708 results about "Dual domain" patented technology

The invention discloses a low-dose CT reconstruction method and system based on global optimization iteration deep learning, and the method comprises the steps: collecting CT cross section data of a normal dose, and generating low-dose CT data; the method comprises the following steps: establishing a dual-domain analysis compression iteration model, constructing maximum posteriori estimation based on a CT projection chordal graph data composite Poisson noise generation mechanism, subdividing CT projection data into a projection domain and a chordal graph domain, sequentially updating the projection data, the chordal graph data and image data, and constructing an iteration algorithm to realize CT global optimization modeling. Expanding an iterative algorithm into a trained reconstruction network; inputting the paired data into the expanded network for training, and storing a model with the minimum output result loss; a sub-network and an attention mechanism are added for CT data features, jump connection weights are finely adjusted, middle layer neural network parameters are finely adjusted and updated through middle supervision, and a global optimization iteration deep learning model NGACI-Net is obtained. According to the method, the problems that the low-dose CT image quality is poor, the reconstruction efficiency is low, and generalization and interpretability are lacked are solved.

The invention relates to the technical field of image denoising, and particularly discloses a dual-domain heterogeneous image denoising method, which comprises the following steps: extracting a preliminary feature map based on depth separable convolution; an encoder of a double-domain heterogeneous cooperative architecture is adopted in a shallow layer of a hierarchical double-drive encoding and decoding architecture to perform double-domain heterogeneous cooperative processing, through hierarchical feature adaptation, the shallow layer gives consideration to details and local structures, a deep layer focuses on global semantics, and dynamic allocation of computing resources is performed, so that the redundant computing burden is remarkably reduced; and splicing the processed image frequency domain information and the image space domain information, fusing features of each layer after hierarchical processing based on a vertical stripe perception fusion attention mechanism module connected between an encoder and a decoder in a jumping manner, and outputting the fused features to the decoder to obtain the sensitivity of denoising image enhancement to vertical stripe noise. And a noise area is suppressed in a targeted manner.

The invention relates to the technical field of intelligent manufacturing and production management, in particular to a self-adaptive production scheduling system based on artificial intelligence, which comprises a data acquisition and reference construction module for analyzing process data to construct a directed acyclic graph representing a non-interference state as a reference map; the theoretical disturbance simulation module is used for converting the interference rule into a graph change instruction, generating a theoretical damaged state graph and obtaining a theoretical difference feature vector; the theoretical difference feature vector comprises, but is not limited to, a vector form obtained after a difference matrix is expanded according to rows or columns in terms of mathematical representation; the real deviation extraction module is used for collecting real-time state data to construct a real-time operation state diagram and calculating a real difference feature vector; a double-domain coupling decision module; an adaptive scheduling execution module; according to the method, the causal relationship is verified by comparing the form of theoretical deduction and actual observation, non-systematic noise is effectively filtered, and accurate response to real faults is realized while the stability of the production rhythm is maintained.

The invention relates to the technical field of grinding machining, and discloses a blade grinding wheel abrasion loss online compensation method which comprises the steps that a dynamic hexagonal detection grid is constructed based on the rotating phase of a grinding wheel, non-repeated high-density sampling is achieved through a Fibonacci spiral expansion path, a detection area is dynamically adjusted to reduce the overlapping rate, and the blade grinding wheel abrasion loss is obtained. Micron-sized wear transition is accurately captured; dispersing the abrasion loss into minimum compensation units, constructing a dynamic state equation in combination with grinding force, dressing displacement and workpiece errors, introducing an attenuation coefficient related to the service life of a dressing wheel, and automatically correcting a model drift error; a grading triggering strategy of residual uncompensated quantity is adopted, dynamic decision making of a processing period is combined, and through integral multiple compensation and a margin temporary storage mechanism, it is ensured that long-term accumulative errors are restrained at the submicron level while overmodulation oscillation is avoided. Through the synergistic effect of space-time coupling detection, discrete-continuous double-domain modeling and intelligent compensation decision, the grinding wheel abrasion compensation precision and the system stability are remarkably improved.

