529 results about "Wave transformation" patented technology

The invention discloses an electric energy metering box fault prediction method and system based on big data analysis, relates to the technical field of smart power grids, and solves the problems of progressive aging missing detection and instantaneous interference misjudgment caused by dependence on single parameter threshold alarm and fault positioning misalignment caused by multi-source data isolated analysis in the prior art. According to the scheme, electrical, environment and equipment state parameters are collected in real time through a multi-dimensional sensing network; the error drift of the mutual inductor is dynamically predicted based on an LSTM-Kalman filtering model, and core breakdown early warning is realized in combination with wavelet transform; outputting a corrected resistance value and a fault mark by using a BP neural network; predicting the life of the piezoresistor by adopting a gradient boosting decision tree and fusing lightning overvoltage characteristics; the transient interference is suppressed through the combination of a Transform self-attention mechanism and dynamic time warping; according to the method, the aging detection precision and the complex environment adaptability are remarkably improved, the misjudgment rate is reduced, and the multi-fault associated positioning and active defense capability is realized.

The invention provides a marine ranch water quality parameter real-time correction and compensation method and system for a multi-source sensor, and relates to the technical field of multi-source sensors, and the method comprises the steps: constructing a double-layer edge computing network, and connecting a sensor through a micro-service architecture to collect water quality data. And carrying out data preprocessing in combination with wavelet transform. And establishing a sensor digital twinborn model, and calculating the real-time credibility. Establishing a multi-dimensional sensor association network, optimizing a weight coefficient by adopting federal learning, and establishing a self-evolution correction parameter matrix; and fusing the sensor data by using a multi-task deep learning model to generate an initial correction value. And calculating a theoretical reference value through a space-time sequence prediction model. A compensation coefficient is adaptively adjusted by adopting a fuzzy decision tree, hierarchical water quality parameter correction is realized, and a closed-loop self-optimization intelligent correction system is formed through verification of a digital twin model. The accuracy and reliability of marine ranch water quality monitoring data are effectively improved.

The invention provides a water level distributed monitoring method and system based on a visual sensing network, and relates to the technical field of water level monitoring, and the method comprises the steps: constructing a water level propagation prediction model through multi-modal comparative learning and a graph neural network, processing a real-time water level image through the combination of adaptive wavelet transform, decoupling representation and a dense connection network, and obtaining a real-time water level prediction model; the water level measurement value and the reliability score are obtained, anomaly analysis and emergency monitoring are carried out based on state estimation, causal discovery and hierarchical reinforcement learning, high-precision and high-reliability distributed water level monitoring can be achieved, abnormal water level changes can be effectively recognized, and timely emergency monitoring parameters are provided.

The invention provides a brushless motor attitude detection and balance control method and system based on a gyroscope, and relates to the technical field of control, and the method comprises the steps: collecting angular velocity and acceleration data through a six-axis gyroscope, carrying out the noise reduction through wavelet transform, and carrying out the attitude calculation through the combination of an extended Kalman filter and a quaternion algorithm. A rotor position signal is obtained through a magnetic encoder, nonlinear compensation is carried out, and rotating speed data are calculated. A motor state is modeled by adopting a long-short-term memory network, a double-layer adaptive fuzzy neural network controller is constructed, and attitude error compensation and rotation speed fluctuation suppression are realized. A controller model is optimized through particle swarm optimization and a genetic algorithm, a compensation current vector is corrected in real time, and the control precision and stability of the brushless motor are improved. According to the method, the operation efficiency and the dynamic response capability of the brushless motor are effectively improved.

The invention discloses a fault detection and root cause analysis system for a multi-source energy system, and relates to the technical field of energy management and fault diagnosis. The system obtains various operation data and image data of the multi-source energy system in real time; preprocessing the operation data and the image data; inputting the preprocessed data into a hybrid algorithm combining wavelet transform, CNN and LSTM, and detecting an abnormal mode and a potential fault in the multi-source energy system; the digital twinborn model is dynamically updated according to data acquired in real time; and in combination with the feature extraction capability of a digital twin model and a hybrid algorithm, positioning key causes causing the fault, simulating a system operation state, analyzing potential causes of the fault, and outputting an analysis result. According to the invention, the comprehensive analysis capability of the system on various data sources is effectively improved, and a comprehensive and accurate solution is provided for intelligent operation and maintenance of the energy system.

