122 results about "Feature dependency" patented technology

The invention provides an intelligent auxiliary diagnosis and maintenance method and system based on multi-path recall, and relates to the technical field of fault processing, and the method comprises the steps: collecting equipment operation data, extracting and enhancing time sequence features after preprocessing, constructing a feature dependency graph, and generating a state feature vector. And carrying out fault feature recall by using the multi-path recall diagnosis network, fusing to generate a fault propagation map, and further generating a fault diagnosis result. And finally, generating a preliminary maintenance scheme based on a fault diagnosis result and the equipment state transition probability matrix, and optimizing the maintenance scheme by adopting a hierarchical reinforcement learning method in combination with a historical case and a fault propagation map to generate an optimal maintenance scheme. According to the invention, the accuracy of fault diagnosis and the effectiveness of a maintenance scheme can be improved, and the equipment maintenance cost is reduced.

The invention provides a robust multi-modal image segmentation method and system based on instance perception query, and the method comprises the steps: obtaining original image data and text description data, and generating training data through semantic analysis and causal graph construction; extracting image features and text features by adopting a multi-layer coding network, and establishing a feature dependency relationship through causal reasoning; generating instance clustering features based on attention calculation, and obtaining fusion features in combination with a multi-scale fusion strategy; and calculating query confidence and generating candidate segmentation masks, and outputting a final segmentation result through post-processing optimization. According to the method, by introducing causal reasoning and instance awareness query mechanisms, the cross-modal feature alignment precision and the robustness of the segmentation result are improved, and the method has better segmentation performance and generalization ability in a complex scene.

The invention provides a multi-modal image fusion method based on modal self-adaption and modal interaction compensation, and the method comprises the following steps: S1, obtaining a multi-modal image fusion data set, and obtaining a training data set through preprocessing; S2, analyzing the modal difference characteristics of infrared and visible light images, and evaluating the correlation characteristics of image pairs in different scenes; s3, capturing a cross-modal feature dependency relationship through a self-attention mechanism; s4, a differential feature extraction strategy is adopted, model parameters are optimized through iterative training, and multi-modal image fusion is completed; s5, a modal interaction compensation module is additionally arranged, unit dynamic balance common features and modal exclusive features are fused, feature complementation is achieved in channel and space dimensions, parameters of the modal interaction compensation module are optimized, the model is made to learn the optimal fusion weight of the multi-modal features in a self-adaptive mode, and multi-modal fusion image generation optimization is achieved through the model; according to the invention, multi-modal image fusion can be accurately and effectively carried out.

The invention discloses a point cloud segmentation method and system based on two channels, and relates to the technical field of computer vision and artificial intelligence, and the method comprises the steps: collecting point cloud data; creating a feature channel according to the three-dimensional point cloud data; and performing feature fusion and classification according to the feature channel. According to the point cloud segmentation method based on the two channels, different types of features are processed in different channels, high calculation efficiency is kept while global feature dependence is captured, and the method is suitable for processing and segmentation tasks of large-scale point cloud data. Two different network architectures are used, the structure and parameters of the model are flexibly adjusted to adapt to different application requirements, and a more efficient and reliable solution is provided for a point cloud segmentation task in a complex three-dimensional scene. The method has better effects in the aspects of segmentation precision, segmentation result accuracy and universality.

The invention relates to a brain load identification method based on electroencephalogram signals, and belongs to the technical field of man-machine interaction. Comprising the steps of obtaining electroencephalogram signals in different mental load states and preprocessing the electroencephalogram signals, constructing a feature map, constructing a mental load recognition model, training a network model, inputting a test set and outputting a test result. According to the method, electroencephalogram data representing brain load are extracted by using electroencephalogram signal equipment and are put into a brain load identification model for brain load classification, and feature dependence of time-frequency-space three-dimensional features of electroencephalogram signals can be extracted simultaneously by a three-dimensional wavelet convolution branch of the brain load identification model; due to the fact that the three-dimensional wavelet convolution and the spectrogram convolution are embedded into the double-branch parallel structure of the gating neural network, the brain load model can extract global features and local features at the same time, the problem that in a traditional neural network, the single-dimensional feature recognition rate is low is solved, meanwhile, the calculation complexity of brain load recognition is reduced through a single channel, and the brain load recognition efficiency is improved. And the operation cost is reduced.

