layer predicts a load trend through LSTM, a middle layer quantifies a DO demand through a microbial metabolism model, and an optimization layer outputs an optimal DO set value through a PPO algorithm; and developing a dynamic decision-making system, adaptively generating a DO interval according to a load state, and adjusting fan parameters and number in a linkage manner. According to the invention, load dynamic adaptation is realized, effluent ammonia nitrogen is guaranteed to reach the standard, aeration energy consumption is reduced, and remote control precision and equipment operation efficiency are improved.

The invention provides an image privacy protection-oriented reversible adversarial sample generation system and method, and relates to the technical field of artificial intelligence and image privacy protection. The system comprises an input module, a semantic guidance mask generation module, a potential confrontation generation module and a reversible steganography embedding module. The input module acquires an original image, generates a binary mask in the semantic guidance mask generation module, and generates adversarial text description by adopting a visual language model; a final image potential representation is generated in a potential adversarial generation module through forward diffusion and reverse sampling of a diffusion model, and adversarial disturbance is introduced by adopting a gradient optimization algorithm to generate an adversarial sample image; and finally, discrete wavelet transform is adopted in the reversible steganography embedding module to separate out a high-frequency sub-band, and a reversible neural network is combined to obtain a secret-containing hidden image. According to the method, the three requirements of visual authenticity, visual anonymity and machine identifiability can be met at the same time, and an innovative solution is provided for privacy protection in an advanced face recognition scene.

The invention belongs to the field of photovoltaic panel hot plate image processing, and particularly relates to a photovoltaic panel defect detection method fusing multi-scale wavelets and a lightweight attention mechanism. According to the method, an RHDWT discrete wavelet transform module based on multi-scale decomposition is added in a model input stage to strengthen image edge and texture detail representation; a Mix Structure Block module is introduced into a backbone network of the YOLOv11, so that multi-scale features are fused, and the feature expression capability is improved; an LWGA lightweight global attention mechanism is introduced into a neural network connection layer to enhance the context modeling capability, and the detection effect on small target defects such as fine cracks and hot spots is improved. According to the model, through collaborative optimization in three aspects of input preprocessing, feature extraction and an attention mechanism, the precision and robustness of defect detection in a complex photovoltaic module infrared or visible light image are remarkably improved, and the model is suitable for scenes such as high-precision photovoltaic panel image detection and intelligent maintenance.

The invention discloses a parallel U-Net-based dual-domain collaborative infrared and visible light image fusion method. The method comprises the following steps: firstly, extracting initial features of a source image by using dense connection blocks; then, parallel frequency domain branches and spatial domain branches are constructed, the frequency domain branches are combined with discrete wavelet transform and fast Fourier convolution to decompose and enhance multi-scale global frequency domain features, and the spatial domain branches capture long-distance spatial dependence with linear calculation complexity by using a convolutional layer and Mama based on a selective state space model; dynamic interaction and weighted fusion of double-domain information are realized through an adaptive feature fusion module; and finally, generating a fused image through an image reconstruction module. According to the method, the problems of high calculation overhead and video domain information negligence in the prior art are solved, and infrared heat radiation maintenance and visible light texture enhancement are effectively considered.

The invention discloses a traffic flow prediction method and system based on TEDDGN, and belongs to the technical field of traffic flow prediction, and the method comprises the steps: carrying out the convolution operation of a traffic time sequence and a wavelet function through employing discrete wavelet transform, achieving the trend-event decoupling of traffic flow data, projecting a signal to different scale spaces, and carrying out the prediction of the traffic flow. Obtaining independent trend components and event components; processing the obtained trend component and event component by adopting a multi-scale time learner; spatial feature extraction is carried out on the processed data, modeling is carried out through adaptive graph convolution and dynamic graph convolution in spatial feature extraction, and the adaptive graph convolution and the dynamic graph convolution are dynamically fused and output through a learnable gating coefficient; inputting the output spatio-temporal characteristics into an output layer to form a TEDDGN prediction model; a TEDDGN prediction model is trained; and performing traffic flow prediction by using the trained TEDDGN prediction model. According to the invention, the accuracy of traffic flow prediction is realized.

