17results about How to "Protect private data" patented technology

The invention provides a meta-universe cross-scene anonymous seamless identity authentication method based on double block chains. The meta-universe cross-scene anonymous seamless identity authentication method comprises an initialization stage, a scene server registration stage, a user registration stage, a stage that a user logs in and creates an avatar in a scene server, and a user cross-scene server authentication stage. The server is set for different application scenes, the load of the server is reduced, and a user can experience various scenes without changing equipment. A fuzzy extractor and block chains are adopted to protect privacy information such as biological information and avatars of a user, meanwhile, a double-block-chain mode is adopted, authentication information of a server and the user is stored on an authentication data chain, the user moves in different services, an interaction data chain stores avatar data of the user in a multi-scene server, and the user experience is improved. And the data consistency is realized. The double block chains realize isolation of user real data and virtual data in the meta universe, privacy data of the user is effectively protected, the user does not need to input information again for login, and seamless authentication is realized.

The invention relates to the technical field of intelligent robots, and discloses a face recognition method of an intelligent robot, which comprises the following steps: step 1, lightweight face detection and feature extraction: performing face detection and feature extraction through a lightweight neural network structure, and combining model compression and edge calculation to reduce calculation complexity; and step 2, illumination and attitude adaptability is realized through multi-modal fusion, and face processing under different illumination and attitude conditions is carried out based on the extracted face features by utilizing fusion of an RGB image and an infrared image, adaptive illumination enhancement and 3D attitude correction. Through a multi-modal fusion technology, 3D attitude correction and self-adaptive illumination enhancement, different illumination conditions and attitude changes can be adapted, then the recognition stability of the system in a complex environment is remarkably improved, it is ensured that face recognition has high accuracy and robustness under the conditions of insufficient illumination, strong light or multi-angle offset, and the face recognition accuracy and robustness are improved. And the application requirements of the intelligent robot in multiple scenes are met.

The present disclosure relates to a data processing method, device, readable medium and electronic device to ensure the security of data and models in the data processing process. The method includes: receiving a data processing task; determining target data corresponding to the data processing task and a target first key corresponding to the target data, wherein the target data is obtained by encrypting the first key to the data to be processed, and the target first key is obtained by encrypting the first key by the first session key; decrypting the target first key by a target computing node according to the first session key to obtain the first key, and decrypting the target data according to the first key to obtain the data to be processed, wherein the target computing node runs in a trusted execution environment; determining a data processing result according to a target model and the data to be processed, wherein the target model is obtained by decrypting the encrypted target model based on the model key, and the model key is stored in a key management service, and the key management service runs in a trusted execution environment.

The invention relates to a commodity recommendation method and system, and belongs to the technical field of commodity recommendation. The commodity recommendation method comprises the following steps: collecting user information data and commodity information data, and desensitizing the user information data; constructing user features through the obtained data features to form a user portrait vector; extracting theme features of the commodity information by using a theme modeling algorithm through the commodity information data, and calculating the distribution condition of each theme feature; mining a commodity relationship between the commodity information data and the user behavior data, and establishing a time function of a change trend based on a complementary substitution relationship of commodities; and based on the user portrait, the commodity information data and the commodity relationship, establishing a recommendation score function of the recommended commodity, and generating a recommendation list of the commodity information according to recommendation scores. According to the method, matching is completed by respectively extracting the features of the user and the commodity, and the comprehensive recommendation score is given based on the relationship between the commodities, so that the accuracy and richness of commodity recommendation are improved.

The invention relates to the technical field of brain-computer interfaces, in particular to a self-adaptive calibration method and system for a brain-computer interface earphone, and the method comprises the following steps: obtaining electroencephalogram data; based on the electroencephalogram data, constructing an initial model through federal learning; selecting a user specific sample by using an active learning strategy; performing dynamic parameter optimization on the initial model; based on the result of the dynamic parameter optimization, adaptive personalized calibration is executed; performing anomaly detection and correction on the result of the self-adaptive personalized calibration; performance evaluation is carried out on the result obtained after anomaly detection and correction, the evaluation result is fed back to the previous steps, and the steps form a closed-loop optimization process; through the combination of federal learning and active learning, the calibration time is shortened from 45 minutes of a traditional method to about 5 minutes, and the efficiency is improved by nearly 9 times; the accuracy rate of the method reaches 95.8% and is improved by 8.6% compared with that of a traditional method.

