17results about How to "Optimize network parameters" patented technology

The present invention relates to the field of smart grid technology, specifically a distribution terminal communication network system based on WAPI, the system includes a distribution data collection module, a distribution data compression module, a data encryption processing module, a network parameter adjustment module, a distribution anomaly identification module and a distribution scheduling adjustment module. In the present invention, by introducing real-time data processing and encryption technology based on WAPI, the security of data and the efficiency of the network are improved, the data encryption processing ensures the security of information during transmission, enhances the system's control ability over power network operation and monitoring, and adjusts the sending frequency of data packets according to the network status, dynamically optimizes network parameters, improves the network's adaptability and stability, and implements anomaly identification and fault response to ensure the continuity and safety of power grid operation. By optimizing the data processing process and strengthening safety measures, the overall performance and reliability of the power system are significantly improved.

The embodiment of the invention provides a groundwater level monitoring method and system based on artificial intelligence, and the method comprises the steps: constructing a first groundwater level prediction network which comprises a focusing node and a transmission node, and generating a plurality of second groundwater level prediction networks based on the incidence relation of different nodes; and the diversity and accuracy of underground water level prediction are obviously improved. And linkage parameter learning is performed on the plurality of second groundwater level prediction networks by using the groundwater dynamic monitoring data of the basic sample, so that the network parameters are effectively optimized, and the prediction precision is further improved. Meanwhile, more abundant and comprehensive target sample groundwater dynamic monitoring data are generated by performing water level prediction learning sample extension on the basic sample data, and data support is provided for training of a target groundwater level prediction network. The finally generated target groundwater level prediction network is excellent in performance in practical application, and an accurate and reliable decision basis can be provided for groundwater management and water resource protection.

This invention relates to the field of digital image security technology, and in particular provides a robust feature embedding image watermarking method based on noise-aware guidance. The method includes acquiring a carrier image dataset, constructing a robust feature embedding image watermarking network based on noise-aware guidance, the robust feature embedding image watermarking network including a watermark embedding architecture and a watermark extraction architecture; inputting the carrier image into the watermark embedding architecture to obtain a watermarked image; inputting the watermarked image into the watermark extraction architecture to obtain watermark information; optimizing network parameters, and simultaneously using a discriminator as an adversarial network, utilizing adversarial loss to ensure that the carrier image and the watermarked image are similar. This method improves robustness against attacks while generating high-quality watermarked images.

The application discloses a kind of underground image monitoring method and equipment based on unsupervised learning network, it is related to mineral exploitation underground image monitoring technical field, the method comprises: the monitoring model corresponding to target monitoring task is target monitoring model;Target monitoring task scene data set is obtained;Based on target monitoring task, unsupervised learning network and the loss function of unsupervised learning network are constructed;Using target monitoring task scene data set, loss function and target monitoring model, unsupervised learning network is trained;Using target monitoring model and the unsupervised learning network trained, executes target monitoring task.The application improves the accuracy of underground image monitoring by constructing and training unsupervised learning network.

The embodiment of the invention provides an Internet information pushing method and system based on big data service, and the method comprises the steps: obtaining an Internet interaction data sequence of a basic template and a business derivation template from the big data service, and carrying out the parameter learning and optimization of an Internet interest prediction network through the data. And finally, a target Internet interest prediction network is generated, so that the accuracy of interest point prediction is improved, and accurate grasp of personalized demands of the user is realized. By generating and fusing various interactive knowledge graph features, the network parameters are further optimized, so that the pushed information better meets the actual interests and requirements of the user, and the Internet information pushing efficiency and the user satisfaction are remarkably improved.

The present invention discloses a prediction reconstruction video anomaly detection method integrating latent space autoregression, which belongs to the field of surveillance video processing, and solves the video anomaly misdetection and missed detection problems caused by the diversity of video information and the scarcity of abnormal events in traditional methods; including firstly performing data preprocessing on the input video sequence by random cropping, random horizontal rotation and standardization operation; defining prediction network constraints, generating prediction frames that meet the constraints; then constructing a reconstruction network in combination with a UNet network and a latent space autoregressive layer, generating a reconstruction frame that meets the constraints; then constructing a discriminator by stacking multiple convolutional layers to discriminate the abnormal values ​​of the input frame; finally adopting a generative adversarial algorithm, training the prediction generation network, the reconstruction network and the discriminator, and optimizing the network parameters. The recognition accuracy obtained by the present invention is relatively high, and the recognition speed is relatively fast, which can meet the real-time requirements of video anomaly detection in large-scale surveillance scenarios.

