22 results about "Linear modulation" patented technology

This invention discloses a method for inverting sea surface wind speed from GNSS reflected signals based on mask perception and image-based Mamba-CNN. The method includes: performing spatiotemporal matching and interpolation processing on multi-source spaceborne GNSS-R observation data and ERA5 reanalysis data; gridding the processed observation parameters according to geographical location to obtain a multi-channel grid image and acquiring a binarized occupancy mask; inputting the multi-channel grid image and the binarized occupancy mask into a mask perception Mamba-CNN model; calibrating the grid features through a mask conditional encoder and a feature linear modulation layer, and performing spatial modeling using feature extraction blocks combining convolutional and Mamba state space paths to obtain the wind speed inversion result; aggregating the extracted features using center-perception pooling, and outputting the final sea surface wind speed inversion image through a regression network. This invention can significantly reduce computational complexity, effectively reconstruct the continuous wind field structure, and improve the inversion accuracy and robustness in high-wind-speed areas.

This invention provides a production monitoring system for mirror cabinet processing equipment based on multi-source data, relating to the field of process control. The system acquires multi-source data from the mirror cabinet through an acquisition module and constructs a discrete mesh model; a deviation topology construction module generates directional selective edge weights and a nonlinear modulation function; a propagation control coupling module achieves synchronous iterative convergence of deviation states and control components; a topology constraint module identifies structural boundaries and divides control sub-regions; a region control module generates region control vectors, which are then decomposed into processing adjustment quantities and coupled in by the instruction module; and a feedback reconstruction module optimizes parameters based on the processing results. This invention enables deviation propagation to match the processing path, achieving bidirectional coupling of state and control, thus improving accuracy and control stability in the processing of complex curved surfaces and multi-material mirror cabinets.

The application discloses a panoramic image depth estimation method combining spherical correction and semi-supervised learning, innovatively designs a spherical correction (SR) mechanism based on spherical coordinate Fourier coding and feature linear modulation (FiLM), a lightweight plug-and-play PanoAdapter adaptive module, a teacher-student semi-supervised cross-domain knowledge transfer framework and a joint optimization strategy fusing pseudo-label distillation and consistency constraint, realizes apparent correction of equidistant cylindrical projection (ERP) nonlinear spherical distortion, alleviates the contradiction between panoramic labeling data scarcity and perspective model cross-domain migration difficulty, and improves the geometric consistency and generalization ability of panoramic depth estimation in diversified real scenes.

This invention provides a motion environment perception method based on biomechanical features and gait adaptation, belonging to the field of human-computer interaction perception. The method includes: acquiring the subject's original motion feature sequence and biomechanical feature vector; scaling and aligning the physical dimensions using a biomechanical scaling matrix to obtain a motion feature tensor; inputting the motion feature tensor and biomechanical feature vector into a multi-task temporal convolutional network for feature-level linear modulation to obtain modulated deep features; processing these features through a backbone network, with the classification branch outputting terrain categories and the regression branch outputting initial values ​​of environmental geometric parameters; freezing the backbone network parameters; fine-tuning the regression branch based on an adaptive loss function driven by gait phase and foot arch features to obtain a motion environment perception model; inputting real-time data from the subject; and outputting terrain categories and environmental geometric parameters. This invention solves the technical problem of low accuracy in environmental geometric parameter estimation caused by biomechanical differences between individual subjects in existing technologies.

The application discloses a washing machine motor operation control method and system based on fundamental wave voltage control, and relates to the technical field of washing machine motor control. The method comprises the following steps: S1, offline calculation of a relationship curve between the fundamental wave voltage and the reference voltage in the over-modulation region and storage of the curve in the memory; S2, calculation of the fundamental wave voltage by using a field-oriented control algorithm, wherein the d-axis and q-axis components of the fundamental wave voltage and the stator voltage are obtained through a speed loop and a current loop; S3, if the fundamental wave voltage is in the linear modulation region or exceeds the over-modulation region, an SVPWM algorithm in a 120-degree coordinate system is used for modulation; if the fundamental wave voltage is in the over-modulation region, an over-modulation algorithm in the 120-degree coordinate system is used to calculate the three-phase voltage action time, and finally, the SVPWM algorithm is used for modulation. The application directly takes the fundamental wave voltage as a control target, avoids the uncertainty caused by the inconsistency between the reference voltage and the fundamental wave voltage, and thus realizes accurate control of the fundamental wave voltage which has an actual effect.