The invention provides a civil air defense engineering concealed conduit leakage positioning method based on acoustic characteristics, and belongs to the technical field of civil air defense engineering. An acoustic signal acquisition network is constructed by deploying a high-sensitivity hydrophone array, and signals are preprocessed by using a time division multiple access technology and a pulse compression technology; and carrying out time-frequency analysis by applying wavelet transform to extract acoustic features. A time reversal mirror technology is introduced to identify direct propagation and multipath reflection signals, a fluid acoustic coupling propagation equation is constructed to analyze a leakage sound source mechanism, and pipe network topological information and a sound wave speed correction function are combined to compensate a measurement error. Wherein the deep learning pipe network acoustic propagation multi-modal model is fused with a pipeline structure encoder, an acoustic feature extractor and a position prediction decoder, and high-precision leakage positioning and degree evaluation in a complex environment are realized through a three-stage pre-training strategy; the technical problem that it is difficult to accurately locate the position of a leakage point in the complex pipe network environment of civil air defense engineering is solved.

The invention relates to the technical field of remote sensing monitoring, in particular to a remote sensing target detection method and system for a low-visibility image. The method comprises the following steps: acquiring multi-modal remote sensing image data; carrying out defogging enhancement processing on the low-visibility input image; normalizing the defogged RGB image and the defogged IR image, and then splicing and fusing the RGB image and the IR image; carrying out layer-by-layer coding on the multi-modal fusion image by adopting a mixed trunk structure fusing Transform, Mamba and CNN (Convolutional Neural Network); performing frequency domain decomposition on the trunk output features based on two-dimensional wavelet transform; generating an HR feature map by adaptively selecting a key region; and carrying out cross-scale aggregation on the HR feature map to obtain a detection target frame. Through the multi-modal image defogging enhancement and feature distillation mechanism, the definition and contrast of the remote sensing image in severe weather such as haze and rainy days are effectively enhanced, the shielding interference of environmental degradation on small target detection is weakened, and the stability and adaptability of the model in complex weather scenes are enhanced.

The invention discloses a night semantic segmentation method and device based on wavelet transform detail enhancement and text prompt, and the method comprises the steps: obtaining a night image, carrying out the preprocessing of the night image, and carrying out the reconstruction of a wavelet image; and inputting the preprocessed night image and the image after wavelet transform reconstruction into a deep learning model for semantic segmentation to obtain a segmentation result of the night scene object. A new three-stage network structure is designed and formed, in the first stage, a three-mode feature extractor composed of an image encoder, a night semantic category encoder and a wavelet image encoder is used for extracting features, in the second stage, a double-branch cross-mode feature interaction module is designed, and the feature extraction is carried out through the image encoder. In the first stage, features of different spatial resolutions and semantic hierarchies and natural language priori of a target object are integrated, all-directional semantic information from coarse granularity to fine granularity is captured, in the third stage, a multi-scale feature segmentation decoder is introduced, details of a low-light area are enhanced, fine texture edges and target contours are captured, and the target object is obtained. Through positioning and understanding of the target area by the natural language prior enhancement model, the precision of night scene semantic segmentation can be effectively improved.

The invention relates to the technical field of deformation early warning of a coal mine TBM tunneling roadway, in particular to a simulation and monitoring method for intelligent early warning of large deformation of the coal mine TBM tunneling roadway. According to the technical scheme, the method comprises the steps of data acquisition and preprocessing, feature mining and fusion, intelligent early warning model construction and real-time monitoring and early warning, and the data acquisition and preprocessing specifically comprises the steps that multiple types of sensors are installed on a coal mine TBM tunneling roadway site to acquire multiple data, three-dimensional reconstruction is conducted on image data, and timestamps and space coordinates of the data are unified; a multi-dimensional spatio-temporal data set is formed by using wavelet transform denoising, and feature mining and fusion comprises the steps of extracting geologic structure image data features by using 3D-CNN and capturing spatio-temporal change features of other data by using LSTM. According to the method, high-quality data is provided, meanwhile, the model is more suitable for complex working conditions and more accurate, the generalization ability of the model is enhanced, real-time monitoring and early warning are timely, effective and visual, decision making is facilitated, and safety accidents and economic losses can be effectively reduced.