According to the invention, the multi-mode time-frequency domain conversion and the deep learning technology are combined, and a multi-component mixed Raman spectrum unmixing method is developed, so that clinical in-vivo and in-situ detection and disease diagnosis of novel Raman probes, instruments and the like are facilitated. The method comprises the following steps: (1) converting a mixed Raman spectrum from a time domain to a frequency domain by using fast Fourier transform (FFT), discrete cosine transform (DCT) and discrete sine transform (DST), and extracting frequency domain features; (2) extracting multi-scale local time-frequency domain characteristics of the mixed spectrum by using short-time Fourier transform (STFT) and discrete wavelet transform (DWT); (3) carrying out spectral unmixing calculation in each mode in combination with a one-dimensional attention mechanism U-shaped neural network model; and (4) fusing various modal unmixing results by using a meta-learning method, and analyzing the weight of each modal to obtain an accurate unmixing spectrum. Compared with a traditional Raman spectrum analysis method, the Raman spectrum multi-component signal unmixing method based on multi-modal time-frequency domain transformation and deep learning can accurately separate independent Raman signals of different tissue structures and biochemical components in a complex environment in a living body, so that the unmixing accuracy of the Raman spectrum multi-component signal is improved. Therefore, convenience is provided for subsequent disease mechanism analysis and diagnosis. The method provides an innovative and potential solution for in-vivo and in-situ detection analysis and disease diagnosis of medical clinical Raman spectroscopy.

The invention relates to the cross technical field of artificial intelligence and medical image processing, in particular to a medical image segmentation system and method based on a wavelet bridge diffusion model and an efficient conditional random field. A WBDM-ECRF framework is constructed and comprises a discrete wavelet transform module, a BDM-T module, a BDM-S module and an ECRF module; decomposing the image through discrete wavelet transform, extracting a low-frequency sub-band, and enhancing the contrast ratio of a focus and normal tissues; the BDM-T takes U-Net as a backbone, integrates a FlashAttention mechanism, and optimizes a variance formula to realize efficient training; the BDM-S adopts a leapfrog sampling strategy, so that the reasoning time is greatly shortened; the ECRF introduces a multivariate potential function of a structural similarity index and smooth operation through edge expansion, and accurately optimizes edge segmentation. According to the method, the dependence of marked data is reduced, the training and reasoning efficiency is improved, the problem of fuzzy edge segmentation is solved, the Dice coefficient and intersection-union ratio performance on the ISIC data set is excellent, and reliable quantitative support is provided for disease diagnosis and treatment.

The invention discloses a generative image steganography method and device based on Stable Diffusion and discrete wavelet transform, and the method comprises the following steps: S1, sampling a potential representation of a to-be-generated image through employing a Stable Diffusion diffusion model, embedding secret information into a potential space of the diffusion model, processing the potential representation through employing discrete wavelet transform to carry out frequency domain information modulation, and carrying out the frequency domain information modulation; obtaining the potential representation after the secret information is embedded; s2, inputting the potential representation embedded with the secret information into a diffusion model, and generating a visual natural steganographic image through a diffusion inversion process; s3, inputting the received steganographic image into a diffusion model, and decrypting the received steganographic image without original model weight modification and empty prompt conditions to obtain embedded secret information; according to the method, the frequency domain embedding technology is combined with the potential diffusion model, so that high-fidelity, high-capacity and high-robustness image steganography is realized.

The invention discloses a priori knowledge and frequency domain perception-based cell micronucleus detection method, and relates to the technical field of computers. The method comprises the following steps: extracting cell micronucleus features of a to-be-detected cell image in a multi-level and multi-angle manner based on priori knowledge of cell micronucleus; the cell micronucleus characteristics comprise the volume of the cell micronucleus, the form of the cell micronucleus and the distribution area of the cell micronucleus; performing continuous channel compression on the cell micronucleus features, and extracting frequency domain information of the cell micronucleus features after channel compression through discrete wavelet transform; the frequency domain information comprises high-frequency information and low-frequency information; convolution processing is carried out on the high-frequency information and the low-frequency information, and convolution processing results are fused to obtain frequency domain sensing fusion features; analyzing the frequency domain perception fusion feature to obtain a plurality of bounding boxes in the to-be-detected image; the bounding box is the predicted position of the cell micronucleus. The method can improve the detection precision of the cell micronucleus.