The application discloses a two-dimensional ferroelectric transistor device based on controllable nucleation to realize linear modulation and a preparation and micro-characterization method thereof, and the method comprises the following steps: sequentially preparing a bottom electrode and a ferroelectric layer on a substrate; preparing a planar structure covering the surface of the ferroelectric layer, wherein the planar structure comprises a conductive layer and a two-dimensional semiconductor layer, and is used for adjusting the internal electric field distribution of the ferroelectric layer to control the nucleation of domains to occur at the interface between the conductive layer and the two-dimensional semiconductor layer; preparing an insulating layer and a metal electrode to complete the preparation of the two-dimensional ferroelectric transistor device; and micro-characterizing the two-dimensional ferroelectric transistor device to verify the domain nucleation and flipping rule and obtain the device resistance state change under different domain distributions. The application realizes linear flipping of ferroelectric domains by introducing a controllable nucleation process, prepares a transistor device with linear conductance modulation characteristics, and micro-verify by using advanced micro-characterization technology, and provides a closed-loop process which can guide process optimization.

The application discloses a kind of based on enhancing geometric semantic GNN's stop block stress field prediction method, this method belongs to the intersection field of computational mechanics and artificial intelligence.The method is aimed at the hatch stop block with hole, curved surface, contact boundary and other significant geometric characteristics, first, construct the geometric semantic graph that fuses node coordinates, principal curvature, normal vector, boundary identifier and hole distance;Further, design the graph neural network of Encoder-Processor-Decoder architecture, wherein the Feature-wise Linear Modulation (FiLM) mechanism is introduced in the Processor module, and the load-driven feature linear modulation (FiLM) mechanism is introduced in the Processor module, to realize the message passing of load-geometry cooperation;Through end-to-end training, the model can output high-precision stress field in milliseconds to seconds.It is suitable for intelligent health monitoring and digital twin system of key components in the fields of aviation, automobile and the like.

This invention discloses a lightweight multi-task context control image processing system, including a first image encoder, a second image encoder, a multi-scale fusion module, and an output feature processing module. The multi-scale fusion module includes N multi-scale fusion units connected sequentially from high to low spatial resolution. The first and second image encoders are used to extract features and downsample images from sample paths and control paths. The multi-scale fusion units are used to fuse query sample features and task control features at different spatial resolutions. Each multi-scale fusion unit includes a residual convolution module, a sliding window cross-attention module, a feature linear modulation module, a refinement module, and a downsampling module. This invention uses lightweight modules such as depthwise separable convolution and group normalization, significantly reducing the number of parameters compared to traditional control networks. The output feature channel is compatible with downstream diffusion models and can be plugged into existing diffusion models without extensive modifications to the original model architecture.

The application discloses a construction site target detection method and system based on multi-scale gating reweighting and contrast learning, which adaptively selects a scale set participating in interaction through a gating coefficient, designs a proposal reweighting module, and combines local branches, sparse attention branches and global context branches to perform in parallel; meanwhile, a refined reweighting module is constructed, a two-dimensional displacement field is predicted to realize geometric alignment of a class prototype, and attention mask and semantic mask are fused for feature weighting; a cross-batch contrast learning strategy is introduced, a class grouping memory bank is maintained, a momentum encoder is used to generate a key vector, and a class balance and difficult example priority strategy are combined; in an embedding space, a cross-batch contrast loss and a prototype contrast loss are jointly used, an N-way K-shot subtask is constructed through meta learning, and cross-domain knowledge is obtained through linear modulation of features. The application significantly improves the detection accuracy and cross-class adaptability in a construction site scene, while ensuring real-time performance.