The invention discloses an anomaly detection method based on a multi-mode denoising diffusion model, and belongs to the technical field of anomaly detection. Aiming at the problem that an existing anomaly detection method is easily influenced by single-mode information limitation and insufficient feature dependency relationship modeling in industrial quality inspection to cause a poor anomaly reconstruction effect, the method comprises the following steps: firstly, constructing a multi-mode large language model and a CLIP encoder, jointly generating an anomaly semantic description, and embedding a text mode prompt into a de-noising diffusion process; secondly, designing a denoising UNet network with a four-stage hierarchical structure, integrating a mixed attention enhancement Transform module in each stage, and synchronously capturing a long-range dependency relationship in space and channel dimensions and strengthening local details by performing traversal connection on channel-level self-attention and a pixel attention mechanism; and finally, realizing accurate reconstruction of the abnormal region through cross-modal feature alignment. Experiments show that the method improves the reconstruction quality of the abnormal image and the accuracy of anomaly detection, and has high robustness in a complex industrial scene.

The invention provides a network security malicious code binary search method and system, and relates to the technical field of data processing, and the method comprises the steps: calculating the multi-dimensional structural similarity between a to-be-detected binary file and each standard binary reference file based on a function level control flow semantic feature vector; determining a finally matched standard binary reference file according to the multi-dimensional structural similarity; performing fine-grained difference analysis on the binary file to be detected and the finally matched standard binary reference file to obtain a fine-grained difference analysis result; and on the basis of a fine-grained difference analysis result, in combination with sensitive data operation behavior feature detection, judging whether the to-be-detected binary file is a maliciously tampered version, and positioning a malicious code injection point. According to the method, the defects that in the prior art, dependence on fixed features is too high, and compilation optimization is sensitive are overcome.

The invention provides a false face image detection method based on misleading learning, and the method comprises the steps: S1, constructing a priori knowledge obtaining module, and constructing a basic counterfeit artifact discrimination capability for a counterfeit feature extractor and a detector; s2, establishing a misleading learning knowledge flow and a bias data knowledge flow; s3, introducing a single-channel attention fusion network, and enabling the model to adaptively select required multi-scale counterfeit latent layer features and multi-scale real picture latent layer features; s4, constructing a high-pass filter specially used for misleading learning, and carrying out preprocessing operation on the forged image; s5, misleading learning loss is introduced, and training constraint is carried out on the forged feature extractor; and S6, performing fine adjustment on the forged feature extractor and an adaptive device. According to the method, the model gets rid of feature dependence bias generated by interference of irrelevant semantic features in the picture on the model as far as possible, the method has excellent detection capability for forged pictures of different race statistical tags, and the optimal performance is achieved under intra-domain and cross-domain tests.

The invention discloses a radiology report generation method and system based on global dependency learning, and belongs to the technical field of natural language processing. According to the method, a global dependency learning module is designed, and the module integrates rotation position coding to enhance original Mama, so that long-sequence visual feature dependency is effectively modeled, and the model calculation efficiency is improved. According to the method, the problems that the visual long-term dependence modeling and calculation efficiency in the radiology image are difficult to effectively balance, and the heterogeneity of the characteristics of different modal information of the image and the text is difficult to effectively align and fuse are solved; the method not only can effectively capture the long-term dependency relationship on the key visual features in the radiology image, but also can improve the model calculation efficiency, and can enhance the alignment and fusion capability among the multi-modal heterogeneous information.

The invention discloses a tumor phenotype analysis method and system based on a disease source database and machine learning assistance, and relates to the field of artificial intelligence, and the method comprises the steps: obtaining clinical, gene and image multi-modal feature vectors of a tumor sample, inputting a multi-modal feature fusion network, and extracting each modal path vector; performing long-distance dependency modeling on clinical and gene features to obtain a first dependency feature map, and generating a second dependency feature map in combination with the feature association strength feature map; and determining a target tumor phenotype feature vector based on the second dependency feature map and the image feature path vector, finally calculating a matching confidence degree with a to-be-selected phenotype feature vector in a tumor phenotype database, and selecting a phenotype corresponding to the highest confidence degree as a target result. Through cross-modal feature dependence modeling and disease source database matching, the accuracy and efficiency of tumor phenotype analysis are improved, and support is provided for clinical precise diagnosis and treatment.

