123 results about "Data diversity" patented technology

The invention provides a ship noise multi-feature classifier data enhancement method and system based on a multi-fine-grained conditional diffusion model. And compressing a waveform to a potential space through VQ-VAE, extracting a ship type / ship name cross semantic vector by using ResNet, and optimizing clustering in combination with a loss function. And a one-dimensional U-Net conditional diffusion model is constructed, unconditional / conditional model output is dynamically weighted and fused, and the weight is adaptively adjusted according to training loss. In the generation stage, a semantic prototype is constructed by using a high-fine-granularity label, parameters are determined by using low / medium-granularity mean value sampling and Bayesian optimization, and fine-granularity controllable waveform generation is realized. After the generated data is converted into multiple features such as MFCC and Lofar, the generated data and original data are combined to train a classifier, and a virtual class strategy relieves class imbalance. Experiments show that the MSE of generated data and real data is reduced, the classification accuracy is improved, the data diversity and the model generalization ability are remarkably enhanced, and the method is suitable for scenes such as underwater target recognition.

The invention discloses a face privacy protection-oriented generated image detection method and a face privacy protection-oriented generated image detection system. The method comprises the following steps of: firstly, preparing face and text pairing data, and finely adjusting a diffusion model; secondly, on the basis of the diffusion model after fine tuning, potential vectors are extracted and clustered, and text prompts and center vectors obtained through clustering form a dictionary; and finally, based on the obtained dictionary, obtaining a pseudo image and a label through the fine-tuned diffusion model, and outputting a detection result through a classifier. According to the method, potential spatial clustering and conditional diffusion generation are combined, privacy protection and data diversity are taken into consideration, and the security and generalization ability of forged face image detection are remarkably improved.

The invention discloses a large-scale language model fine tuning method, device and equipment and a storage medium, and relates to the technical field of large-scale language model fine tuning. According to the method, a preset check point is firstly subjected to large-scale language model fine tuning based on training loss performance, sample embedding space distribution and sample reply score performance of a current model; and performing multi-dimensional self-reference diagnosis on the training data set, identifying a suboptimal sample which is not matched with the capability of the current model, and then processing the suboptimal sample through an adaptive optimization engine to complete dynamic evolution of the training data set. Therefore, dynamic evolution of the data set adaptation model can be realized, the problem of redundancy or insufficient adaptation of static data is avoided, and the training efficiency is improved; meanwhile, data waste is reduced, data diversity is reserved, the model generalization ability is promoted, manual intervention is not needed, and the data optimization cost is reduced.

The invention discloses an improved YOLOv12-based aircraft surface damage detection algorithm and system, and relates to the technical field of aircraft surface damage, and the method comprises the following steps: S1, data collection and preprocessing: employing an unmanned plane and an unmanned vehicle to collect the related data of the aircraft surface damage, precisely marking the aircraft surface image, and increasing the data diversity; s2, model improvement: improving a YOLOv12 aircraft surface damage detection model according to an aircraft surface detection task; a high resolution input size is employed to capture small defect features. According to the method, the YOLOv12 aircraft surface damage detection model is improved, the unmanned aerial vehicle and the unmanned vehicle can comprehensively record key structural components on the surface of the aircraft under various illumination conditions, the data integrity is guaranteed, and through the CLAHE algorithm, diversified data and the imaging effect in various environments, the detection accuracy of the aircraft surface damage detection model is improved, and the detection accuracy of the aircraft surface damage detection model is improved. And the robustness of subsequent model training can be obviously improved, so that the detection algorithm has higher environmental adaptability.

The application provides a bearing data expansion method and system based on a Gaussian mixture model and a particle swarm optimization, and belongs to the field of data expansion. In order to solve the problem of how to effectively expand the data set, increase the data diversity, alleviate the data scarcity, and improve the generalization ability and detection performance of the model in the existing small sample bearing defect detection, the application adopts a Gaussian mixture model (GMM) to model the original data, optimizes the parameters of the GMM through a particle swarm optimization (PSO) algorithm, generates new data points, and expands the data set. The method can effectively increase the data diversity, alleviate the data scarcity problem, and improve the generalization ability and detection performance of the model.