The invention discloses a time prediction method and system based on double-domain feature fusion. The method and system adapt to long and short term time series prediction requirements of multiple scenes such as weather forecast, energy scheduling, traffic flow and financial exchange rate. The method comprises the steps that a multi-field data set is obtained and preprocessed, and instance normalization is carried out; performing double-domain multi-scale characteristic decomposition by adopting down-sampling and discrete wavelet transform to obtain continuous trend and high-frequency mutation details; a local unit is obtained through patch cutting and embedding, local time sequence association is mined through depth separable convolution, cross-patch global interaction is achieved in combination with a multi-layer perceptron, and local-to-global progressive fusion is completed; and constructing bidirectional attention flow enhanced cross-domain and cross-scale collaboration, and combining with standardized data training to obtain a prediction model. According to the invention, the method can improve the depiction capability of non-stable and non-linear complex time sequence data containing abrupt change and multi-period superposition, gives consideration to the adaptability of long and short term prediction, remarkably improves the accuracy of multi-field time sequence prediction, and promotes the application of the prediction technology in multiple scenes.

The invention provides a multivariate time sequence prediction method and system and a medium, an input time sequence is adaptively decomposed into an approximate coefficient sequence and a detail coefficient sequence by adopting an adaptive discrete wavelet transform and inverse transform mode, and compared with the previous wavelet transform depending on a predefined wavelet basis function, the multivariate time sequence prediction method and system have the advantages that the time sequence prediction efficiency is improved; the mode is more flexible, and filtering kernels of discrete wavelet transform and inverse transform can be updated in a data driving mode in the training process according to the characteristics of the time sequence, so that the filtering kernels are more suitable for the current sequence. Wavelet transformation realized by carrying out convolution operation on a time dimension lacks extraction and modeling of correlation between global information and different variables in a multivariate time sequence; therefore, a coefficient mixing module and a wavelet domain attention channel enhancement module are provided to carry out supplementary modeling on an approximation coefficient and detail coefficient sequence obtained after wavelet transformation. The method can meet the prediction requirements of the time series in actual production such as power load.

The invention discloses a method and system for constructing an interferometric phase unwrapping model based on discrete wavelet transform, and relates to the field of image phase unwrapping, and the method comprises the specific steps: constructing an InSAR synthetic data set based on an SRTM DEM, a two-dimensional Gaussian curved surface and an InSAR measurement error source; an interferometric phase unwrapping model based on discrete wavelet transform is constructed, the interferometric phase unwrapping model comprises an encoder network, a bottleneck layer, a decoder network and an output layer which are connected in sequence, and the encoder network comprises a plurality of cross-stacked dual-channel data down-sampling modules and pyramid pooling modules; and training the interferometric phase unwrapping model based on the InSAR synthetic data set, and optimizing model parameters by minimizing a loss function to obtain a final interferometric phase unwrapping model. According to the method, the interferometric phase unwrapping of the low-coherence region is realized under the condition of lightweight parameters, the phase continuity hypothesis is not depended on, and the interferometric measurement precision of the synthetic aperture radar is improved.

The invention discloses a wide-spectrum interference weak signal enhancement method for high aspect ratio micro-nano structure measurement, and relates to the technical field of semiconductor measurement. The method comprises the following steps: acquiring a wide-spectrum interferogram, and performing two-layer two-dimensional discrete wavelet transform to obtain a first-stage wavelet coefficient group; performing nonlinear enhancement on the wavelet coefficient, and outputting a nonlinear enhanced wide-spectrum interferogram; performing three-layer two-dimensional discrete wavelet transform on the wide-spectrum interferogram after nonlinear enhancement to obtain a second-stage wavelet coefficient group, and calculating a Bayesian threshold value; a neighborhood sliding window function is set, and a Bayesian threshold is combined to carry out local denoising enhancement on the second-stage wavelet coefficient; reconstructing the second-stage wavelet coefficient after local de-noising enhancement to obtain a wide-spectrum interference pattern after local de-noising enhancement; and determining an optical path position by adopting an improved gravity center method according to the wide-spectrum interferogram after local denoising enhancement, and calculating a depth value of the to-be-measured sample. According to the method, the stripe contrast is enhanced, the noise is suppressed, and the depth measurement capability is improved at the same time.