The invention discloses an industrial part defect sample accurate generation method based on a conditional diffusion model, and belongs to the field of image processing and artificial intelligence. The method forms a closed-loop cooperative system by constructing four deep coupling modules of physical constraint noise scheduling, multi-scale feature coupling, double-domain feedback optimization and adaptive weight adjustment; a defect physical forming mechanism is converted into a dynamic noise scheduling strategy, deep interaction between condition information and a feature map is established at multiple levels of a diffusion network, quality closed-loop optimization is achieved through dual evaluation of a pixel domain and a frequency domain, and training weight is dynamically adjusted according to defect scarcity. And multi-scale accurate control is realized, a quality guarantee closed loop is established, the problem of data imbalance is effectively solved, and the performance of an industrial defect detection model is remarkably improved.

The invention provides a human body skeleton point positioning identification method and system based on OpenPose, and the method comprises the steps: carrying out the preprocessing of an input human body image, and obtaining the preprocessing image data; performing skeleton point detection on the preprocessed image data by adopting a double-domain multi-path self-supervised diffusion model in combination with a convolutional neural network to obtain two-dimensional human skeleton point coordinate data; performing signal-to-noise ratio evaluation and noise reduction processing on the two-dimensional skeleton point coordinate data to obtain two-dimensional skeleton point data after noise reduction; performing two-dimensional to three-dimensional conversion through a triangulation principle and a feature matching technology to obtain preliminary three-dimensional skeleton point coordinate data; performing skeleton point optimization through rate perception analysis and a three-dimensional Gaussian compression algorithm; and carrying out shielding prediction and completion processing on the optimized three-dimensional skeleton points to obtain a complete human skeleton point positioning identification result. According to the invention, the problems of accuracy and stability of skeleton point positioning in a complex dynamic scene are solved.

The invention discloses a complex scene traffic sign detection method and system based on dynamic frequency band focusing and double-domain attention screening. The method comprises the following steps: carrying out data preprocessing and data enhancement on a collected road traffic sign image; a CADPCM module and a CHAttention cross coordination attention mechanism are used to construct a CACHNet backbone network; designing a DWMSN neck network, and establishing a dynamic fusion mechanism of multi-scale features; a CACHNet and a DWMSN neck network are used to construct a CDWN model, and a traffic sign enhancement data set is used to train the CDWN model to determine the optimal model weight thereof. Compared with the prior art, the method has the advantages that the average detection precision is improved by 3.3% while the light weight of the model is maintained by constructing a three-level framework of the feature extraction unit, the attention feature expression enhancement and the dynamic feature fusion, the complex scenes such as illumination variation and shielding can be effectively dealt with, and the method is suitable for popularization and application. And high-precision traffic sign detection support is provided for a vehicle-mounted intelligent auxiliary driving system.

The invention relates to the technical field of computer vision, and provides a DETR-based double-domain pseudo-tag generation cross-domain target detection method, a multi-scale decoding query clustering module performs multi-scale clustering on queries output by a decoder, and can accurately capture feature information of targets of different scales, so that the confidence coefficient of pseudo tags is effectively evaluated. The method has a strong cross-domain migration capability, can generate a pseudo label with higher quality between the style data of the target domain and the style data of the source domain, and optimizes the confidence of the pseudo label, thereby improving the accuracy and robustness of target detection, and remarkably reducing the problems of missing detection and false detection of pseudo labels with different scales. The double-domain cooperation and double-domain verification module enhances the reliability of pseudo label generation by combining pseudo labels under different confidence levels of style images of a source domain and a target domain, and effectively filters noise. The adaptability of cross-domain data is improved, it is ensured that the model can smoothly migrate among different fields, and the detection precision and generalization ability are further improved.

The invention discloses a power transmission line state simulation and prediction method based on digital twinning, and the method comprises the following steps: accessing meteorological data, electrical data, space and asset data and historical state data, carrying out the preprocessing, and generating a multi-source spatial-temporal feature sequence set; constructing a double-domain cross-coupling time convolution twinborn model, and obtaining corresponding feature representation through a quick response path and a slow response path; physical consistency state representation is generated through a physical constraint cross gating layer; outputting a prediction result set through a simulation engine; online observation data are obtained, and the model is corrected based on virtual and real residual errors; and generating a current prediction result set by using the corrected model, and converting the current prediction result set into a scheduling and operation and maintenance strategy set. According to the method, the double-domain cross-coupling time convolution twin model and a virtual-real closed-loop correction mechanism are adopted, multi-scale state simulation prediction of the power transmission line is achieved, and the method has the advantages of being high in precision, robustness and performability.