The invention discloses a process scheduling method and a computer storage medium. The method comprises the following steps: acquiring target interaction feature data, data dependence feature data and running context feature data of a target process, and respectively determining feature dependence intensity indexes of the target interaction feature data, the data dependence feature data and the running context feature data on the target process; determining a comprehensive dependence intensity index corresponding to the target process according to the feature dependence intensity indexes corresponding to the target interaction feature data, the data dependence feature data and the operation context feature data; and obtaining a scheduling overhead proportion of the target process, and adjusting process scheduling parameters of the target process according to the scheduling overhead proportion and the comprehensive dependency intensity index, thereby solving the problems of high scheduling overhead, low resource utilization rate and limited optimization effect of a scheduling method in related technologies, improving the resource utilization rate and the response speed, and improving the user experience. And the overall performance and stability of the system are obviously improved.

The invention provides a special disease first-aid full-process auxiliary decision-making method based on multi-modal perception and AI, and relates to the technical field of medical first aid, and the method comprises the steps: obtaining first-aid field multi-modal data, constructing a feature dependence graph, and recognizing the type and symptom features of a special disease; indexing special disease diagnosis and treatment data based on the symptom feature vector, and determining a reference case and a treatment path; the action space is constructed in combination with clinical specifications, the expected income is predicted by applying Monte Carlo search, and the optimal treatment scheme is generated, so that the first-aid special disease recognition accuracy can be improved, the treatment decision time can be shortened, and the first-aid intervention effect can be improved.

The invention provides a high-precision phase unwrapping method fused with a Transform self-attention mechanism. The high-precision phase unwrapping method is suitable for digital speckle interferometry, optical imaging and medical image processing. According to the method, on the basis of a U-Net structure, position coding and multi-head self-attention (MHSA) modules are introduced into a bottleneck layer, so that global feature dependence is established, and a long-range context is captured; a multi-head cross attention (MHCA) module is added in a decoding stage, attention weighting and noise suppression are performed on jump connection features, and selective fusion of key features is realized. A composite loss function is adopted in training, error variance loss and total variation loss are included, and phase periodic characteristics and smooth constraints are considered. Through simulation data training containing noise and complex phase jump, the unwrapping precision and robustness are significantly improved.

The invention relates to the technical field of sales prediction, in particular to an electronic product sales data prediction method and system based on artificial intelligence, and the method comprises the steps: obtaining historical sales time sequence data of a to-be-predicted product; constructing an initial sales prediction model according to the historical sales time sequence data; based on one or more predefined market impact morphological functions, generating an adversarial training sample, the market impact morphological functions being used for simulating external events causing abnormal fluctuation of sales volume; performing adversarial training on the initial sales prediction model by using a training data set containing adversarial training samples to obtain a prediction model with enhanced robustness; and predicting the future sales volume of the to-be-predicted product based on the prediction model with enhanced robustness. Therefore, the problems of strong external feature dependence, weak generalization ability to unknown impact, single model structure and the like in the prior art are solved.

The invention discloses an image defogging method based on multi-scale feature fusion. The method comprises: obtaining a foggy image and dividing the foggy image into a training set and a test set; constructing an original multi-scale feature fusion image defogging model, and performing training evaluation by using the training set and the test set to obtain a multi-scale feature fusion image defogging model; and inputting a to-be-processed foggy image into the multi-scale feature fusion image defogging model, and outputting a defogged image. According to the method, the problem that the feature information is easy to lose in the traditional image defogging process is effectively relieved, so that the accuracy of the defogging effect and the detail reduction capability are remarkably improved. A multi-scale feature fusion mechanism is introduced, and image semantic and texture information of different scales is cooperatively integrated. Compared with a single-scale feature dependence scheme, the method has the advantage that the information expression integrity and the detail retention capability are remarkably improved. In addition, the completeness and robustness of feature representation are further enhanced through the synergistic effect of the modules, so that the model shows higher performance and stability in dealing with a defogging task in a complex scene.