The invention discloses a UWB radar multi-target sensing and tracking method and system. The method comprises the steps that multi-scene original echo signals are collected through a UWB radar, a synchronous optical motion capture system obtains real position information of a target, data diversity is enhanced through noise injection and time-frequency transformation, and a multi-modal data set is constructed; de-noising and signal enhancement are carried out by adopting a three-stage cascade processing flow, and features are extracted by utilizing a CNN and Transform hybrid encoder; based on a graph neural network and a dynamic filtering theory, constructing a model of an improved PointPill detection head and a GAT tracking head to carry out target detection and track association; and finally deploying to edge equipment to realize real-time multi-target sensing and tracking by combining a multi-task joint loss function with a strategy optimization model such as adaptive loss balance. According to the method, the traditional multi-sensor dependence is broken through, the track continuity in a complex scene is improved in a target shielding scene, the probability of wrong tracking and missing tracking is reduced, and high-precision and high-real-time multi-target sensing and tracking are realized in the complex scene.

The invention belongs to the technical field of image processing, and particularly relates to a graphite ore image segmentation method based on an improved YOLO11-seg model, which adopts a balanced design of precision and light weight, optimizes redundancy in a network structure, introduces GSConv and C3k2-Faster modules, effectively reduces calculation overhead and memory occupation on the premise of keeping strong feature representation capability, and improves the image segmentation efficiency. The real-time reasoning efficiency of the model is obviously improved; a Mosai c data enhancement method is fully utilized, the data diversity is improved while the number of samples is increased, and the robustness and generalization ability of the model in a complex environment are enhanced; the trained optimization model can be deployed on intelligent edge equipment of graphite ores, supports real-time and accurate grade estimation on site, remarkably improves the mineral separation efficiency, and provides an efficient and reliable computer vision solution for intelligent mining scenes.

The invention discloses a domain generalization personnel re-identification method and system based on multi-modal fusion and structure perception enhancement. According to the method, a grey-scale map mode is innovatively introduced to extract biological characteristic information irrelevant to dressing, so that the identification limitation of a visible light image under the conditions of uniform shielding and severe illumination is effectively made up; meanwhile, a structure perception data enhancement strategy is provided, a key identification area is protected through a semantic segmentation technology, and damage to effective features is avoided while data diversity is improved; besides, by introducing targeted alignment loss, uniformity loss and intra-domain uniformity loss, the distribution characteristics of the feature space are directly optimized and regularized, and the generalization performance in an unknown substation scene is significantly improved. According to the method, superior detection performance can be realized under the condition of limited annotation data, the dependence on large-scale annotation data is effectively reduced, and the deployment cost is reduced.

The invention provides a multi-task space-time fusion Wi-Fi fingerprint positioning system aiming at the problems that Wi-Fi signals are easily interfered in a complex indoor environment and the generalization ability of a traditional method is weak. The system takes spatio-temporal feature fusion as a core, deep features are extracted from space and time dimensions, building classification, floor classification and coordinate regression targets are optimized through weight collaboration, and the generalization and discrimination performance of multi-scale positioning is remarkably improved. In order to further enhance the robustness of the model, channel attention, a generative adversarial network and a de-noising auto-encoder are introduced to enhance feature expression and data diversity. Experiments on a UJIIndoorLoc data set show that the method is superior to some traditional methods and existing neural network models in the aspects of building and floor recognition accuracy and average positioning error, and the effectiveness of the method in a complex indoor environment is verified.