The invention discloses a multispectral image fusion model based on a double-branch self-attention-generative adversarial network and a fusion method thereof, and the model sequentially comprises an input preprocessing module which is used for carrying out the same-amplitude mapping, normalization and overlapping block embedding of visible light and infrared original images, and generating a to-be-fused feature block; the double-branch encoder module captures a cross-modal long-distance dependence and overall brightness structure through multi-head deep convolution transpose attention (MDTA) and a gated deep convolution feedforward network (GDFN), and extracts high-frequency texture and edge information by using a reversible residual gating layer and detail DetailNode iteration; the fusion decoder module is used for carrying out multi-level self-attention-convolution reconstruction on the two paths of features after channel dimension splicing, and outputting a single-frame high-resolution fusion image; and the double-domain discriminator module comprises a visible light domain discriminator and an infrared domain discriminator which are respectively used for carrying out adversarial evaluation on the fused image and the corresponding modal truth value image so as to improve the detail authenticity and the thermal target contrast ratio of the fusion result. The technical problems that an existing infrared-visible light image fusion method is insufficient in detail reservation, unbalanced in brightness and contrast, poor in unsupervised training stability and the like are solved, the method can be deployed on embedded platforms needing real-time and multi-modal information enhancement such as night monitoring, unmanned driving and edge security and protection, and high-contrast and high-information-amount fusion imaging is achieved.

The invention discloses a pole-mounted transformer state evaluation and fault early warning method fusing multi-source data, and relates to the technical field of intelligent power grid monitoring, and the method comprises the steps: executing dual state judgment according to a multi-source fusion feature set, extracting a statistical deviation degree of operation fluctuation abnormity, and calculating a physical deviation degree of multi-physical field coupling imbalance; the statistical deviation degree and the physical deviation degree are fused in a double-domain cooperation mode, and a real-time health index is obtained; the degradation rate and multi-scale energy characteristics of the real-time health index are extracted, a potential degradation aggregation mode is identified by referring to a historical slow-varying baseline, and a hidden danger trend factor is generated; performing composite attribution matching on the hidden danger trend factor and the multi-source fusion feature set, executing grading risk criteria according to a multi-physical field anomaly mechanism chain, and outputting an early warning grade and an early warning reason; according to the method, diagnosis is carried out by embedding a multi-physical field anomaly mechanism chain, and the physical root and evolution stage of the fault can be accurately positioned.

The invention discloses a hyperspectral image and LiDAR data collaborative classification method based on double-domain mask and multi-scale local reconstruction, and belongs to the field of remote sensing image classification. According to the method, the problems of scarcity of annotation data and insufficient multi-source feature fusion precision in cross-modal classification of a traditional method are solved. According to the invention, feature learning is carried out through mask random image blocks and channels; a hierarchical multi-scale reconstruction architecture is designed, a lower-layer encoder learns fine-grained features, an upper-layer encoder recovers macroscopic semantic information, and multi-level feature space alignment is realized in combination with deconvolution up-sampling and adaptive pooling. According to the method, a multi-modal feature interaction mechanism and a cross-modal attention module are constructed by fusing the local feature extraction advantages of a convolutional neural network (CNN) and the global modeling capability of Transform, and the complementarity of heterogeneous data is enhanced. According to the method, through multi-level feature dynamic fusion and adaptive weight distribution, the collaborative classification precision of the hyperspectral image and the LiDAR data is improved. The method can be applied to remote sensing image classification.

The invention relates to the technical field of remote sensing landform enhancement, and discloses a remote sensing landform enhancement algorithm, which comprises the following steps of: S101, preprocessing an existing high-resolution DEM (Digital Elevation Model) image and an existing low-resolution DEM image to construct a training data set; step S102, constructing a double-domain multi-scale attention fusion super-resolution network; step S103, training the double-domain multi-scale attention fusion super-resolution network through the training data set; according to the method, the edge information and spatial continuity characteristics of the DEM image are comprehensively considered, the high-frequency and low-frequency information of the DEM image is extracted through Haar wavelet transform by utilizing the characteristic that a wavelet domain is highly sensitive to the high-frequency and low-frequency information, and spatial domain feature information obtained by a parallel branch convolutional layer is combined, so that the high-frequency and low-frequency information of the DEM image is obtained. Through fusion of the two, the network has a feature enhancement function in a shallow layer, and meanwhile, the importance relationship between different depth feature maps is obtained through the multi-scale attention fusion module group, so that the precision of DEM super-resolution reconstruction is improved.