The invention relates to the technical field of power monitoring, in particular to a station power utilization system safety and energy efficiency monitoring system based on multi-source information fusion, which comprises a signal analysis module, an anomaly identification module, a risk assessment module and an emergency response module. According to the method, the fineness and the response speed of power station monitoring are improved by integrating wavelet transformation and information entropy analysis, the time-frequency characteristics of signals are deeply analyzed through wavelet transformation, the capturing capacity for short-term and long-term changes is improved, the integrity and the accuracy of data analysis are further ensured, and the accuracy of power station monitoring is improved. The information entropy is introduced to assess the stability and uncertainty of the power station, a scientific basis is provided for predicting and recognizing potential risks of the system, and by combining real-time anomaly recognition and dynamic risk assessment, equipment faults and system failures can be effectively predicted, and energy consumption is optimized, so that early warning and prevention and control are performed in advance, the downtime is remarkably shortened, and the system reliability is improved. The maintenance cost is reduced, and the level and effect of energy efficiency management are improved.

The invention discloses a human body posture reconstruction and tracking method based on a neural network, and relates to the technical field of posture reconstruction and tracking. Target echo data are acquired by adopting a multiple-input multiple-output (MIMO) antenna array, and a low-resolution three-dimensional radar image is constructed by adopting space-time weighting in combination with wavelet transform noise reduction and self-adaptive gain compensation; in the feature extraction stage, a Transform-based feature extraction network is introduced, short-distance joint details are processed in combination with a local convolutional neural network (CNN), the spatial correlation of a far-end joint is calculated by adopting a global self-attention mechanism, and a global feature enhancement (GFE) module performs feature completion on a low-confidence region; a hybrid memory network HMN is adopted to carry out short-time and long-time information fusion, a short-time memory unit STM is used to store current frame features, a long-time memory unit LTM is used to store historical key point data, a drift error is corrected in combination with time sequence alignment, and time sequence consistency of attitude estimation is guaranteed.

The invention relates to the technical field of wireless communication, and particularly discloses a self-adaptive anti-interference communication method and system based on star flash-Bluetooth dual-mode, and the method comprises the steps: monitoring channel interference parameters in real time through a dual-mode radio frequency module on a communication terminal, including the out-of-band radiation power and the pulse interference duty ratio at the switching moment; instantaneous interference intensity features are extracted by using synchronous compression wavelet transform and sparse time-frequency analysis, periodic pulse interference features are identified by combining Hawkes process modeling, and a comprehensive anti-interference feature vector is constructed by fusion; the feature vectors are input into a pre-trained gradient boosting tree model for communication quality evaluation, an optimal communication mode is dynamically selected according to a scoring result, intelligent response to a complex interference environment is achieved, the stability and reliability of a communication system in a high-interference scene are improved, energy efficiency optimization and resource scheduling are considered, and the method is suitable for large-scale popularization and application. The method is suitable for industrial, medical, Internet of Things and other application scenes with strict requirements on communication quality.

The invention discloses a cross-modal eye fundus image generation method and system based on a generative adversarial network, relates to the technical field of medical image processing, and constructs an eye fundus focus perception and edge consistency generative adversarial network by taking a cyclic consistency generative adversarial network as a baseline. The core of the method is that a lesion perception mixed attention module is embedded in a bottleneck layer of a generator so as to strengthen the extraction capability of fine features of a lesion area; an edge information extraction module is designed, and key edge features are accurately extracted in combination with Roberts edge detection, wavelet transform and non-local mean denoising; and a joint loss function containing edge consistency loss is constructed, and the semantic consistency of a focus structure during cross-modal generation is ensured by minimizing the feature difference between the source image and the generated image. According to the method, the problems of disordered content, inconsistent structure and unstable training of the generated image in the prior art are effectively solved, and the simulation degree and clinical availability of the generated image are remarkably improved.

The invention relates to the technical field of vehicle safety monitoring, and discloses a non-contact type in-vehicle driver respiratory rate detection method and device. Physiological signal data are collected through a millimeter wave radar, infrared thermal imaging and a microphone array, breathing frequency band signals are extracted through self-adaptive wavelet transform, and breathing signal waveforms are reconstructed through time-frequency joint analysis. A model formed by combining a multi-channel convolutional neural network and a bidirectional gating circulation unit is adopted to extract a respiratory rhythm feature vector, interference is eliminated based on a dynamic time warping algorithm, and a respiratory frequency value is estimated by applying methods such as variational mode decomposition. And processing abnormal points and missing data through a sliding window mechanism, and outputting a stable monitoring result. The invention further provides a sensor data synchronous calibration and model online updating method. According to the invention, the breathing frequency can be detected with high precision in a complex in-vehicle environment, and the safety and intelligent level of the vehicle are improved.

