14 results about "Feature dependency" patented technology

The application discloses a kind of multi-modal data quality evaluation method and system of fusion meta-model and big model, first construct the evaluation meta-model including object set, morphism set, combination sub-set and feature dependency graph, then receive multi-modal data evaluation task and automatically convert multi-modal data evaluation task into a composite morphism in evaluation meta-model;The application realizes the function of multi-modal data quality evaluation with the object set, morphism set, combination sub-set and feature dependency graph for constructing the evaluation meta-model based on category theory, and by abstracting evaluation operation as morphism, evaluation process as composite morphism and shared computing as dependency graph, a combinable, optimizable and reusable evaluation framework can be constructed, and by evaluation functor, heterogeneous score output can be unified standardized, the application can unify the logic of multi-modal data quality evaluation, automatic scheduling and knowledge sedimentation.

The application discloses a kind of nuclear instance segmentation model and method based on attention and ellipse regularization, belong to deep learning image processing technical field, model includes from shallow to deep window attention mechanism feature extraction module, long-distance feature dependent attention fusion module and ellipse regularization module, input image is extracted feature by from shallow to deep window attention mechanism feature extraction module, then by long-distance feature dependent attention feature fusion, finally by ellipse regularization module constraint model.The application uses the above instance segmentation model and method, for the small and elliptical shape of cell nucleus is constrained training, strengthens the image feature extraction capability based on bounding box instance segmentation, more effectively separates the adherent cell nucleus, can extract more complete cell nucleus density, morphology and location information.

The application relates to the technical field of structural health monitoring and engineering safety, in particular to a dam automatic multi-element data fusion monitoring system and method based on the Internet of Things; the method comprises the following steps: an initial state reference without historical sample dependence is established by deploying checked heterogeneous sensing units on site and constructing a dynamic baseline mechanism; then, cross-scale mapping matrices and feature-dependent correction coefficients are used to perform multi-domain analysis and scale self-adaptive adjustment on time series data, so that the difference capturing capability is enhanced; a dynamic credibility adjustment and multi-source weighted fusion method is adopted at the edge node, preliminary event discrimination is performed in combination with physical connectivity and neighborhood cooperation; finally, abnormal diffusion trend analysis and dynamic early warning level determination are realized based on a space-time evolution matrix and a probability prediction model, and a response strategy and system parameter optimization are automatically generated through a closed-loop feedback mechanism. The application significantly improves the accuracy, self-adaptive capability and risk control level of dam monitoring.

This invention discloses a dynamic scattering medium imaging method based on Mamba channel feature dependency and dual-domain learning, belonging to the field of scattering medium imaging technology. It employs a ScatMamba network with an encoder-decoder architecture to extract initial features from speckle images formed by dynamic scattering media. Multi-scale high-level semantic features are extracted through downsampling and channel filtering enhancement of the Mamba module, and frequency domain global feature modulation is achieved using a spectral modulation module at the encoder bottleneck layer. The decoder fuses multi-scale features through upsampling and skip connections, outputting a clear reconstructed image. End-to-end training is completed using a spatial-frequency dual-domain composite loss function. This invention achieves joint modeling of channel feature dependency and spatial long-range dependency through channel filtering and enhanced Mamba layers, utilizes a spectral modulation layer for adaptive control of high and low frequency components, and strengthens the robustness of the reconstruction mapping through a dual-domain loss function. This enables efficient decoupling of speckle features in dynamic scattering environments, restoring the global structure and fine texture of the target.

The application discloses a kind of perception decision integration target detection and tracking method based on high-low view angle cooperation, comprising: high-low video stream access and timing multimodal alignment based on network time synchronization protocol;Based on the backbone layer network target detection and depth feature extraction of CSPDarknet;Based on homography correction and geographic coordinate registration, cross-view space mapping;Based on the track association integration of unified geographic spatial distance measurement and bipartite graph matching;Based on the perception decision integration generation of multi-branch spatio-temporal graph convolution network ST-GCN.The application can solve the problem of feature collapse and matching recognition error rate soaring caused by the dependence of existing technology on re-identification depth appearance features in response to high-low pitch angle sharp crossing, realize the effect of uninterrupted and high-precision entity re-anchor in the target appearance distortion state across the scale deformation barrier.