The invention belongs to the technical field of path planning, and provides a dynamic path planning method, system and equipment based on multiple mobile robots and a medium based on the multiple mobile robots in order to solve the problems that the calculated amount is increased sharply and unexpected situations cannot be handled in the current path planning of the multiple mobile robots, and the independent path planning of each robot is completed by adopting mixed A *-epsillon. A dual-domain priority queue is introduced, child nodes are generated through node expansion, then a path is re-planned for the robot with constraints, and therefore a conflict-free path of the robot is generated; and performing control sequence sampling on each robot, taking a conflict-free space-time path as a reference path, obtaining an optimal control sequence according to the cost value of each predicted trajectory, and adjusting the path in real time through rolling optimization so as to cope with a dynamic obstacle. The method can actively avoid a dynamic obstacle or an obstacle which is not considered during planning.

The invention discloses an artificial intelligence abnormity early warning diagnosis method for industrial equipment operation state monitoring, and belongs to the technical field of industrial equipment intelligent monitoring, and the method comprises the steps: collecting a multi-source heterogeneous signal of industrial equipment, and carrying out the time-frequency dual-domain feature extraction; constructing a multi-scale time window based on the time-frequency features, and executing self-supervised contrast learning by injecting a preset abnormal mode to obtain cross-scale contrast feature representation; constructing a dynamic adjacency matrix according to the comparison features, extracting space-time correlation features through a graph attention network, and determining an abnormal score based on joint evaluation; according to the method, a deep coupling closed-loop cooperative system is formed, multi-dimensional state characterization, adaptive anomaly detection and root cause diagnosis are realized, and the problems of single data source and lack of fault analysis capability in the prior art are effectively solved.

The invention provides an image super-resolution reconstruction method based on double-domain feature fusion and implicit representation, and relates to the field of image processing and computers, and the method comprises the steps: obtaining a low-resolution remote sensing image, and carrying out the preprocessing of the low-resolution remote sensing image; performing feature extraction on the preprocessed remote sensing image through Haar discrete wavelet transform and a Transform-based pyramid structure to obtain frequency domain features and spatial domain features; performing double-domain cross attention fusion on the frequency domain features and the spatial domain features to obtain local detail features and global structure features; and through an implicit representation network, the fused local detail features and global structure features are mapped to any space coordinates, a final three-channel high-resolution image is obtained, and high-quality reconstruction of a remote sensing image of any scale is realized. According to the technical scheme of the invention, the implicit neural representation is guided to realize higher-precision image reconstruction through the cooperative expression of the frequency domain information and the spatial domain information.

The invention discloses a compressed sensing MRI (Magnetic Resonance Imaging) reconstruction method and system based on a double-domain fusion expansion model. The method comprises the following steps: acquiring an MRI image, preprocessing the MRI image to obtain a preprocessed image and a corresponding compressed sensing measurement value, and constructing a data set; training, testing and verifying the double-domain fusion expansion model by using the data set; and outputting a reconstructed image from the MRI image after mask sampling by using a double-domain fusion expansion model. According to the double-domain fusion expansion model, a complex convolutional neural network and a K-space attention mechanism are introduced for complex value data processing and double-domain information fusion in compressed sensing MRI reconstruction. The complex value data is processed through the complex network, the amplitude and phase information of the complex field is fully utilized, the detail recovery capability and the noise suppression effect of the model are enhanced, and particularly, excellent robustness is shown at a low sampling rate. The method is excellent in performance in complex MRI data reconstruction, and has high reconstruction precision and generalization ability.