This invention discloses a dynamically densely connected spatiotemporal feature decoupling network for recognizing cross-view gait, relating to the field of computer vision. It includes: an initial feature processing module, a dynamically dense spatiotemporal decoupling feature extraction module, and a feature enhancement processing module. The module uses dense spatiotemporal feature decoupling blocks and concatenation operations to achieve the sharing of shallow and deep network features, thereby addressing the problem of insufficient representation ability. Simultaneously, it employs an enhanced convolutional block attention mechanism to allow the network to focus on more important gait features. Finally, the feature enhancement processing module processes the five-dimensional feature mapping into multiple lateral features and performs batch standardization of the features to enhance representation ability and model generalization ability. This allows for the mining of correlations between shallow and deep features, alleviating the problem of insufficient information representation ability.

The invention relates to an underwater mechanical arm control method for optimizing an RBF (Radial Basis Function) neural network through an improved Harris Hawk algorithm (PTHHO). The method comprises the following steps: firstly, constructing an RBF (Radial Basis Function) network which comprises an input layer, a hidden layer and an output layer, and then generating a group of initial solutions (weight and bias) by using a PTHHO algorithm; and then, by simulating hunting behaviors of eagles and iteratively updating and optimizing the parameters, the loss function is minimized, and the approximation precision is improved. The combination utilizes the global search capability of the PTHHO algorithm and the fitting capability of the RBF neural network, so that the performance and generalization capability of the neural network can be improved to a certain extent. Compared with a traditional control method, the method has the advantages that the robustness of a control system is improved, hydrodynamic interference is effectively compensated, and the purpose of accurately controlling the underwater mechanical arm is achieved.

The present invention provides a lung nodule recognition model training method and device, and a lung nodule recognition device. The method is applied to the field of image processing. The method comprises: using a lung nodule recognition model to recognize a lung image, the lung nodule recognition model comprises a backbone network module, a dual-feature pyramid module and an output head module, the backbone network module is used to extract a multi-level feature map from an input lung image, the dual-feature pyramid module is used to fuse the multi-level feature maps, the output head module is used to classify and regress the multi-scale feature map generated by the dual-feature pyramid module to obtain classification results and regression results, and calculate the loss function to obtain classification loss and regression loss; weighted sum the classification loss and regression loss to obtain the total loss, and use the back propagation algorithm to calculate the gradient of the total loss; and update the network parameters of the lung nodule recognition model according to the gradient. The present invention improves the detection and positioning accuracy of lung nodules.

The application discloses a pancreas disease multi-time window risk prediction system based on routine physical examination queue data and relates to the technical field of medical big data and disease risk prediction. The system comprises data acquisition, preprocessing, feature fusion, risk calculation and model optimization modules; the acquisition module acquires and integrates data to output a structured original data set; the preprocessing module processes the original data set into a personal longitudinal time sequence feature vector set; the feature fusion module outputs a unified feature vector; the risk calculation module obtains a multi-time window disease probability; and the model optimization module constructs a label, calculates a loss and optimizes and updates parameters of each module. The application integrates routine physical examination and pancreas disease outcome data to construct a data set, mines data value through standardization preprocessing and the like, inputs a unified feature vector into a multi-layer perception machine, constructs an incremental risk vector, realizes multi-time window disease probability stratified prediction, breaks through single limitation, introduces L2 regularity to improve performance, and helps early screening and intervention of pancreas disease based on routine physical examination data.

The application discloses a human pose reconstruction method based on a distance-guided double-branch network. The purpose is to solve the problems of the loss of joint information of slow movement or almost static, the lack of unified range prior between different representations, and the difficulty in effectively combining human motion saliency and structural integrity. The method comprises the following steps: acquiring a multi-channel millimeter wave radar echo signal; estimating human distance prior based on absolute median difference; constructing distance-guided point cloud features and distance-guided heat map features; and based on a double-branch network of point cloud and heat map, and using a cross-modal attention mechanism to fuse the two representations. The application improves the accuracy and robustness of human pose reconstruction of the millimeter wave radar, effectively reduces the loss of human information caused by filtering, and is suitable for scenes such as smart elderly care and smart home.