The invention discloses an electric energy quality comprehensive treatment method and device based on multilevel collaborative optimization, and belongs to the technical field of electric energy quality treatment of an electric power system. According to the method, voltage and current data of a power grid are collected in real time and multi-parameter monitoring is performed; processing the data by using an adaptive wavelet transform and singular value decomposition combined algorithm, extracting disturbance characteristics and identifying the type and position of a disturbance source; a multi-objective optimization function including harmonic suppression, voltage stabilization, three-phase balance and power factor improvement is constructed, a Pareto optimal solution set is solved, and an optimal solution is selected; and according to an optimization result, generating and executing a cooperative control strategy of each type of electric energy quality treatment equipment. Cooperative treatment of various electric energy quality problems and cooperative control of various treatment devices are realized, the accuracy, self-adaptability and economical efficiency of electric energy quality treatment are improved, and the reliability and stability of an electric power system are enhanced.

The invention provides a spatio-temporal data fusion-based power distribution network load prediction method and system, and relates to the technical field of power distribution network load prediction, and the method comprises the steps: collecting related data of a power distribution network, carrying out the wavelet transform decomposition of historical load data, obtaining a load feature matrix, constructing a hierarchical graph convolution network based on topological structure data, and extracting topological correlation features; and generating a spatial-temporal feature tensor through tensor decomposition fusion, training a depth probability prediction model adopting a variational auto-encoder structure, and adjusting prediction probability distribution in combination with environmental data. According to the method, the prediction precision is improved, a complex space-time dependency relationship can be captured, and reliable uncertainty quantization is provided.

The invention discloses a water source chlorophyll a concentration prediction model design method based on machine learning. The method comprises the following steps: acquiring chlorophyll a concentration data in a to-be-predicted region for a continuous period of time; carrying out data preprocessing on the chlorophyll a concentration data, and filtering high-frequency noise by adopting wavelet transform preprocessing; constructing a concentration prediction model, carrying out data preprocessing on chlorophyll a concentration data, and filtering high-frequency noise by adopting wavelet transform preprocessing; constructing different concentration prediction models, and inputting the processed chlorophyll a concentration data and physicochemical parameters into the prediction models to obtain a chlorophyll a concentration data prediction result; and comparing prediction results of different prediction models, and determining the prediction model. According to the prediction model design method, the WT-GRU model is adopted to preprocess the data through wavelet transform, the wavelet transform effectively extracts key time scale characteristics through signal decomposition, and the accuracy of chlorophyll a concentration prediction is remarkably improved.

The invention relates to the technical field of image processing, in particular to a remote sensing image segmentation method based on multilayer wavelet transform and a dynamic memory network, which comprises the following steps: decomposing an image into a low-frequency sub-band and a plurality of high-frequency sub-bands through multilayer Haar wavelet transform, and performing feature extraction on each frequency band in combination with convolution operation to obtain a final feature map; constructing a dynamic memory unit network, performing dynamic feature extraction on the feature map by adopting a query-key-value structure, generating a dynamic memory enhanced feature with global context information, performing global-local feature modeling on the feature map, and fusing the dynamic memory enhanced feature and the global-local feature to obtain a fused feature map; and segmenting the fused feature map to obtain a final segmentation result. According to the method, low-frequency smooth information and high-frequency detail information of the image can be accurately extracted, efficient fusion of global and local features can be realized, and the accuracy and robustness of remote sensing image segmentation are improved.

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 relates to the technical field of image processing, in particular to a road anti-corrosion effect detection and evaluation method based on an image processing technology, and the method comprises the steps: obtaining to-be-detected grayscale images corresponding to two different types of images on the surface of a road anti-corrosion layer, and carrying out the segmentation to obtain to-be-detected sub-images; respectively obtaining two difference images corresponding to the to-be-detected sub-image according to the row gray difference and the column gray difference of the gray difference image corresponding to the to-be-detected sub-image; obtaining the roughness degree of the to-be-detected sub-images, obtaining an optimal scale based on the roughness degree, decomposing the two to-be-detected grayscale images by using wavelet transform according to the optimal scale, and fusing according to decomposition results to obtain a fused feature image; and obtaining a defect region of the fused feature image, and obtaining a road quality detection result according to area features of the defect region. According to the invention, a more accurate road quality detection result can be obtained.