A ship pipe fitting recognition system based on multi-source data enhancement and intelligent recommendation comprises a multi-source data enhancement module, a model training module, a recommendation decision module, a federated learning platform and a block chain evidence storage module, the system is of a'cloud + side 'double-layer architecture, the cloud is used for intelligent evolution, distributed knowledge is integrated through federated learning, and the block chain evidence storage module is used for storing the distributed knowledge. The global model is continuously updated and optimized, the edge end is used for edge real-time response, and efficient pipe fitting feature detection and pipe fitting number recommendation are carried out at the equipment end. According to the method, the detection precision and the non-standard pipe fitting recognition capability in complex illumination and shielding scenes can be remarkably improved, and the misjudgment rate is reduced; the decision-making problem of similar part hybrid scenes is solved, and the assembly efficiency and accuracy are improved; meanwhile, the data privacy security is ensured; the technical blank in the aspects of data diversity, algorithm robustness and industrial adaptability in the field of ship pipe fitting recognition is filled, the efficient requirement of a ship assembly scene is met, and technical support is provided for ship manufacturing intelligence.

The steam generator state real-time monitoring method based on artificial intelligence belongs to the field of artificial intelligence and comprises the following steps: S1, data acquisition and labeling; S2, sample generation is performed by using a quantum generative adversarial network based on random projection embedding to realize data expansion; S3, the expanded data is input into a feature extraction model to perform training of the feature extraction model, and a five-layer fully connected neural network is used for feature extraction; S4, the feature-extracted data is input into a feature dimension reduction model to perform training of the feature dimension reduction model, and a self-encoding neural network algorithm based on local preserving projection is used to realize feature dimension reduction; S5, the dimension-reduced data is input into a classifier to perform training of the classifier model; and S6, steam generator state recognition and monitoring are performed.The steam generator state real-time monitoring method based on artificial intelligence can solve the problems of insufficient sample quantity and lack of data diversity and enhances the robustness of the model when the model has noise or fuzzy classification boundary data.

The invention provides a training data segmentation method and an electronic device. The method includes the following steps. Training data including a plurality of training images is acquired. Acquiring a plurality of feature vectors of the plurality of training images by using an encoder of the pre-training model; the feature similarity between any two of the plurality of feature vectors is evaluated. And selecting a plurality of target training images from the plurality of training images according to the feature similarity associated with each feature vector. Dividing the training data into a training set and a verification set comprising a plurality of target training images, wherein the training set and the verification set are used for training a machine learning model; therefore, the data diversity of the training set can be improved, so that the generalization ability of a machine learning model is improved.

The invention relates to an unbalanced traffic accident data analysis method based on a mixed diffusion model. The analysis method comprises the following steps: S1, preprocessing original accident frequency data; s2, performing marking processing on the accident frequency data; s3, constructing and training a variational auto-encoder model, and obtaining low-dimensional hidden space embedding representation; s4, constructing and training a mixed feature diffusion model; s5, performing oversampling on accident frequency data by using the trained mixed feature diffusion model; s6, training a traffic accident frequency model based on the enhanced accident frequency data set; and S7, outputting a result by applying the SHAP value interpretation model, and identifying key factors influencing traffic safety. Compared with the prior art, the method has higher mixed variable modeling and generating capacity, data with higher quality can be generated on the basis of keeping the diversity of accident frequency data and the consistency of variable relations, and the effect of accident frequency modeling and analysis is remarkably improved.

The invention provides a training data generation method and device, electronic equipment and a storage medium. The method comprises the following steps: constructing a three-dimensional simulation environment constrained by a natural language navigation instruction; executing the natural language navigation instruction in a simulation environment corresponding to the three-dimensional simulation environment to verify the construction correctness of the three-dimensional simulation environment; if the navigation route is correct, verifying whether the navigation route described by the natural language navigation instruction is reachable in the three-dimensional simulation environment or not based on a navigation grid corresponding to a ground walkable area in the three-dimensional simulation environment; and if yes, generating training data of a visual language navigation task for the preset intelligent agent according to the natural language navigation instruction and the three-dimensional simulation environment. According to the method and the device, the diversity of the VLN training data and the consistency of the three-dimensional space can be considered, and the accessibility of the path corresponding to the navigation instruction in the three-dimensional environment is ensured while the data diversity is improved.