The invention provides a working face mine pressure prediction method based on domain generalization small sample learning and a model construction method thereof, and belongs to the technical field of signal processing. Constructing a small sample data set, and designing an efficient feature extraction network to perform feature extraction on the preprocessed mine pressure data; based on a feature extraction network, a generative adversarial network model for generators and discriminators of different domains is designed. Meanwhile, in domain generalization problem processing of source domain data and target domain data, an attention weighted mine pressure prediction network with a long and short-term memory network coding module, a guiding attention weighting module and a soft attention weighting module is constructed by utilizing the instructive coding function of a generator. The problem that the feature dependency relationship in the training data cannot be captured on the premise of small samples is further solved. According to the method, the prediction robustness can be improved, a new solution is provided for the complexity of mine pressure display of the coal mine working face, and the continuity and reliability of coal mine safety production are ensured.

The invention relates to the field of computer vision and artificial intelligence, in particular to an image event multi-mode semantic segmentation method, device and equipment, which can be applied to scenes such as automatic driving, robot perception and intelligent traffic. Time slices are divided through a fixed time window, event information is accumulated, and asynchronous event streams are converted into T * H * W voxel tensors; a selective state scanning mechanism of a Mama framework is used for replacing a self-attention mechanism of a traditional Transform, the calculation complexity is reduced while modeling global feature dependence is achieved, and the problems of video memory and delay in a Transform high-resolution scene are solved; besides, image textures and event edges are aligned through cross-space interaction, an event dynamic time sequence is captured through cross-time interaction, and modal inherent characteristics are retained through residual connection, so that feature degradation caused by excessive fusion is effectively avoided; finally, the image segmentation precision is effectively improved, and the processing efficiency and the model robustness are improved at the same time.

The embodiment of the invention discloses a multi-mode battery detection method and device, electronic equipment and a storage medium, and relates to the technical field of battery management, and the method comprises the steps: obtaining multi-dimensional original data such as battery temperature and voltage, carrying out the preprocessing, dividing the data according to a working state mode through KMeans clustering, constructing a data set through combining diversity sampling, and carrying out the detection of a multi-mode battery; and carrying out standardization processing and constructing a sliding window to obtain a time series data set. A single-layer LSTM network and a graph neural network are adopted to construct a dual-path encoder joint model based on a variational auto-encoder framework, and training is carried out. After to-be-detected data is obtained, the data is segmented by a sliding window, and the dependency relationship between the time sequence and the features is extracted respectively for splicing and compression. And restoring the sequence into a reconstruction sequence through a three-layer full-connection network, and calculating a mean square error and a regularization term to obtain a reconstruction error. And setting a dynamic threshold value, judging that an error is abnormal if the error exceeds the threshold value, calculating a feature contribution degree and generating a visual report. According to the method, the problems of low accuracy and efficiency and poor result interpretability in the prior art are effectively solved.

The invention discloses a heavy truck electric drive bridge fault diagnosis method based on a generative adversarial network, and the method achieves the precise analysis and diagnosis of the operation state of a heavy truck electric drive bridge through data collection and marking, data expansion, feature extraction model training, classifier model training and fault diagnosis. In the data expansion stage, a generative adversarial network based on adaptive stochastic disturbance is adopted, an entropy weight dynamic back propagation mechanism is introduced, the authenticity of generated data is improved by dynamically adjusting a learning rate, and distribution of the generated data is optimized by using a Gaussian mixture model and a feature dependency matrix; the feature extraction model is based on a full-connection neural network optimized by a bionic algorithm, a training rule is adjusted by adopting a self-correction mechanism, and the training efficiency and the model generalization ability are improved; the classifier adopts a high-order neural network of a reverse backtracking sparse strategy, the network structure and the feature weight are dynamically adjusted, and the sensitivity to abnormal data and the accuracy of fault classification are effectively enhanced. The method is high in fault diagnosis efficiency, high in precision and high in stability.

A point cloud semantic segmentation method based on a cross-modal Transformer belongs to the field of semantic segmentation technology. The method focuses on guiding the dense visual information of the camera image into the point cloud semantic segmentation task to complete the point cloud semantic segmentation task. First, the three-dimensional point cloud is unified to the camera image coordinate system according to the coordinate system transformation relationship, and then the perspective projection is used to obtain the two-dimensional representation of the three-dimensional point cloud. Then, the multi-scale feature map is calculated and extracted, and then cross-modal attention fusion is performed. The camera image and the projected point cloud image are fused at the feature level. Finally, the image is upsampled and classified. The classification result is projected onto the three-dimensional point cloud according to the inverse projection transformation relationship to complete the point cloud semantic segmentation task. The Transformer self-attention mechanism is used to establish cross-modal feature dependencies. The feature information of the two modalities of image and point cloud is combined to enhance the feature expression ability of the model. A boundary loss function is designed to emphasize the boundaries of semantic objects, thereby improving the boundary accuracy of object segmentation.