The invention discloses a wavelet attention shadow removal method based on soft prior guidance, and belongs to the field of image processing. The method comprises the following steps: constructing a multi-scale wavelet attention codec architecture, decomposing an input image into a low-frequency component representing illumination characteristics and a high-frequency component containing texture details by using discrete wavelet transform, and establishing multi-scale frequency domain representation; proposing probabilistic soft prior modeling, adaptively generating a continuous differentiable probability distribution diagram by using a convolutional sub-network and a double-slope Sigmoid function, and guiding differentiated repair of high and low frequency features in a dynamic weight mode; and cross-domain illumination correction based on gray world hypothesis is introduced, spatial domain enhancement is performed on a frequency domain restoration result, illumination consistency of cross-domain processing is ensured, and finally a shadow-free image is output. According to the method, wavelet domain feature optimization is guided through the probabilistic soft mask, high-quality shadow removal is realized, and the robustness and visual quality of shadow removal are remarkably improved while the light weight of the model is kept.

The invention discloses an abnormal heart beat detection method and system based on a magnetic cardiogram, and relates to the technical field of medical data processing, and the method comprises the steps: carrying out the preprocessing of original magnetic cardiogram data; magnetic cardiogram features are extracted based on discrete wavelet transform, model parameters are adjusted, and an abnormal heart beat detection model is obtained; and judging whether abnormal heart beat exists or not and the position. According to the method, the problem that the magnetocardiogram data lacks a large number of heart beat level labeling samples is effectively solved, dependence on labeling data is reduced, automatic, efficient and accurate detection of abnormal heart beats is achieved, the good clinical application prospect is achieved, and powerful support can be provided for early screening, auxiliary diagnosis and risk assessment of arrhythmia.

The application provides a method for identifying mud shale fractures based on conventional logging data, comprising the following steps: step 1, obtaining acoustic travel time logging curves and natural gamma logging curves of a mud shale section; step 2, performing discrete wavelet transform and box dimension calculation on the acoustic travel time curves and the natural gamma curves to obtain wavelet coefficients and box dimensions; step 3, calculating the difference between the wavelet coefficients of the acoustic travel time curves and the wavelet coefficients of the natural gamma curves to obtain wavelet coefficient differences, and calculating the difference between the box dimensions of the acoustic travel time curves and the box dimensions of the natural gamma curves to obtain box dimension differences; step 4, averaging the wavelet coefficient differences and the fractal dimension differences to construct a fracture development response index; and step 5, identifying fractures in the mud shale section according to the relatively high value of the fracture development response index. The method for identifying mud shale fractures based on conventional logging data can accurately and reliably identify mud shale fracture development sections.

The invention discloses a component multi-dimensional information measurement method and system based on multi-source information fusion, and the method comprises the steps: 1, obtaining a fluorescence speckle image and a texture image before deformation, and decomposing the two images into low-frequency features and high-frequency features through discrete wavelet transform; 2, fusing low-frequency features of the fluorescence speckle image and the texture image before deformation; fusing the high-frequency features of the fluorescent speckle image and the texture image before deformation; 3, reconstructing the fused low-frequency features and high-frequency features through inverse wavelet transform to obtain a fused image before deformation; 4, acquiring a fused image after deformation; 5, the three-dimensional shape of the component is obtained through an FPP method; then calculating a sub-pixel-level displacement field, and constructing a three-dimensional displacement field according to the three-dimensional morphology and the sub-pixel-level displacement field; and extracting a partial derivative of the three-dimensional displacement field to the pixel coordinate, and calculating the Green strain tensor. According to the invention, the interference of environmental factors on displacement and strain measurement is effectively overcome, and the measurement precision and robustness are improved.

The invention relates to the technical field of analog current fault diagnosis, in particular to an analog current fault diagnosis method based on redundancy feature suppression, which comprises the following steps: acquiring voltage and current data of a circuit output position according to a circuit simulation model, and constructing a circuit fault state data set according to the voltage and current data; performing time-frequency analysis on the circuit fault state data set through discrete wavelet transform; performing redundant feature suppression processing on the hierarchical frequency domain component set to obtain a circuit fault state data feature set; a DCNN-BiLSTM model is constructed, the DCNN-BiLSTM model is trained according to the circuit fault state data feature set, and the DCNN-BiLSTM model meeting a preset fault diagnosis correct rate is output as an analog power fault diagnosis model; and obtaining target to-be-detected voltage and current data, and performing analog current fault diagnosis on the target to-be-detected voltage and current data according to the analog current fault diagnosis model to obtain a corresponding fault type classification result. According to the invention, redundant features can be effectively suppressed, and the accuracy of analog power supply fault diagnosis is remarkably improved.