The application discloses a micro-service time sequence prediction method based on multi-scale two-way fusion coding condition diffusion. First, a local segment of a micro-service key performance indicator and a time characteristic matrix thereof are acquired; local historical characteristics and global historical characteristics of the micro-service key performance indicator are extracted and fused to obtain historical fusion characteristics of the micro-service key performance indicator; then, a condition diffusion network is constructed, including a Transformer decoder, a gate linear unit and a characteristic linear modulation, and the historical fusion characteristics of the micro-service key performance indicator are taken as condition information of multi-head attention of the Transformer decoder; finally, based on multi-scale autoregressive diffusion, the condition diffusion network is used to gradually and reversely denoise the local segment of the micro-service key performance indicator at a prediction starting position, weighted fusion is performed on micro-service key performance indicator prediction results at various steps, and finally, micro-service key performance indicator prediction results are obtained. Through capturing local and global information and fully fusing, the prediction accuracy is improved.

The application discloses a human body kinetic energy collection device based on nonlinear magnetic response and belongs to the technical field of energy collection and wearable devices. The human body kinetic energy collection device comprises an upper stator disc, a lower stator disc, a rotor magnet disc, an eccentric mass block and a nonlinear modulation magnet. The upper stator disc, the rotor magnet disc and the lower stator disc are coaxially stacked to form a containing cavity for the rotation of the rotor. The eccentric mass block is fixed to the outer edge of the rotor magnet disc and provides an eccentric driving torque by using the gravity component when the human body arm swings. The nonlinear modulation magnet is fixed to the gap of the trapezoidal coil on the stator disc, and the periodic interference between the alternating magnetic poles of the rotor magnet disc and the rotor magnet disc is generated when the rotor rotates, thereby constructing a single-dual stable nonlinear potential energy field. The nonlinear electromagnetic micro energy capture device is used for efficiently collecting the low-frequency and random arm swing kinetic energy of the human body, and the limitation of the narrow resonant frequency band and high starting threshold of the conventional single stable linear system is broken through.

The application discloses a building concrete internal defect ultrasonic detection signal processing method and system, belongs to the material nondestructive testing technical field, and the method comprises the following steps: configuring a full matrix capture array to collect multi-channel original time domain data; performing a time domain flip operation on the data and mapping to the full matrix capture array to perform a synchronous retransmission algorithm; extracting sum frequency and difference frequency amplitudes of specific frequency components in the matrix, and constructing a nonlinear amplitude coordinate mapping table; performing statistical analysis on a full-field energy focusing distribution matrix, performing a difference operation with the sum frequency component amplitudes to generate a defect response coordinate set; and feeding back the coordinate set to the full matrix capture array to trigger supplementary scanning and update the matrix. The application adopts dual-frequency nonlinear modulation and time reversal focusing technology, can effectively suppress aggregate scattering noise interference, and improves the three-dimensional imaging precision and detection signal-to-noise ratio of concrete micro defects.

The application provides a pitch correction method and device based on fundamental frequency modulation, equipment and medium, relates to the technical field of speech processing, and comprises the following steps: extracting a speech content sequence from an original song audio to be corrected; obtaining a timbre feature vector of a target speaker corresponding to the original song audio; performing fundamental frequency coding on a target fundamental frequency curve of a target song corresponding to the song audio to obtain a fundamental frequency condition feature of the target song; performing first feature linear modulation on the speech content sequence and the fundamental frequency condition feature to obtain a pitch mean condition; performing condition diffusion according to the pitch mean condition, the timbre feature vector and the fundamental frequency condition feature to generate a target mel-frequency spectrum; and performing audio waveform reconstruction according to the target mel-frequency spectrum to obtain a target song audio after pitch correction. The application can improve the pitch correction effect of the song.