The present invention relates to a lip reading recognition method and system based on staged cross-training. The method comprises: obtaining encoding network parameters and decoding network parameters in the first stage based on an end-to-end training method; in the second stage, training a model according to single identity data, generating initialization parameters of an encoding network and a decoding network with overfitting, adaptively generating weighted regularization constraints of a decoding network using the encoding network initialization parameters, and adaptively generating weighted regularization constraints of a decoding network using the decoding network initialization parameters. The method of the present invention ensures the independence of the encoding network and the decoding network by such cross-training method, thereby improving the generalization ability of the lip reading recognition network, and further, when the lip reading recognition model trained by the present invention is applied to lip reading recognition, the robustness and accuracy of the lip reading recognition of unknown speakers can be improved.

The invention discloses a privacy protection image feature compression method and a privacy protection image feature compression system for edge cloud collaborative reasoning, which are based on a novel privacy protection feature compression framework, are combined with an attention mechanism, and guide the compression of intermediate features through an adversarial collaborative training strategy in a feature domain set. Selective aggregation and feature disturbance suppression of a feature domain are introduced into a feature domain focused confrontation cooperative training strategy, the security and anti-attack ability of data in the transmission process are enhanced, and the risk that data privacy is leaked on a cloud honest but curious server is effectively reduced. In order to improve the generalization ability and adaptability of the algorithm, system evaluation is carried out on a plurality of image data sets and different deep learning models. Experimental results show that the method is remarkably superior to an existing feature compression method in the aspects of intermediate feature transmission efficiency, analysis precision and privacy protection level, and powerful technical support is provided for application of a large-scale deep learning model in a resource-limited edge device environment.

The present invention discloses a method and system for constructing a deep network model based on difficult sample mining. The deep network model is constructed and training samples are input to train the deep network model. The feature vector of the training sample is extracted during each iterative training process, the angular boundary margin between the feature vector and its class center weight vector is calculated, and whether the feature vector is a difficult sample within the class is judged according to the angular boundary margin, and the feature vector is marked as a difficult sample according to the judgment result. The angular boundary margin loss value and the difficult sample loss value within the class output by the deep network model are calculated based on the marked feature vector, and the network parameters of the deep network model are optimized by combining the angular boundary margin loss value and the difficult sample loss value within the class. The present invention takes into account the influence of difficult samples within the class on the model accuracy, and supervises the training process based on the angular boundary margin loss and the difficult sample loss within the class, which can improve the compactness of the distribution of features within the class, thereby improving the discrimination of the model.

The invention discloses a training method of a gesture motion translation model, a gesture motion translation method and device, setting and a storage medium. The method comprises the following steps: when a target person performs a gesture action, acquiring video data corresponding to the target person and target information corresponding to the gesture action; performing at least one action type identification on the video data to obtain first feature information; performing gesture motion prediction on the gesture motion based on the target information to obtain second feature information; inputting the first feature information and the second feature information into the gesture motion translation model to obtain a loss function corresponding to the gesture motion translation model; and updating parameters corresponding to the gesture motion translation model based on the loss function.

The present invention relates to the field of artificial intelligence technology, and in particular to a method for post-disaster restorative maintenance of cluster systems based on the AC-MCTS algorithm. The present invention improves the reliability and maintainability of cluster systems by real-time prediction of system status and dynamic selection of optimal maintenance strategies. The system can not only effectively solve various faults in cluster systems, but also improve the intelligence level of maintenance strategies, reduce system downtime, and improve the overall operating efficiency of the system through continuous learning and optimization.

The invention provides an image super-resolution method based on a novel network convolution architecture, and aims to improve the spatial resolution and detail recovery capability of an image. According to the method, a super-resolution generative adversarial network is taken as a baseline model, and the network is improved and optimized by using a novel network convolution architecture. In the improved network, batch normalization in the original network is replaced by a new layer normalization module, so that the network can perform independent normalization on a single sample, the dependence on statistical information of the whole batch is avoided, and the feature extraction capability is enhanced. In addition, a new module is introduced in the method, the structures of the generator and the discriminator are further optimized, and the performance of the network during high-resolution image generation is improved. According to the method, the paired image data sets are adopted for training, the model obtained after improved network training can effectively reconstruct the high-quality image, the definition and detail performance of the image are remarkably improved, and good adaptability and wide application prospects are shown.