The invention discloses an infrared corn drought identification method based on wavelet boundary enhancement, and belongs to the field of agricultural informatization and plant phenotype identification, and the method comprises the following steps: S1, obtaining an infrared image of a corn plant; s2, executing two-dimensional discrete wavelet transform to generate a boundary response diagram; s3, generating a binary mask according to the characteristics of the boundary response diagram to the blade edge; s4, connected domain analysis is executed, candidate leaves are obtained, and independent single-leaf masks are generated; s5, the small holes are removed, a mask is obtained, and the blade curvature is quantified; s6, constructing a multi-dimensional feature vector reflecting the curling characteristics of the blade; and S7, outputting the drought grade corresponding to the corn leaf through the multi-layer perceptron model. By adopting the method, automatic identification and early warning of the early-stage water shortage state of the corn are realized.

The invention relates to the field of electrocardiosignal processing, and particularly discloses an electrocardiosignal classification method, device and equipment based on multi-mode electrocardiosignal characteristics and a medium thereof.The method comprises the steps that firstly, an original electrocardiosignal is preprocessed, including denoising, resampling and heartbeat segmentation; then three types of complementary features are extracted in parallel: wavelet energy features are extracted through discrete wavelet transform, deep abstract features are extracted through a neural network comprising a multi-scale residual block, a Transform encoder and a windowed global-local attention module, a knowledge graph is constructed based on clinical prior knowledge, and clinical concept features are extracted through a graph convolutional network; adaptive weighted fusion is carried out on the three types of features through a multi-modal feature fusion module; and finally, classification is performed based on the fusion features, and an unsupervised domain adaptation strategy combining antagonism domain alignment and class condition alignment is adopted in the training process. The method effectively solves the problems that a traditional method is single in feature and insufficient in cross-domain generalization ability.

The invention relates to the technical field of image processing, and discloses a warping process monitoring method and system based on image recognition, and the method comprises the steps: obtaining a yarn image in the operation process of a warping machine, and inputting the image into a monitoring model; generating a direction gradient prior graph, and calculating a structure tensor representing yarn texture complexity; discrete wavelet transform is carried out on the preorder layer feature map of each dense layer, and a high-frequency detail component and a low-frequency contour component are obtained through separation; performing weighted fusion on the high-frequency detail component and the low-frequency contour component to generate a fusion feature map; performing channel splicing on the directional gradient prior image and the fusion feature image, and inputting a splicing result into a sub-pixel convolution layer for up-sampling to obtain a high-resolution monitoring image; calculating a local anisotropy index of each pixel; and the local anisotropy index and the pixel gray value are combined to identify the yarn broken end or hairiness defect, and a monitoring result is output. According to the method, the reliability of detecting tiny defects such as broken ends and hairiness under a complex background is improved.

The invention discloses a remote sensing image hiding and recovering method and device based on space-frequency collaborative modeling. The method comprises the following steps: acquiring a to-be-hidden remote sensing image and a carrier image; inputting a to-be-hidden remote sensing image and a carrier image into the pre-embedded residual attention module to obtain condition features, and inputting the condition features into the image hiding network to obtain a secret-containing image; performing discrete wavelet transform on the secret-containing image to obtain a high-frequency feature and a low-frequency feature, and inputting the high-frequency feature and the low-frequency feature into a Gaussian cross-frequency fusion module to obtain a dynamic fusion feature; and inputting the dynamic fusion features into a reconstruction recovery module to obtain a secret image. According to the method, the fidelity, the robustness and the detection resistance of the secret image can be improved.

The invention discloses a multi-modal medical image processing method and device, a storage medium and computer equipment. Comprising the following steps: performing three-dimensional discrete wavelet transform on a brain MRI image of a target patient to generate an MRI wavelet coefficient; inputting the MRI wavelet coefficient into a diffusion model to obtain a reference PET wavelet coefficient of the brain of the target patient in a healthy state; performing inverse wavelet transform on the reference PET wavelet coefficient to generate a reference PET image; and comparing the brain PET image of the target patient with the reference PET image, and determining the metabolic deviation index of the brain of the target patient. Therefore, each patient can take the condition without the neurodegenerative change as a contrast, space standardization does not need to be carried out on a group template, anatomical structure distortion caused by the space standardization is greatly reduced, voxel-level accurate analysis of the neurodegenerative disease is realized, tiny pathological change aiming at the patient can be identified, and the accuracy of voxel-level accurate analysis of the neurodegenerative disease is improved. And clinical doctors are assisted in early diagnosis.