The invention relates to the field of encryption search, and discloses an anonymous search system suitable for a large-scale multi-valued mapping database. The system comprises an original database used for storing a plurality of key-value pairs composed of keywords and corresponding values; the server is used for preprocessing all key value pairs in the original database to obtain a target matrix, generating grouping information and sending the grouping information to the client; the client is used for generating a query statement by adopting a Regev encryption algorithm according to a query keyword and grouping information submitted by a user and sending the query statement to the server; the server is also used for calculating a response vector according to the query statement and the target matrix and sending the response vector to the client; and the client is also used for calculating a query result according to the response vector by adopting a Regev decryption algorithm. According to the system provided by the invention, anonymous search of key fonts for the multi-valued mapping database can be realized, the calculation cost of anonymous search is reduced, and the response performance of the system is improved.

The invention discloses a longitudinal federated learning framework based on secret sharing and public key encryption, and belongs to the technical field of federated learning. The longitudinal federated learning framework is applied to federated training and contribution degree calculation, and participating entities are an aggregation server and two or more participants. According to the method, privacy data of the aggregation server and the participants are protected through secret sharing, and public key encryption is used on a transmission channel to enhance privacy protection.

The invention discloses a task execution method based on robust distillation, electronic equipment and a medium, and the method comprises the steps: inputting a clean data sample into a teacher model, and carrying out the detoxification to obtain a detoxification feature corresponding to each distillation layer; the first loss function is the variance of the detoxification feature corresponding to the last distillation layer of the teacher model; the second loss function is the distance between the detoxification feature corresponding to each distillation layer in the teacher model and the feature representation output by the corresponding distillation layer in the student model; the third loss function is cross entropy loss calculated based on the detoxication feature corresponding to the last distillation layer of the teacher model and the label corresponding to the clean data sample; taking the sum of the first, second and third loss functions as a total loss function; training a teacher model and a student model by using the total loss function; taking the trained student model as a task execution model; wherein the task execution model is used for executing a downstream target task.

The present invention relates to the technical field of vehicle data protection, and discloses a method and system for protecting vehicle privacy data based on collective signature. It is used when the management center and each node in the same collective communicate; the node includes a vehicle application node and a vehicle node, the vehicle application node collects the vehicle status information of the vehicle node, and the vehicle application node and the vehicle node request and receive messages; the management center is used to generate and maintain the collective public key and public private key. By constructing a collective public key and a public private key for a collective, when the vehicle application node and the vehicle node communicate, any node uses its own private key and public public key to perform a signing operation and send a signed message, and another node in the collective can only use the public public key without any information related to the signer to verify the signature, which fully protects the privacy data of the node while ensuring the credibility of the source of the message and the integrity of the information.

The invention relates to a student privacy data protection method based on artificial intelligence, and the method comprises the steps: configuring a plurality of databases which are used for storing student privacy data of different privacy levels, and each database is configured with a corresponding access level; obtaining student privacy data, and carrying out privacy grade division on the student privacy data; storing the divided student privacy data into a database of a corresponding privacy level; responding to an access request of a user, acquiring user information of the user, and judging whether the access permission of the user is higher than or equal to the access level of the accessed database; if yes, calling corresponding student privacy data; if not, refusing to access; according to the method, data are classified in an artificial intelligence mode, leakage of student privacy data possibly caused by manual data classification is avoided, and due to the fact that the artificial intelligence mode is adopted for data classification, the classification accuracy is higher than that of manual classification, and the classification efficiency is higher.

This invention provides an AI-powered companion robot control system and method based on mobile communication devices, relating to the field of intelligent robot control technology. It collects multi-dimensional behavioral data of users using mobile communication devices and extracts behavioral feature vectors. Based on the behavioral feature vectors, it matches them with historical behavioral patterns to generate clone commands containing predicted actions and their confidence levels, emotional state values, and their confidence levels. The robot itself can execute the predicted actions based on these clone commands, providing proactive reminders and companionship to the user. Furthermore, by dynamically evaluating the computing power contribution level using the operating status parameters of the mobile communication device, it performs differentiated task scheduling, achieving the coordinated utilization of three levels of computing resources. This protects user privacy data, reduces the hardware cost of the robot itself, and organically unifies the high intelligence of cloud-based large-scale models with low-latency, secure local control, improving resource utilization, saving computing power, and achieving a three-in-one collaborative system.