The invention provides a diversity signal phase deviation compensation method and system based on pilot frequency and data combination, and relates to the field of digital communication, and the method comprises the steps: calculating a difference value between two paths of diversity signal sequences received at a target symbol position through a receiving signal sequence model, and carrying out the average noise reduction of a difference result between all data symbols; estimating a phase offset result between two paths of diversity signals based on pilot frequency coherent noise reduction and a data diversity inter-symbol difference post-averaging algorithm, and calculating to obtain a parameter estimation linear signal-to-noise ratio; and carrying out noise reduction on a phase offset result estimated by an average algorithm after pilot frequency coherent noise reduction and data diversity inter-symbol difference by adopting a maximum ratio diversity to obtain an improved linear signal-to-noise ratio of phase offset parameter estimation. On the premise of not increasing pilot frequency overhead, the phase offset estimation precision can be obviously improved, finally, the decoding performance is obviously improved, the practical equipment performance competitiveness is improved, and meanwhile, the pilot frequency overhead can be reduced in waveform design by adopting the technology, so that the equipment performance is improved.

An adaptable deep learning method is provided that delivers sound hepatic lesion identification in NETs, while significantly reducing human effort for data annotation and improving model generalizability for PET image quantification. A region-guided GAN (RGGAN) model conducts image-to-image translation between list-mode simulated PET images and real-world clinical data, while preserving semantic content of interest, e.g., lesions. The RG-GAN model is integrated with a lesion detection model into an end-to-end, unified framework for joint-task learning, such that the two models can benefit from each other. The RG-GAN translates the list-mode simulated data into real world-style images, which appear to be drawn from the real clinical PET image dataset, and feeds the translated images into the lesion detection model for training. In order to deal with the limited diversity of list mode-simulated PET image data, a specific data augmentation module is incorporated into the unified framework to improve model training.

This invention provides a method for processing real-device data of an intelligent agent and related equipment. The method includes: acquiring a target action acquisition task and a target action type; in response to the target action type being a long-term task, acquiring key continuous operation segment data in the long-term task; acquiring failure scenario data and corresponding recovery operation data during the execution of the target action acquisition task; identifying and deleting invalid static frames from the key continuous operation segment data and recovery operation data using a pre-trained invalid frame recognition model to obtain cleaned data; evaluating the quality of the cleaned data based on a preset data quality assessment model to obtain a quality assessment score; and training the target intelligent agent using the cleaned data in response to the quality assessment score meeting a preset quality score threshold. This application can effectively reduce the redundancy of real-device data for intelligent agents and enrich data diversity, thereby improving the generalization ability of visual language action models.

This application discloses a method and apparatus for use in nodes for wireless communication and artificial intelligence. The node receives a first information block configured for training data collection for a first cell set and a second cell set; the maximum buffer size used by the node for the collection is a first size; the sizes of the training data collection for the first cell set and the second cell set are a first target size and a second target size, respectively; the first target size plus the second target size is greater than the first size; the node reports training data for the first cell set and the second cell set with reference to a first reference size and a second reference size, respectively; the second reference size is equal to the smaller of the candidate size and the second target size; the first reference size is equal to the first size minus the second reference size. This application improves data diversity.

The invention belongs to the technical field of sonar data processing, and particularly relates to a small sample sonar reverberation data enhancement and target detection method and system. According to the small sample sonar reverberation data enhancement and target detection method, spectrum sensing loss and multi-scale short-time Fourier transform feature constraints are introduced through WGAN-GP; the cosine similarity between the time-frequency domain distribution of the generated reverberation signal and real data is greater than or equal to 0.92, so that the problem of mode collapse of a traditional generative adversarial network model under a small sample is solved, and the data diversity is improved; a sequential structure and a state transition rule of a sonar detection data set are coded into a two-dimensional image through a Gramer angle field and a Markov transition field, the feature discrimination degree of a target and reverberation is improved in combination with a CBAM attention mechanism, the detection accuracy can be improved by a multi-modal convolutional network in a small sample data scene, the multi-modal convolutional network does not depend on a large sample data volume any more, and the detection efficiency is improved. And overfitting caused by less sample data is avoided.