The invention relates to the technical field of audio identification, and discloses a forged audio identification method and system, and the method comprises the steps: carrying out the framing and windowing of original audio data, and obtaining frame sequence audio data; performing short-time Fourier transform on the frame sequence audio data to obtain frequency domain representation data and an initial acoustic feature vector; performing depth feature reconstruction on the initial acoustic feature vector, and performing multi-modal feature fusion on the reconstructed feature tensor and the frame sequence audio data to obtain a joint identification feature set; performing time sequence context analysis on the joint identification feature set, and capturing a feature dependency relationship of the frame sequence audio data to obtain a context enhanced feature sequence; quantifying the cosine similarity between the context enhanced feature sequence and the real voiceprint template, and analyzing the feature mutation amplitude of the context enhanced feature sequence to obtain dual identification measurement; performing comprehensive scoring on the dual identification measurement to obtain an audio identification conclusion; according to the invention, the efficiency of counterfeit audio identification can be improved.

This invention provides a fake face image detection method based on misleading learning. S1: Construct a prior knowledge acquisition module to build a basic fake artifact detection capability for the fake feature extractor and detector; S2: Build a misleading learning knowledge flow and a biased data knowledge flow; S3: Introduce a single-channel attention fusion network to enable the model to adaptively select the required multi-scale fake latent features and multi-scale real image latent features; S4: Construct a high-pass filter specifically for misleading learning to preprocess the fake images; S5: Introduce a misleading learning loss to constrain the training of the fake feature extractor; S6: Fine-tune the fake feature extractor using an external adaptor. This invention enables the model to minimize the feature dependence bias caused by irrelevant semantic features in the image, exhibiting excellent detection capabilities for fake images with different statistical labels, achieving optimal performance in both intra-domain and cross-domain tests.

This invention discloses a 3D human-object interaction reconstruction method based on multi-scale correction and frequency-domain adaptive graph convolution, belonging to the field of computer vision and 3D reconstruction technology. By extracting features from the input data, a multi-scale correction mechanism is used to capture human-object interaction features at different scales, optimizing the completeness of feature representation. A frequency-domain adaptive graph convolution module is introduced to adaptively learn frequency-domain feature dependencies, effectively suppressing high-frequency noise interference. Through feature fusion and reconstruction, an accurate 3D human-object interaction model is output. This invention, through the synergistic effect of multi-scale correction and frequency-domain adaptive graph convolution, improves the reconstruction stability and geometric accuracy in complex interaction scenarios, and can be widely applied in virtual reality, augmented reality, film and television production, etc., possessing good practicality and promotional value.

This invention discloses a gearbox fault diagnosis method based on time-frequency feature extraction and majority voting decision-making, belonging to the field of wind power equipment condition monitoring and fault diagnosis technology. The method first segments the acquired raw vibration signal and simultaneously extracts the time-domain and frequency-domain statistical features of each data segment. Then, a one-dimensional convolutional neural network model is constructed, using the raw data segment as input to the convolutional layer to extract deep features. The statistical features are then fused with the deep features and input into a classifier. In the diagnosis stage, this invention employs a majority voting decision-making strategy, statistically analyzing the classification results of multiple sample segments from a long signal segment, and determining the final health status of the equipment based on the "majority rule" principle. This method, through physical feature fusion and voting mechanisms, effectively solves the problems of strong dependence on single features and poor anti-interference ability in traditional methods, significantly improving the accuracy and robustness of fault diagnosis.