The invention discloses an intelligent energy consumption distribution method for a multi-source energy system. The method comprises the steps of energy supply data fusion, dual-channel joint prediction, energy response lag correction, multi-objective optimization and intelligent energy consumption distribution. The invention relates to the technical field of data processing of power management and resource scheduling, and the method comprises the steps: carrying out the unified collection and archiving of operation control data and environment information of power grid energy, distributed photovoltaic energy and an energy storage system, and constructing an energy supply fusion data set; respectively extracting energy availability and user load characteristics by adopting a double-domain attention decoupling fusion joint supply and demand prediction method, and realizing double-channel joint modeling; modeling, estimating and correcting response lags of different energy sources in combination with a dynamic response curve reconstruction method; and further constructing a layered target penalty function containing real-time response error, economic cost, carbon emission and response penalty terms, and embedding the layered target penalty function into a particle swarm optimization algorithm to obtain an energy distribution strategy under multi-target optimization.

The invention discloses a dynamic scene reconstruction method and device based on a multi-scale Gaussian sphere, and relates to the field of computer vision, and the method comprises the steps: employing a motion recovery structure algorithm to carry out the processing of a to-be-reconstructed video frame sequence, generating a sparse point cloud, carrying out the initialization of the sparse point cloud, and generating a 3D Gaussian sphere set; processing the 3D Gaussian ball set by adopting a double-domain deformation model and an adaptive timestamp to obtain a deformed 3D Gaussian ball set; performing multi-scale Gaussian processing on the deformed 3D Gaussian ball set to generate a multi-scale Gaussian ball set; gaussian screening based on the pixel coverage rate is carried out on the multi-scale Gaussian ball set, and an optimized multi-scale Gaussian ball set is obtained; and performing Alpha hybrid processing based on the optimized multi-scale Gaussian ball set, and reconstructing to obtain an anti-aliasing dynamic rendering scene image. According to the method, the problems of high calculation overhead, aliasing effect and the like of the current dynamic scene reconstruction are solved.

The invention relates to the technical field of intelligent wearable devices, and discloses a multi-modal positioning and health monitoring combined method and device, and the method comprises the steps: obtaining first multi-modal sensor data of an intelligent wearable device, and carrying out the two-domain signal representation separation, and obtaining a physiological signal representation vector and a motion artifact representation vector; performing comparison loss optimization of motion posture modulation to obtain a first physiological signal representation and a first motion artifact representation; performing feature enhancement to obtain a second physiological signal representation and a second motion artifact representation; performing position-sensitive contrast characterization joint optimization and layered contrast characteristic distillation to obtain a lightweight model, processing data of the second multi-mode sensor through the lightweight model, and outputting a pure physiological signal and a target positioning result. The problem that the performance of a traditional attention mechanism is reduced at a motion conversion point is solved; and the adaptive capacity to a complex motion scene is obviously enhanced.

The invention discloses a rainfall field image extraction method based on double-domain collaboration and progressive feature decoupling. The method comprises the following steps: acquiring a rain image; respectively inputting the rain image into the rain layer branch and the background branch, carrying out corresponding image block embedding operation, and generating an initial rain layer feature and an initial background feature; performing feature extraction on the initial rain layer features through a learnable wavelet transform algorithm to obtain rain layer features, and performing feature extraction on the initial background features through frequency domain separation convolution to obtain background features; performing multi-scale progressive coupling feature extraction on the rain stripe features and the background features to obtain processed rain layer features and processed background features; and performing multi-scale up-sampling on the processed rain layer features and the processed background features to reconstruct a rain layer image and a rain-free background image, and performing weighted fusion on the rain layer image and the rain-free background image through learnable residual gating to obtain a reconstructed rain image and a rainfall field image. According to the invention, accurate separation of the rainfall field can be realized.

The invention discloses a real-time back break control method of a heading machine based on visual association and digital twinning, which relates to the technical field of intelligent control of coal mining equipment, and comprises the following four steps: processing binocular visual data through a double-domain double-branch Transformer model, extracting coal rock and roadway structure characteristics and outputting cutting head offset; a GCN-LSTM model is adopted to fuse visual, laser radar and fiber-optic gyroscope data, and dynamic calibration and output of a three-dimensional pose are carried out; a roadway three-dimensional model is constructed based on the SLAM technology, and real-time updating of a digital twin model is achieved through edge-cloud collaboration; and in combination with a reinforcement learning strategy, generating a cutting parameter adjustment instruction according to the deviation of the digital twin model, and correcting model parameters through error feedback. The method solves the problems of low visual recognition precision, sensor data drift, digital twinning update lag and the like in an underground complex environment, improves the back break control precision, tunneling efficiency and equipment stability, and is suitable for tunneling operation under complex geological conditions.