The invention provides a dynamic sequence recommendation mechanism for multi-scale wavelet transform of an intelligent maritime reconnaissance instrument, which belongs to the technical field of communication information services and comprises an embedded layer, a multi-scale wavelet decomposition layer, a wavelet neural network layer, a multi-view contrast learning module and a final recommendation layer. According to the method, the discrete wavelet transform is introduced to replace the traditional Fourier transform and discrete cosine transform, so that the dynamic interest change in the user behavior sequence is more accurately captured. According to the method, time and frequency localization analysis can be carried out on the signals at the same time, and the method is especially good at processing non-stable user behavior data containing mutation, so that long-term stable interests and short-term sudden interests of users are extracted and distinguished on different time scales. In addition, a wavelet neural network and an enhanced multi-view contrast learning mechanism are introduced, and the feature processing ability, generalization ability and recommendation precision of the model are further improved.

The invention discloses an anti-multi-attack medical image robust watermarking method based on a Mama architecture, which comprises the following steps: an original medical image and watermark information are processed through an encoder network to generate a watermark-containing image, and an encoder comprises a hierarchical feature refined sampler module and a channel perception Mama architecture bottleneck layer; a watermark is recovered from a watermark-containing image which may be attacked through a decoder network, and a decoder carries out dual-path processing through a wavelet domain Mama module and a sequence context dependence module; randomly extracting 2-4 attacks from the attack library to apply dynamic composite attacks to the watermark-containing image in a random sequence and intensity; and performing end-to-end optimization through a composite loss function including reconstruction loss, clean decoding loss, robust decoding loss and perception loss. According to the method, the image frequency band is decoupled through discrete wavelet transform, meanwhile, the watermark is embedded at the high frequency, the long-range dependency relationship of the wavelet coefficient is modeled by using the selective state space mechanism of Mamba, and the robustness of the watermark is improved.

The invention discloses a video depth forgery detection method and system based on unsupervised learning, and the method comprises the steps: extracting frames from a training set according to a set interval to construct a frame image set, and carrying out the clustering of the constructed frame image set based on the texture artifacts of a gray-level co-occurrence matrix, extracting frequency domain features of the clustered frame image by using hierarchical discrete wavelet transform, constructing an attention map by using the obtained frequency domain features to perform RGB domain feature enhancement and model training, performing counterfeit recognition on the to-be-recognized image by using the trained model, obtaining the inter-frame similarity in the to-be-recognized image, and performing counterfeit recognition on the to-be-recognized image by using the inter-frame similarity in the to-be-recognized image. According to the method, deep counterfeit classification is carried out according to the obtained similarity, the problems of high labeling cost and cross-domain failure of a supervision model in a traditional method are solved by exploring the internal difference of data and automatically generating a pseudo label, generalization detection can still be effectively carried out under the condition that a large amount of labeling data does not exist, and the detection efficiency is improved. The dependence on manual annotation is greatly reduced, and the applicability of the model in different fields and scenes is enhanced.

The invention relates to a remote sensing image change detection method based on frequency domain distribution alignment, a program, equipment and a storage medium, and the method comprises the steps: carrying out the preprocessing of two remote sensing images of different time phases in the same region, respectively carrying out the multi-scale feature extraction, obtaining a first feature map of each remote sensing image under each scale, and obtaining a second feature map of each remote sensing image under each scale; projecting to different frequency domains by using two-dimensional discrete wavelet transform to obtain sub-bands corresponding to the frequency domains; subtraction is carried out on the sub-bands of the two remote sensing images in the same frequency domain under each scale to obtain a frequency domain component of the scale; performing deconvolution on the frequency domain component according to a convolution kernel of two-dimensional discrete wavelet transform by adopting inverse wavelet transform, and fusing the frequency domain component into a second feature map of the scale; for each remote sensing image, splicing the first feature map and the second feature map under the minimum scale to obtain a fusion feature map of the remote sensing image; and respectively aligning the fusion feature maps of the two remote sensing images with the original input resolution, connecting the fusion feature maps and generating a change region detection map through a classifier, thereby realizing change region detection of the two remote sensing images of different time phases in the same region.