A physical constraint spectral codebook-based intelligent fault diagnosis method for aero-engines belongs to the fields of aero-engine fault diagnosis and artificial intelligence technology. First, the multi-source vibration signals of the aero-engine are normalized in the order domain and injected with rotational speed information through feature linear modulation. Then, a one-dimensional convolutional neural network and spectral attention mechanism are used to extract fault-sensitive features. Next, a physical constraint spectral codebook is constructed, embedding the order ratio of aero-engine bearing fault features into the vector quantization codebook structure in the form of a sinusoidal basis template, and using hierarchical dual-space quantization to separate sensor features and fault representations. Finally, prototype matching inference is used in the codebook activation histogram space to achieve fault diagnosis across operating conditions and models with few samples. This invention integrates fault physics knowledge into discrete representation learning at the structural level, giving codebook entries interpretable physical semantics, thus solving the problems of missing physical knowledge and insufficient generalization ability in aero-engine fault diagnosis across operating conditions and models.

This application provides an absolute pressure sensor with an F-P cavity structure and a method for detecting air pressure. The absolute pressure sensor includes: a light source module for linearly modulating the received external sawtooth wave modulation signal to output continuous laser light; an optical fiber transmission module for unidirectional transmission of the continuous laser light to block the reflected light of the continuous laser light from returning to the light source module; an F-P interference cavity including an exit surface of an optical fiber collimator coated with a low-reflectivity film and a total reflection mirror fixed on a pressure-sensitive elastic element. The pressure-sensitive elastic element senses changes in air pressure and deforms, causing the total reflection mirror to shift. The continuous laser light is reflected on the exit surface and the total reflection mirror to form coherent light, generating a beat frequency optical signal carrying air pressure information; a photoelectric detection module for processing the beat frequency optical signal carrying air pressure information transmitted via the optical fiber transmission module to obtain an electrical signal that meets demodulation requirements; and a signal demodulation module for processing the electrical signal that meets demodulation requirements to obtain the air pressure value.

The application discloses a zero sequence injection modulation method of a modular multilevel matrix converter and relates to the technical field of power electronic control. The modular multilevel matrix converter is connected with two AC systems with different frequencies; the method comprises the following steps: obtaining modulation wave components of three-phase systems on input and output sides respectively through capacitor voltage outer loop control and input current inner loop control of a sub-converter and overall output current control; for each bridge arm, corresponding modulation waves are calculated based on the two modulation wave components; a zero sequence modulation voltage is calculated based on the modulation wave components of the three-phase system on the input side; the zero sequence modulation voltage is injected into each modulation wave to obtain a zero sequence modulation wave; all the zero sequence modulation waves are modulated by using a carrier wave to generate trigger pulses which are distributed to switch devices of corresponding bridge arms. The method can expand the linear modulation range and operation range of the modular multilevel matrix converter and improve the utilization rate of capacitor voltages of sub-modules by injecting zero sequence components into modulation waves.

This invention discloses a method and system for enhancing ultra-high resolution low-light images based on block overlap inference, belonging to the field of image processing technology. The method segments the ultra-high resolution image into sub-blocks with overlapping regions, and uses a deep learning model based on the UNet architecture to independently enhance each sub-block. The model incorporates a Feature Linear Modulation (FiLM) module to adaptively modulate the underlying features using target brightness parameters. It calculates the ratio of the local average brightness of the sub-block to the global average brightness, and uses logarithmic scaling to determine the local target brightness, ensuring brightness continuity. The overlapping regions are progressively fused using a cosine window function to eliminate block artifacts. This invention solves the problems of memory limitations, block artifacts, and brightness discontinuities in low-light enhancement of ultra-high resolution images, achieving high-quality real-time enhancement, and is suitable for scenarios such as wide-area monitoring and astronomical observation.