This invention provides a method for building extraction from remote sensing images that integrates convolutional neural networks (CNNs) and Transformers, relating to the intersection of remote sensing image segmentation and computer vision technologies. This method involves creating a remote sensing image dataset from acquired remote sensing images, labeling each image individually, and dividing the dataset into training, validation, and test sets. The preprocessed remote sensing images are then preprocessed to increase data diversity. Feature extraction is performed on the preprocessed images to collect feature maps containing building information and obtain global features of the image. This method can significantly reduce false positives and false negatives for small target buildings, improve the completeness of segmentation for large target buildings, and mitigate boundary blurring caused by insufficient extraction of target building edge information.

The application provides a DAS signal generation method and device based on an autoencoder and a conditional diffusion model, and relates to the technical field of oil and gas pipeline monitoring. The method comprises the following steps: constructing a trained autoencoder; obtaining a threat event signal data set to be enhanced; inputting one-dimensional time sequence signal data in the data set into an encoding module of the autoencoder to obtain a two-dimensional signal feature map; inputting the two-dimensional signal feature map into a conditional diffusion model to obtain a generated two-dimensional signal feature map; inputting the generated two-dimensional signal feature map into a decoding module of the autoencoder to obtain new one-dimensional time sequence signal data, and constructing an enhanced threat event signal data set according to the new one-dimensional time sequence signal data. The data enhancement method provided by the application can provide more training samples for threat event identification, improve data diversity while keeping signal feature consistency, and significantly improve the classification accuracy and robustness of the identification model under the condition of few samples.

In the field of machine learning, there may be challenges associated with constructing a comprehensive and diverse training data set. For example, the data that is available may not be sufficiently diverse, which may cause issues such as overfitting in a model trained using the available data. A computer-implemented method and system are provided to use an embedding function as a tool in assessing the diversity of a data set. The embedding function may be employed in constructing a training data set having a high degree of data diversity for training a model.

The invention provides a human body posture data set construction method based on a large model. A large language model is combined with a fine-grained attribute database to generate motion description, then the motion description is converted into 3D human skeleton motion data through a MoMask model, and quality control is carried out. And then view control and gait speed control are performed on the motion data to increase data diversity. And then radar data simulation is carried out, including human body modeling, RCS estimation, radar signal generation and processing, and micro-Doppler spectrogram generation. Meanwhile, a VQ-VAE model is adopted to generate background noise and add the background noise, shielding is simulated, and the data trueness is enhanced. And finally, constructing a data set containing a plurality of activities. The data set generated by the method has semantic and physical diversity, a real scene can be simulated with high fidelity, the performance of a human body posture recognition model is effectively improved, and powerful support is provided for application of related technologies in multiple fields.

The invention relates to the technical field of artificial intelligence, and discloses a neural network optimization method for product recommendation, which comprises the following steps: step S01, fusing short-term behaviors of users to obtain historical session information, step S02, constructing a user session graph according to the historical session information of the users, step S03, constructing a user global graph according to the historical session information of the users, s04, splicing dual-channel features to generate user dynamic interest representation, S05, adding random noise enhancement features, and S06, deploying a lightweight reasoning model to generate recommendation content, by mining a potential interest transfer path of the user, the dynamic interest of the user is comprehensively described, the complexity and diversity of the interest of the user are better reflected, and the user experience is improved. The user interest understanding ability of a recommendation system is improved, random noise is added to increase data diversity, the problem of recommendation lag caused by long-term interest stability is avoided, and recommendation timeliness and accuracy are improved.

The application belongs to the technical field of sonar data processing, and particularly relates to a small-sample sonar reverberation data enhancement and target detection method and system. The small-sample sonar reverberation data enhancement and target detection method introduced the spectral perception loss and multi-scale short-time Fourier transform feature constraint through WGAN-GP, the cosine similarity of the time-frequency domain distribution of the reverberation signal generated by the method is greater than or equal to 0.92 with the real data, the mode collapse problem of the traditional generative adversarial network model under small samples is solved, and the data diversity is improved. The time sequence structure and state transition law of the sonar detection data set are coded into a two-dimensional image through the Gram angle field and the Markov transition field, and the CBAM attention mechanism is combined to improve the feature distinction degree of the target and the reverberation. The multi-modal convolution network can improve the detection accuracy under the small sample data scene, is no longer dependent on a large sample data volume, and avoids overfitting when the sample data is small.