The invention discloses a high-resolution image semantic segmentation network for underwater scene design. A high-resolution image semantic segmentation network for an underwater scene is provided based on HRNetV2. The core of the method comprises: a WDCM feature extraction module for enhancing the anti-noise capability by using wavelet transform convolution, inhibiting irrelevant features by dynamic gating, and improving the performance of a complex scene in combination with an RCA module; a multi-scale slice segmentation head is adopted, global and local features are fused, and key information is dynamically selected; and 3) proposing a sliding combination loss function, and optimizing target boundary semantic information on the basis of variable weight focus (Focal) loss and dice (dice) loss. The target boundary segmentation precision is effectively improved, and the segmentation performance in a complex scene is improved.

The invention discloses an intelligent primary and secondary fusion pole-mounted circuit breaker fault monitoring method, relates to the technical field of power equipment monitoring, and is used for solving the problem of insufficient real-time performance of fault monitoring under complex environment interference. According to the invention, electrical, mechanical vibration and environmental data are synchronously acquired through a multi-mode sensor array, and a timestamp synchronization mechanism is applied; performing multi-source interference classification on the original data, separating noise by using wavelet transform, and identifying interference types by clustering; kalman filtering adaptive compensation is applied based on an interference result, and environment feedback is introduced to ensure low delay; extracting multi-dimensional fault features, and forming a robust matrix through time-frequency analysis and principal component dimensionality reduction; inputting a deep learning model to carry out space-time modeling and rapid classification; a response mechanism is triggered to execute isolation or alarm, and closed-loop optimization is formed. The method effectively solves the problem of insufficient real-time performance under the interference of a complex environment, and improves the monitoring precision and the response speed.

The invention provides a distribution network type lightning arrester leakage current optimization measurement method. The method comprises the steps of signal acquisition, preliminary signal denoising and preprocessing, wavelet transform signal decomposition, high-frequency noise component elimination and processing, input convolutional neural network CNN descrambling, integrated learning model optimization, output of descrambled leakage current signals, performance evaluation and result feedback. Through a multi-level processing method combining wavelet transform and a convolutional neural network, high-frequency noise in a weak leakage current signal is effectively removed, effective characteristics of the signal are reserved in the denoising process, and compared with a traditional method depending on a single signal processing technology, signal descrambling can be more accurate and effective, and the signal descrambling efficiency is improved. According to the wavelet transform signal decomposition method, the signal can be decomposed into components with different frequencies, high-frequency noise components are effectively eliminated, low-frequency and intermediate-frequency effective signals are reserved, then the convolutional neural network further improves the definition and reliability of the signal through depth feature extraction, and accurate detection of the leakage current signal is ensured.

The invention relates to an image three-dimensional reconstruction method, in particular to a multi-view three-dimensional reconstruction method based on frequency perception feature enhancement and cost aggregation. The objective of the invention is to overcome the defects of lack of frequency sensing capability and limited processing capability for problems of weak texture, noise, illumination variation, color distortion and the like in an existing learning-based multi-view three-dimensional reconstruction method. According to the method, multi-view three-dimensional reconstruction is realized through the steps of acquiring a multi-view image, calculating a multi-scale frequency sensing feature, calculating an initial cost body, embedding frequency information into the initial cost body, calculating depth estimation, performing back projection and the like in sequence; when multi-scale frequency sensing features are calculated, a double-branch frequency component enhancement module is arranged to process wavelet transform layer decomposition to obtain lossless approximate low-frequency components and high-frequency components of an input image, so that global consistency and local detail expression of depth estimation are improved.

The invention relates to the technical field of infrared spectroscopic analysis, in particular to a spectroscopic analysis-based corn kernel nutrient abundance and deficiency state evaluation method, which comprises the following steps of: irradiating corn kernels by using near-infrared and mid-infrared light sources, collecting reflected spectral signals, and recording spectral signal data; and based on the spectral signal data, extracting key wavelength region data to obtain key spectral data. According to the invention, the near-infrared light source and the mid-infrared light source are combined, so that the capability of acquiring nutrient information of the corn kernels is enhanced, the reflection characteristic of spectral signals is utilized, the interference of environmental factors on data acquisition is reduced, and the stability of spectral data is improved. The screening of key wavelength regions reduces irrelevant information, so that the spectral data is more concentrated on the characteristic wave bands influencing the nutrient state. The spectral signals are processed through wavelet transform, the resolution ratio of the characteristic peak value is improved, the local change trend of the spectral signals is enhanced, and characteristic extraction is more accurate.