The application relates to the technical field of target detection, and discloses a light detection method and device, computer equipment and a storage medium. The encoder comprises at least one of an interactive control design module and a quality attention module. A backbone network is used to extract features of a target scene image, and initial feature data of different scales is obtained. The initial feature data of different scales is subjected to pyramid bidirectional fusion processing through the encoder. In the pyramid bidirectional fusion processing, the number of channels is controlled through a channel mapping convolution layer, linear modulation processing, low-contrast and low-brightness feature enhancement processing are performed through the interactive control design module, and / or global quality coding and internal scale feature interaction processing are performed through the quality attention module, so that fusion enhanced feature data of different scales is obtained. The fusion enhanced feature data of different scales is classified and regressed through a detection head, and target detection results of different scales are output.

The application discloses an auditory brainstem response automatic classification method and device based on a time-frequency cross attention network, equipment and a storage medium, relates to the technical field of hearing monitoring, and through double-branch parallel processing of ABR signals: a one-dimensional convolution is adopted in a time domain branch to extract local waveform form features, and a short-time Fourier transform is adopted in a frequency domain branch to obtain a time-frequency spectrum feature; a feature linear modulation mechanism is introduced into both branches, a stimulus intensity is taken as condition information to adaptively scale and offset intermediate features; complementary fusion of time-frequency features is realized through a bidirectional cross attention mechanism, time domain features can query frequency domain information, and frequency domain features can query time domain information; global average pooling, stimulus intensity late fusion and a multilayer perception machine are combined to make a binary classification decision. In the automatic classification, multi-modal information and physiological priori can be fully utilized without manual feature engineering, and the accuracy, objectivity and efficiency of hearing screening and diagnosis are significantly improved.

The objective of this invention is to enable the identification of various conditions in a vehicle and driver condition estimation system using an LLM (Limited Linear Modulation) even without defining the scenarios to be identified. The LLM-based vehicle and driver condition estimation system comprises: a conversion unit that converts input data of a first type related to at least one of the vehicle's condition and the user's condition into input data of a second type that is easier to understand in an LLM than the first type; and an estimation unit that takes the converted input data as input and uses an LLM to estimate data related to the at least one of the conditions.

This application provides a metasurface array structure, relating to the field of communication technology. The metasurface array structure includes a first metal layer, a first dielectric layer, a second metal layer, a second dielectric layer, and a phase-shifting network layer. The first metal layer includes multiple array units, each including a reflective patch. The first dielectric layer is located on one side of the first metal layer; the second metal layer is located on one side of the first dielectric layer; the second dielectric layer is located on one side of the second metal layer; the phase-shifting network layer is located on one side of the second dielectric layer, and includes a first phase-shifting network structure and a second phase-shifting network structure. The first and second phase-shifting network structures are electrically connected to the reflective patches, respectively. In this way, not only can bandwidth and linear modulation be achieved over a wider frequency band, but also dual-polarization independent operation can be achieved, effectively reducing the size of the array units, improving the scanning capability of the array units, and avoiding interference from the first and second phase-shifting network structures to the reflective patches.

The application discloses a building concrete internal defect ultrasonic detection signal processing method and system, belongs to the material nondestructive testing technical field, and the method comprises the following steps: configuring a full matrix capture array to collect multi-channel original time domain data; performing a time domain flip operation on the data and mapping to the full matrix capture array to perform a synchronous retransmission algorithm; extracting sum frequency and difference frequency amplitudes of specific frequency components in the matrix, and constructing a nonlinear amplitude coordinate mapping table; performing statistical analysis on a full-field energy focusing distribution matrix, performing a difference operation with the sum frequency component amplitudes to generate a defect response coordinate set; and feeding back the coordinate set to the full matrix capture array to trigger supplementary scanning and update the matrix. The application adopts dual-frequency nonlinear modulation and time reversal focusing technology, can effectively suppress aggregate scattering noise interference, and improves the three-dimensional imaging precision and detection signal-to-noise ratio of concrete micro defects.