This invention relates to a fault tracing and analysis method for garment production line equipment. Addressing the challenges of processing multi-source heterogeneous sensor data and the interpretability of causal models in garment production lines, it proposes a fault identification and causal attribution method based on operational condition semantics and disturbance analysis. This method includes multi-source sensor data normalization and operational condition semantic mask construction, generation of physically constrained causal disturbance samples, dual-channel feature encoding to decouple causal features, counterfactual reasoning to measure the contribution of the causal subspace to the diagnostic results, and finally, generation of a structured fault report containing a causal inference chain. Furthermore, it can dynamically optimize the model feature space partitioning based on new fault modes. This scheme achieves efficient alignment and utilization of data diversity and dynamism in complex industrial environments, improves the accuracy of fault diagnosis and the transparency of decision support, and significantly enhances the interpretability and adaptability of the model.

The invention provides a multi-modulation signal robust identification method based on a time-frequency diagram and data enhancement. Original baseband signals of the traditional modulation mode and the novel modulation mode are generated and preprocessed; performing data set division according to a double sampling mode; performing STFT on the data to generate a time-frequency graph; performing enhancement operations of time axis offset, frequency axis offset, noise injection and tiny rotation on the time-frequency graph of the training set; a ResNet-18 model adaptive to single channel input is constructed; and training the model and carrying out a performance test of unknown signal modulation mode identification. According to the method, the novel data set comprising a traditional modulation mode and an emerging modulation technology is constructed, the defect that most of the current modulation data sets are the traditional modulation mode is overcome, a data basis closer to a future communication scene is provided for modulation recognition in a complex scene, multi-dimensional enhancement is performed on data, data diversity is improved, and the method is suitable for being applied to modulation recognition in a complex scene. The model generalization ability is significantly improved, and the identification precision of the modulation signal is also effectively improved.

A data diversity augmentation method includes inputting a sentence into an encoder and extracting a feature vector for the sentence, inputting the extracted feature vector into an attribute classifier to form a decision boundary for the feature vector, and moving the extracted feature vector based on the decision boundary to generate a transformed feature vector and inputting the transformed feature vector into a decoder to restore a transformed sentence for the sentence.

The application provides a project recommendation method, device, system and storage medium, and belongs to the field of item recommendation. The method comprises the following steps: training original knowledge graph triples by using a TransH model to obtain a scoring function; performing parameter update analysis on the TransH model according to the scoring function to obtain an initial loss function and an updated TransH model; and performing initialization processing on the original knowledge graph triples by using the updated TransH model to obtain an initialized project node vector, an initialized user node vector and an initialized relationship vector. The application realizes more accurate, interpretable and diversified user preference recommendation, alleviates the conflict between data noise and data diversity, and improves the accuracy and robustness of recommendation.

The application belongs to the field of nuclear magnetic resonance sounding signal noise suppression methods, and is a ground magnetic resonance multi-type noise denoising network construction system and denoising method, which comprises a magnetic resonance signal construction module, a plurality of groups of magnetic resonance effective signals and environmental noise are simulated, and a noisy data set affected by three types of noise is constructed; a denoising neural network building module, the magnetic resonance signal generated by the magnetic resonance signal construction module is trained and optimized based on three types of noisy data sets and noise data sets, three denoising networks for different types of noise are obtained, a classification and discrimination model adopts a support vector machine method to judge the noise type according to the characteristics of different noises, and outputs the judgment result to the corresponding denoising network model in the denoising neural network building module. The application can reduce the scale of the label data amount, shorten the training time of the model, and remove the noise in a targeted manner, so that the dependence of the network model on data diversity can be reduced.