The invention discloses a double-domain artifact correction method based on multi-energy-level data and physical prior fusion, and the method achieves the efficient and precise removal of dispersive metal artifacts through the construction of a complete technical scheme of the combination of multi-energy-level data collection, double-domain cooperative correction and physical model constraint. The system has the beneficial effects that the system covers multiple fields of medical treatment, industry, security and protection, aerospace and the like, and has extremely high universality and adaptability. On the basis of a multi-energy-level slow switching scanning protocol, full-angle scanning of at least two energy levels is completed by dynamically adjusting radiation source parameters, and obtained complete multi-energy-level projection data is converted into a high-dimensional tensor through a channel dimension splicing image fusion method. The constructed multi-channel virtual image completely retains attenuation characteristics and structure information of a target object under each energy, provides a more comprehensive input source with discrimination for a deep learning network, builds a data basis for accurate correction in different fields fundamentally, and adapts to various imaging scenes containing metal targets.

The invention provides an unmanned aerial vehicle full-time perception image reconstruction method based on multi-mode cooperative reinforcement learning and degeneration decoupling. A frequency perception feature modulation model, a dual-mode dual-domain transformation module and a dynamic bidirectional guide mechanism are included. According to the system, firstly, feature information of different frequency bands is adaptively separated and modulated through a frequency sensing feature modulation model, and decoupling and compensation of composite unknown degradation are achieved; realizing cross-domain interaction and information fusion of visible light and infrared characteristics in a spatial domain and a channel domain by using a bimodal dual-domain transformation module; and finally, realizing collaborative enhancement of cross-modal degradation perception through a bidirectional dynamic guide mechanism, and generating an unmanned aerial vehicle visible light reconstruction image and an infrared super-resolution image with higher structural consistency and texture fidelity. According to the method, deep fusion and degeneration decoupling of multi-modal information can be realized in a complex degeneration environment, and the imaging quality and the environmental adaptability of an unmanned aerial vehicle full-time sensing system are remarkably improved.

The invention relates to a low-illumination underwater image enhancement method based on double-domain adaptive collaboration, and the method employs a multi-scale mixed attention mechanism to collaboratively integrate the spatial domain and frequency domain information of an image, and obtains the key feature information of the image. And on the basis of the obtained image key feature information, adaptive parametric convolution is used, convolution kernel parameters are dynamically adjusted by using a multi-branch convolution structure, and image dynamic key information is obtained. And a dynamic content perception Markov discriminator is used to cooperatively perceive space and frequency characteristics, and the authenticity of the enhanced image is discriminated, so that the fidelity of the texture is improved. And finally, optimizing the generator and the discriminator by using adversarial training, and finally outputting an enhanced underwater image by using the optimized generator. The method can be embedded into an underwater robot platform, image quality enhancement is achieved with fewer computing resources in a low-illumination complex water environment, image definition, contrast and color consistency are improved, the enhanced image is closer to human eye visual perception, and the visual perception capability and operation reliability of an underwater robot are remarkably improved.

The invention discloses a prospective preprocessing method and system in six-axis mechanical motion automatic control, particularly relates to the field of six-axis mechanical motion automatic control, and is used for solving the problem of insufficient track prediction and control precision in a load sudden change scene. The inertia tensor is updated in real time in combination with dual-channel Kalman filtering, dynamic inertia data are accessed, and the sensitivity of trajectory prediction to load change is improved; double-domain feature extraction of time-domain thermal stress density and frequency-domain energy dispersion is adopted, acceleration pulse layout is dynamically optimized by using a complementary integral mechanism, and the overload risk of a driver is reduced; meanwhile, the instruction dispatcher avoids cache overflow through advanced rhythm regulation and control, and continuity of instruction streams is guaranteed; the inertia model is continuously refreshed through closed-loop feedback, and synchronous evolution of inertia change and trajectory prediction is achieved; the overload shutdown risk caused by inertia lag of six-axis mechanical motion is reduced, and the adaptive capacity of the prospective preprocessing method in a dynamic complex operation environment is enhanced.