The application discloses a process scheduling method and a computer storage medium. The method comprises the following steps: obtaining target interaction feature data, data dependency feature data and running context feature data of a target process, respectively determining characteristic dependency strength indexes of the target interaction feature data, the data dependency feature data and the running context feature data on the target process; determining a comprehensive dependency strength index corresponding to the target process according to the characteristic dependency strength indexes corresponding to the target interaction feature data, the data dependency feature data and the running context feature data; obtaining a scheduling overhead ratio of the target process, and adjusting a process scheduling parameter of the target process according to the scheduling overhead ratio and the comprehensive dependency strength index, thereby solving the problems of high scheduling overhead, low resource utilization and limited optimization effect in the related art scheduling method, improving the resource utilization and response speed, and significantly improving the overall performance and stability of the system.

The invention provides a battery fault diagnosis method and system based on a hybrid neural network model, and belongs to the field of new energy automobile power battery fault diagnosis. According to the method, a CNN-LSTM hybrid neural network model of deep learning is adopted, and firstly, feature extraction and feature enhancement are performed through a convolution block and a residual attention module; then, time sequence modeling is carried out through a long-short-term memory network; and finally, global feature modeling is carried out through a Transform neural network algorithm, a global feature dependency relationship is established through a multi-head self-attention mechanism, and finally synchronous identification and accurate positioning of four typical faults of self-discharge abnormity, connection abnormity, sampling abnormity and insulation failure are realized.

The application belongs to the technical field of vehicle information anonymization detection, and particularly relates to a multi-type image anonymization annotation dataset construction and target coverage rate determination method. The method is based on a face and license plate image dataset with double annotation of theoretical anonymization region and anonymization features, introduces an anonymization feature extraction branch in an improved YOLOv5-L model and performs cross-modal feature fusion, combines a plurality of loss functions with theoretical anonymization region positioning loss as the core and a two-stage progressive training and difficult example mining mechanism, and realizes precise learning of the anonymization features. After normalizing the input anonymization image, the model inference obtains the theoretical anonymization region coordinates and maps them back to the original size, and through non-maximum suppression and matching of the IoU threshold, the region coverage rate is calculated to determine the positive detection, missed detection and statistical false detection rate. The application effectively overcomes the feature dependency failure and model robustness problem, and realizes high-precision, low-misjudgment anonymization detection and evaluation under various anonymization conditions.

The invention discloses an improved color image tampering detection and self-recovery method, which relates to the technical field of image processing, and is characterized by comprising the following steps: S1, generating a watermark; s11, an initial step; s12, authentication watermark generation: establishing a feature dependency relationship between blocks by adopting a Hash authentication feature algorithm so as to construct an authentication bit; s13, generating an abstract image: generating three minimum values by using a SparseViT algorithm; respectively scrambling and rearranging the three chaotic sequences by using a dynamic diagonal mapping algorithm so as to improve the recovery quality and enhance the safety; s2, embedding a watermark; s3, tampering detection; and S4, performing self-recovery. The invention aims to solve multiple defects of the existing watermarking technology in color image processing, and cope with new challenges brought by digital media content diversification and network environment complexity at the same time.

The invention discloses a deep learning tropical cyclone center positioning method based on multiband fusion. The method comprises the following steps: preprocessing a collected satellite image; constructing a deep learning model which comprises a backbone network feature extraction module, a feature fusion module and a feature enhancement module; extracting features of each tropical cyclone image through a backbone network feature extraction module; carrying out feature fusion on the extracted feature maps through a feature fusion module; the fused features are enhanced through a feature enhancement module; and the enhanced features are sent to ViT, global feature dependence in the image is captured through a self-attention mechanism, and a final output result is generated. According to the method, the information of the tropical cyclone on different wave band images is fully utilized, the spatial features are extracted, the spatial features extracted on the different wave band images are fully fused, the features of the different wave band images neglected by an existing model can be made up, and the positioning precision of the tropical cyclone center is effectively improved.