The invention relates to the technical field of time sequence data processing, in particular to a drainage basin flood process automatic deduction method based on time sequence analysis, which comprises the following steps: obtaining drainage basin historical hydrological data, processing by applying adaptive wavelet transform, and extracting an intermediate frequency wavelet feature sequence and a transform residual sequence as data features; establishing a first calculation model, and generating a first predicted water level sequence; obtaining an error sequence based on the first predicted water level sequence and the actual water level sequence; constructing a second calculation model by using the data features and the rainfall data, and fitting the error sequence by using the second calculation model to obtain a prediction error sequence; performing weighted fusion on the first predicted water level sequence, the predicted error sequence and the rainfall sequence to generate a second predicted water level sequence; and combining the second predicted water level sequence with a hydrodynamic numerical calculation model, and obtaining predicted data through flow field calculation and analysis. According to the method, high-precision automatic deduction is realized through automatic processing and hybrid modeling.

The invention relates to an AC-DC integrated grounding fault positioning method and system, and the method comprises the steps: collecting the potential change data of a DC power supply system and the frequency characteristic data of an AC power supply system, and determining the electrical signal characteristics of stray current corrosion; decomposing the electric signal by using wavelet transform, and extracting direct-current and alternating-current components; determining a fault type according to the component amplitude; analyzing the potential gradient and the stray current flow direction in combination with a line topological structure, and determining the position of a fault point; measuring resistance to determine a grounding loop, and analyzing the correlation between corrosion distribution and electrical parameter characteristics; evaluating the influence of corrosion on the grounding path in combination with the material resistivity; according to the method, faults can be accurately recognized, fault points can be accurately positioned, the corrosion degree and the influence of the corrosion degree on the system can be evaluated, and effective support is provided for safe operation and maintenance of a subway power supply system.

The invention discloses a wafer process link pollution source identification method based on FOUP inner interlayer AMC distribution, and relates to the technical field of pollution source identification, the method comprises the following steps: analyzing airflow disturbance and pollutant diffusion characteristics, formulating an external detector deployment strategy, and collecting pollutant concentration and associated equipment operation parameters in real time; establishing a space-time matrix of pollutant concentration, extracting gradient characteristics of pollutant concentration change by using wavelet transform, and obtaining initial position data of a pollution source in combination with operation parameters of associated equipment; constructing a space-time gridding input matrix, designing a double-branch deep learning model, generating a pollution source positioning probability graph, and obtaining position data of a pollution source; the contribution weight of each associated device to pollutant diffusion is evaluated, and a directional treatment strategy is formulated and executed. According to the invention, the position of the pollution source can be accurately positioned, and the accuracy and response speed of pollution source identification are improved.

The invention discloses an adaptive fine-grained ocean ship rescue positioning method based on severe weather, and belongs to the technical field of ocean ship rescue positioning, and the method comprises the steps: collecting photo data, carrying out the feature extraction of pollution elements, and carrying out the fitting with the photo data to form a data set; based on a multi-scale linear attention decoupling head, the fine-grained classification features and the fine-grained regression features are decoupled; using wavelet transform features to capture bottom texture information, and aligning the bottom texture information with the features of the decoupling head; a loss function and a training model are provided, a total loss function value is calculated, back propagation is carried out, optimization is carried out through a selected optimizer and corresponding parameters, and a final ship positioning model is obtained. According to the method, a multi-scale linear attention decoupling head, a wavelet transform feature decomposition space semantic alignment module and a progressive reinforcement learning loss function fine-grained module are introduced, the fine-grained learning ability of the method is improved, the detection ability and robustness of the model are improved, and finally the ship positioning accuracy and efficiency in severe weather are improved.

The invention discloses a non-tracking freestyle three-dimensional ultrasonic image reconstruction method based on deep learning. The method comprises the following steps: obtaining an image data set; constructing an ultrasonic sequence coding module; constructing a wavelet transform convolution module; constructing a mixed attention module; constructing a loss function; constructing a deep learning model by combining an ultrasonic sequence coding module, a wavelet transform convolution module and a mixed attention module, and training the deep learning module by using the training set to obtain a three-dimensional ultrasonic image reconstruction model; scanning a diagnosis and treatment area of a patient by using an image acquisition system and adopting a non-linear scanning track to obtain two-dimensional ultrasonic image sequence data; and inputting the two-dimensional ultrasonic image sequence data as input data into a three-dimensional ultrasonic image reconstruction model for image reconstruction, and outputting a high-quality three-dimensional ultrasonic image. According to the method, the efficiency and the accuracy of three-dimensional reconstruction of the ultrasonic image can be remarkably improved, the adaptability and the robustness of processing the ultrasonic image under different conditions can be improved, and more accurate diagnosis information can be provided for clinic.