The invention relates to the technical field of remote sensing target detection, in particular to a remote sensing image directed target detection method based on double-domain feature fusion, and the method comprises the steps: obtaining a remote sensing image data set, and obtaining a training set; constructing a remote sensing target detection network, wherein the remote sensing target detection network comprises a feature extraction branch, a double-domain feature fusion branch for performing feature fusion on spatial domain features and frequency domain features of the remote sensing images, and a target detection branch for judging corresponding categories and positions of the remote sensing images according to different fusion features; training a remote sensing target detection network by using the training set to obtain a remote sensing target detection model; and inputting a to-be-detected remote sensing image into the remote sensing target detection model, and outputting a corresponding detection category. According to the method, the effect of the spatial domain and the frequency domain on remote sensing classification is fully considered, and through adaptive selection of the spatial domain and the frequency domain and feature interaction fusion between the two domains, fusion of global context information and local information is effectively enhanced, and the deficiency of target information is made up.

The invention discloses a multi-scale ReID network and double-domain joint measurement target tracking method for a complex scene. The method comprises the following steps: S1, constructing and training a target detection model YOLOv5; s2, designing an improved ReID network IncepSPA-DSC fusing a depth separable convolution and a spatial pyramid channel attention mechanism; s3, a DeepTrack-SPAE tracking framework is constructed, and a DeepTrack-SPAE tracking framework is According to the invention, by constructing a multi-scale feature fusion mechanism and an attention enhancement module, on the premise of maintaining the real-time processing speed, an anti-interference feature vector with strong discrimination is generated, and the cooperative capture capability of the network on local features and global context information is significantly improved. A space-feature double-domain joint measurement method is innovatively proposed, and by establishing a feature similarity matrix fused with Euclidean distance constraint, the spatial proximity and feature consistency of a target are considered in cost calculation, so that the problem of trajectory breakage caused by short-time shielding is effectively solved.

The invention provides an intelligent traceless erasing method and system for video picture characters, and relates to the technical field of video stream processing, and the method comprises the steps: carrying out the space-time dual-domain feature extraction of a preprocessed first video frame sequence, and obtaining a first space-time feature; and establishing a space-time double-domain attention model according to the first space-time feature, the space-time double-domain attention model calculating fusion space domain and time domain features through cross attention, dynamically distributing fusion weights based on the character movement speed, obtaining a second video frame sequence of the to-be-processed video stream, preprocessing the second video frame sequence, and extracting a second space-time feature. Texture and motion characteristics of front and back frames are synchronously referenced through a space-time double-domain attention model, so that the time sequence coherence of a repaired area is improved, a time sequence consistency compensation mechanism driven by an SSIM is combined, the flicker frequency is reduced, and the time-space domain weight is adaptively adjusted according to the character speed through a dynamic weight distribution strategy, so that the time sequence coherence of the repaired area is improved. And stable erasing and rapid moving of characters are realized.

The invention belongs to the technical field of image processing and deep learning, and particularly relates to an image defogging method based on dynamic wavelet prior and double-domain learning. Aiming at the requirements of all-weather clear imaging in the fields of intelligent traffic systems, safety monitoring and the like, and in order to overcome the defect that a static convolution kernel adopted by a traditional defogging method is difficult to adapt to different haze degradation, the invention provides a method for dynamically generating a convolution kernel by using haze priori contained in a multi-scale wavelet LL sub-band; and an efficient, robust and accurate image defogging model is constructed. According to the invention, based on a multi-scale U-shaped coding-decoding architecture, a dynamic wavelet depth separable convolution module DyWConv is embedded in front of each level of a coder to realize content adaptive feature extraction, and a double-domain feature learning module SPAFormer Block cooperatively utilizing Fourier domain global modulation and wavelet domain multi-scale decomposition is designed. And double-domain features are fully fused through an adaptive gating fusion mechanism, and finally a clear image is reconstructed and output step by step. According to the method, a method for explicitly encoding frequency domain degradation prior into dynamic convolution kernel parameters is innovatively provided, the complementary advantages of Fourier transform and wavelet transform are cooperatively utilized, spatial non-uniform haze can be effectively removed, image details can be recovered, leading performance is achieved in a synthetic data set and a real scene, and the method has a wide application prospect.