The invention provides a surface crack laser ultrasonic quantitative detection method, device, equipment and medium, and relates to the technical field of nondestructive detection.The method comprises the steps that laser ultrasonic B-scanning data are obtained through synchronous scanning, a multi-modal characteristic wave propagation mechanism is analyzed, a flight time analysis model is established, and the flight time analysis model is analyzed; and a theoretical and experimental multi-modal characteristic mode is constructed, and discretization similarity matching is carried out, so that the depth and the inclination angle of the surface crack are stably inverted in a full-parameter space, and quantitative detection is realized. According to the method, instability caused by dependence of a single mode or a single characteristic can be avoided, synchronous and high-precision quantitative detection on the surface crack depth and the crack inclination angle under the condition of complex multi-mode wave aliasing is realized, and the problems of insufficient characterization accuracy and robustness of geometric parameters of cracks in the prior art are solved.

A lightweight dual-branch convolutional neural network for image object detection and its detection method, which solves the lightweight image object detection task by constructing a dual-path network, consists of a dual-path backbone network, a feature pyramid and a lightweight detection head. The core module of the backbone network is an enhanced shuffle module ESB, and an efficient self-attention module ESAM is used to capture the global dependency of features while maintaining very few model parameters and computational complexity. The backbone network adopts a dual-branch network, one branch maintains high resolution, and the other branch uses continuous downsampling to obtain abstract semantic information. In the feature pyramid, ESAM is expanded to ECAM to capture the relationship between features of different resolutions. The detection head uses lightweight depth-separable convolution to predict borders and categories. Experimental results show that this network architecture and detection method achieve the best balance between detection accuracy and implementation efficiency, and become an effective method for solving lightweight image object detection tasks.

The invention relates to the field of intangible cultural heritage protection, and provides a non-abandoned pattern deconstruction regeneration system and method based on a multi-scale recurrent neural network. The system comprises a data acquisition and preprocessing module, a multi-scale feature extraction module, a recurrent neural network module, a pattern deconstruction module and a pattern regeneration module. The method comprises the steps of collecting non-abandoned pattern images through multiple channels, and constructing a data set through preprocessing; extracting pattern multi-scale features by using a multi-scale convolutional neural network, and capturing a feature dependency relationship by using a recurrent neural network; the pattern deconstruction module is used for decomposing the pattern into basic elements and analyzing the relationship of the basic elements to obtain structured representation; the pattern regeneration module generates a new pattern by means of a variational auto-encoder in combination with deconstruction information and random noise, and can control the style and complexity of the new pattern. The method effectively solves the problems of research and innovative application of traditional non-abandoned patterns, realizes efficient deconstruction and diversified regeneration of the non-abandoned patterns, and provides powerful technical support for non-abandoned culture inheritance and innovative development.

The application provides a disease-specific emergency whole-process auxiliary decision-making method based on multi-modal perception and AI, relates to the technical field of medical emergency, and comprises the following steps: acquiring multi-modal data of an emergency scene, constructing a feature dependency graph, and identifying disease types and symptom characteristics; indexing disease diagnosis and treatment data based on a symptom characteristic vector, determining reference cases and treatment paths; combining clinical norms to construct an action space, applying Monte Carlo search to predict expected returns, and generating an optimal treatment plan, which can improve the accuracy of emergency disease identification, shorten the treatment decision-making time, and improve the emergency intervention effect.

The invention discloses an open source software supply chain security risk intelligent assessment method, system and device based on deep learning and a medium, and belongs to the technical field of software security, and the method comprises the steps: obtaining feature data of an open source software component, and extracting software prior features, dependency relationship information and component identification information; software feature vector representation is obtained through embedded coding processing; identifying a node information missing and error dependency relationship, and carrying out noise reduction processing to obtain noise mark information; performing robust enhancement processing based on a graph neural network to obtain an enhanced dependency graph; inputting the software feature vector representation and the enhanced dependency graph into a semantic vector model to obtain fusion risk features; and a safety risk assessment result is obtained through analysis of the language model adjusted and optimized by the low-rank adapter. According to the method, the problem of noise interference in the dependency relationship is solved, effective compensation for incomplete or wrong dependency information is realized, and the accuracy and the reliability of open source software supply chain security risk assessment are improved.