Machine learning using dependency priors includes determining task-dependent feature dependencies for a group of features extracted from non-imaging data received with the computer hardware. The non-imaging data can be reformatted based on the task-dependent feature dependencies. The reformatting can convert strongly dependent features among the group of features into one or more subgroups based on task-specific, feature dependency-based priors. Based on the reformatted data, a machine learning prediction can be generated by a machine learning model.

This invention discloses a conditional compilation and dynamic feature management system, belonging to the field of software build and configuration management technology. The system includes a build-time conditional import module configured to selectively import modules using conditional expressions. These conditional expressions are evaluated during the build phase, and the module packaging tool completely removes module dependencies corresponding to inactive code branches during the build process. The system also includes a runtime feature checking module and a hierarchical feature control module, distinguishing between build-time feature control and runtime feature control. Build-time feature control achieves code elimination through conditional expressions, while runtime feature control dynamically controls the execution path through status query callbacks. The system further includes an on-demand loading module for delayed initialization of heavy subsystems; a code elimination verification module to verify the complete elimination of inactive code; a feature dependency management module to manage dependencies and mutual exclusion relationships between feature flags; a feature documentation generation module to automatically generate feature configuration documents; and a feature flag definition module to centrally define feature flags to ensure type safety. This invention achieves collaborative work between compile-time code optimization and runtime behavior control, balancing streamlined build artifacts with runtime flexibility, and can be widely applied to feature management and build optimization of large-scale software systems.

The present application relates to the technical field of audio discrimination, and discloses a counterfeit audio discrimination method and system, the method comprising: performing frame windowing processing on original audio data to obtain frame sequence audio data; performing short-time Fourier transform on the frame sequence audio data to obtain frequency domain representation data and an initial acoustic feature vector; performing deep feature reconstruction on the initial acoustic feature vector, and performing multi-modal feature fusion on the reconstructed feature tensor and the frame sequence audio data to obtain a joint discrimination feature set; performing timing context analysis on the joint discrimination feature set to capture feature dependency relationships of the frame sequence audio data, and obtaining a context-enhanced feature sequence; quantifying the cosine similarity between the context-enhanced feature sequence and a real voiceprint template, and analyzing the feature mutation amplitude of the context-enhanced feature sequence to obtain a double discrimination metric; and performing comprehensive scoring on the double discrimination metric to obtain an audio discrimination conclusion; the present application can improve the efficiency of counterfeit audio discrimination.

This invention discloses a PCB surface defect detection method based on context enhancement and multi-feature fusion, relating to the fields of image processing and machine vision. The method includes the following steps: inputting a preprocessed PCB image into a defect detection model; enhancing the model's feature extraction capability through an improved lightweight backbone network, reducing computational requirements and expanding its application scenarios; simultaneously establishing long-range feature dependencies to improve the recognition ability of objects difficult to detect in complex scenes; introducing a three-feature fusion module into the improved neck network and enhancing shallow features, integrating multi-scale information, reducing feature redundancy, and enhancing the ability to capture surface defects; finally, outputting the corresponding defect category and location. This invention effectively improves the detection accuracy of small and multi-scale defects on PCB surfaces, exhibiting good adaptability and robustness, especially in complex circuit backgrounds.

The application relates to the field of artificial intelligence and computer image generation technology, in particular to a multi-dimensional feature extraction and generation system for an animation character based on deep learning; the system comprises a multi-dimensional feature decoupling and graph construction unit, an implicit space topology and physical constraint unit, a semantic and style coupling generation unit, and an entity consistency evaluation and interception unit; multi-modal input including at least target semantic labels and target visual angle parameters is received, apparent features, structural features and semantic variable features are extracted and a feature dependency graph is constructed, a pseudo-three-dimensional coordinate system is established in a latent space and visual angle conversion and physical correction are performed, a candidate image is generated based on a diffusion model, a topology deviation risk value and a style deviation risk value are calculated, a redrawing instruction is generated for local redrawing when the values exceed a threshold, and a final image and a consistency evaluation report are output when the values meet the threshold; the application can realize identity consistency control of the same animation character under multi-view and multi-semantic conditions.