The invention provides a personalized federal learning method and system, and the method comprises the following steps: a client carries out the local training of a received global model, then uploads model parameters, and a central server carries out the dynamic weighting optimization updating according to all received model parameters; wherein the step of local training comprises the steps of learning global information under the guidance of local category embedding, and guiding learning of a personalized model through cooperative training of global and personalized features. The global information is introduced into the local training, and the global features and the personalized features are cooperatively trained, so that the accuracy of the personalized model is improved. Index tilt is used to flexibly adjust personal loss to introduce consideration of fairness, and it is ensured that in the model optimization process, features of individual clients do not dominate a global model.

Embodiments of the present application relate to the field of data security, and provide a face image processing method, related device and storage medium, the method comprising: a business server acquiring a target face image to be processed; inputting the target face image into a target model to obtain a target adversarial sample, wherein the target model is obtained based on unsupervised learning, and the similarity between the target adversarial sample and the target face image is lower than a preset threshold; publishing the target adversarial sample, or updating the target face image to the target adversarial sample. The present scheme can improve the generation efficiency of adversarial samples and protect user privacy data.

The invention discloses a microgrid privacy protection method and system based on a novel homomorphic encryption algorithm, and relates to the technical field of microgrid control. The method comprises the steps of designing a micro-grid distributed cooperative control method based on a distributed control theory, and realizing frequency recovery of each distributed power supply in a micro-grid; defining a privacy protection connotation and a privacy protection range; according to the defined privacy protection connotation and privacy protection range, a microgrid privacy protection method based on a novel homomorphic encryption algorithm is designed, an initial value and a real-time value of each distributed power supply in the microgrid are protected from being leaked to node neighbors, and the distributed cooperative control target of the microgrid is achieved; and verifying the microgrid privacy protection method based on the novel homomorphic encryption algorithm by using a simulation experiment. According to the invention, the credible perception capability and privacy security of distributed cooperative operation of the micro-grid can be improved.

The invention relates to the technical field of optimal scheduling of an integrated energy system, and discloses a fully-distributed combined heat and power optimal scheduling method, device and equipment and a medium. In order to solve the problems of linear convergence, low calculation efficiency, dependence on a coordination center and the like of a traditional distributed optimization algorithm in electric heating system scheduling, on the basis of reserving an original electric heating coupling model core architecture, a distributed collaborative interior point conjugate gradient method is introduced as an optimization algorithm, and fully-distributed scheduling with efficient convergence is achieved. According to the method, an electric heating comprehensive energy system optimization model is converted into a convex optimization problem, a correction equation is decoupled, a coefficient matrix is ensured to be symmetrical and positive definite, a distributed conjugate gradient method is adopted for solving, and collaborative optimization of an electric power system and a thermodynamic system can be achieved without a coordination center. Simulation verification shows that the calculation efficiency is improved by more than 90% on the premise of ensuring that the optimization precision is consistent with a centralized algorithm, the privacy of each energy main body can be protected, and the method is suitable for real-time scheduling of electric heating comprehensive systems of different scales.

The invention discloses a vehicle network interaction-oriented network traffic anomaly detection method, which belongs to the technical field of traffic anomaly detection, and comprises the following steps of: 1, collecting and preprocessing data; step 2, carrying out deep learning model training; step 3, constructing a federal learning framework; and step 4, real-time flow anomaly detection and alarm are carried out. The first step comprises a data source, a data collection process and data preprocessing; the data source is used for communication between a vehicle and the Internet, including uplink traffic, downlink traffic, data transmission of a vehicle internal network and external interaction data; the data transmission of the vehicle internal network comprises communication between CAN bus data and an ECU (Electronic Control Unit); the external interaction data comprises communication data with the drive test unit and interaction data with the traffic control system. According to the invention, deep learning and federated learning are combined, so that the abnormal traffic in vehicle network interaction can be efficiently identified in real time.

The invention relates to the field of service optimization, and provides a service optimization system and method for a service provider, and the method comprises the steps: obtaining data storage node information, and building a virtual storage model according to the data storage node information; acquiring source data, transmitting the source data to a server to obtain target storage data, dividing the target storage data to obtain block data, and setting a database; disorderly arranging the target storage data to obtain disordered storage data, and storing the disordered storage data into a data storage node; calculating data chaos indexes of the data storage nodes, and taking the data storage nodes of which the data chaos indexes do not meet a preset index as unsafe nodes; and enhancing chaos arrangement of the block data of the unsafe nodes to obtain enhanced chaos storage data, and taking the unsafe nodes corresponding to the enhanced chaos storage data as safe nodes. When the method is used, the cloud storage service of a cloud computing service provider can be optimized, and the risk of data leakage is effectively avoided.