38results about How to "Improve privacy and security" patented technology

The invention discloses a federal learning data processing method based on privacy protection, and the method comprises the following steps: S1, a client uses local data to train a generator of a generative adversarial network, and generates virtual data similar to the local data in distribution; s2, setting three privacy protection levels according to the privacy sensitivity of the local data; s3, performing model updating operation by using a homomorphic encryption technology; s4, utilizing a dynamic weight optimization mechanism based on a multi-arm bandit algorithm to obtain a priority processing data set; s5, preliminarily updating the global model, analyzing generated data through a reverse inference detection mechanism, and judging whether a privacy disclosure risk exists or not; s6, the server feeds back the updated global model to the client, and the client continues personalized model training in combination with local data; and S7, generating new virtual data, and carrying out the next round of global model training. According to the method, the generative adversarial network, differential privacy and homomorphic encryption are combined, so that efficient federated learning under privacy protection is realized.

The invention relates to a privacy-protected distributed acceleration resource allocation method and system, and belongs to the technical field of information communication. According to the method and the system, a parameterized distributed momentum method and a push-pull distributed resource allocation algorithm are combined to realize accelerated resource allocation in a directed network; a privacy-related variable is decomposed into two variables through a state decomposition scheme, one variable is used for information interaction, and the other variable directly related to privacy is used for local updating, so that privacy is protected, and noise disturbance is introduced in communication by adopting a differential privacy strategy to further enhance privacy protection. The algorithm disclosed by the invention is specially designed for a complex network scene, is suitable for applications needing privacy protection and efficient resource allocation, such as a smart power grid and energy management, shows wide applicability and innovation in distributed optimization, and has a wide application prospect.

The invention relates to a virtual power plant-oriented power reserve market privacy protection transaction method, which comprises the following steps that: a quotation party performs integer modification on sensitive power data such as declaration price and declaration quantity in T time periods of the next day, and performs encryption protection by adopting a PaillierTD partial homomorphic encryption algorithm to obtain privacy quotation data; the power data is uploaded to a transaction block chain comprising n auxiliary nodes through the power data storage and calculation all-in-one machine for solidification, and the credible problem of the uplink data is solved from the source; the auxiliary node accumulates the privacy declaration amount data with the same time period and the same price in the privacy quotation data, and sends the accumulated privacy declaration amount data to the power dispatching control center; and the power dispatching control center obtains the sum of the electric quantity declared by each quoting party at the same time period and the same price, performs sorting according to a transaction rule, and determines the clearing price and the bid winning quantity of each time period. According to the invention, high privacy security requirements can be met.

The invention relates to the technical field of block chains, and discloses a block chain data encryption query method, device, equipment and medium, the method comprises the following steps: extracting data content information of to-be-chained data, using a homomorphic encryption algorithm to encrypt the data content information to obtain ciphertext content information; taking the ciphertext content information and the corresponding ciphertext hash value as on-chain evidence storage data to be stored in a block chain in a chaining manner; performing association mapping on the on-chain evidence storage data and the ciphertext access rule through the smart contract to generate an authority association rule corresponding to the on-chain evidence storage data; performing permission verification on the access permission in the data query request based on a permission association rule; after the permission verification is passed, querying the on-chain evidence storage data in a homomorphic encryption state according to the data query request to obtain an encrypted query result; and decrypting the encrypted query result to obtain target plaintext query data. The reliability of data encryption query can be improved.

The invention belongs to the technical field of privacy protection, and discloses a false position privacy protection method and system based on moving path semantic relevance, which utilizes a neural network to learn spatial-temporal features and semantic features among positions in a real moving path of a user, and combines transition probability among position sequences to obtain a false position privacy protection result. The influence of query time on spatial-temporal features and semantic features among positions is quantified, and a false position privacy protection scheme based on moving path semantic relevance is provided. Theoretical analysis and a large number of experiments show that the scheme can efficiently help a user to generate a false position sequence which cannot be distinguished from a real position sequence in terms of spatial-temporal features and semantic features, and effectively protect the position privacy of a requesting user.

The invention discloses an image processing method and device, electronic equipment and a medium, and belongs to the technical field of image processing. The method comprises the following steps: receiving an encrypted image corresponding to an original image and encrypted information, wherein the encrypted information is a transformation relation adopted for transforming the original image into the encrypted image; and inputting the encrypted image into an image processing model to perform image feature data reduction operation on the encrypted image based on the encrypted information to obtain image feature data corresponding to the encrypted image, the image feature data corresponding to the encrypted image being the same as the image feature data corresponding to the original image.

The application relates to a video compression communication method based on video space-time semantic decoupling and generative reconstruction. Frame preprocessing is performed on a to-be-transmitted video to generate a video frame sequence; semantic decoupling is performed on the video frame sequence to extract time features and spatial anchor points; the time features are input into a video description model to generate a text description, and the text description is subjected to semantic arrangement and lossless compression to output time semantic loads; key frames are extracted according to a key frame selection strategy and subjected to image semantic encoding to output spatial semantic loads; the time semantic loads are sent to a receiving end via a reliable communication path; the spatial semantic loads are sent to the receiving end via a semantic communication path; the receiving end performs semantic decoding to recover time semantic information and spatial anchor point information, combines receiving end preference information to construct a generation condition; and the generation condition is input into a generative video model to perform video reconstruction, and a target video with consistent semantics with the to-be-transmitted video is output. The method can significantly reduce the video communication data volume and transmission delay and enhance privacy protection.

The invention provides a model training method and device, an advertisement putting processing method and device, a computer program product and electronic equipment, and relates to the technical field of computers. The model training method comprises the steps of performing anonymization processing on an original data set to obtain an anonymous data set, and determining an initial feature set according to the anonymous data set; evaluating the importance of the features in the initial feature set to obtain an importance evaluation result; random noise is added to the importance evaluation result, and a feature subset is determined from the initial feature set according to the importance evaluation result added with the random noise; and training the advertisement putting model based on the feature subset to obtain a trained advertisement putting model. According to the invention, the accuracy of the advertisement putting model can be improved under the condition of ensuring the safety of user data.

The invention relates to a privacy protection technology, and discloses a distance query method based on local differential privacy, which comprises the following steps: acquiring a social network graph, and initializing a distance vector of each client in the social network graph; the client adds a noise value to the distance vector, distributes the distance vector to a neighbor client, and updates the distance vector of each client after aggregation operation; and repeating the steps of adding noise to the distance vector of the client, distributing the distance vector to the neighbor client and updating the distance vector of each client through aggregation operation until a repeat stop condition is met, thereby obtaining a distance query result. The invention further provides a local differential privacy-based distance query device, equipment and a medium. According to the invention, the security and efficiency of distance query based on local differential privacy can be improved.

The invention relates to the field of federated learning, in particular to an aircraft manufacturing equipment abnormity diagnosis method based on federated learning. According to the method, firstly, a federal learning training framework is adopted for training of an aircraft manufacturing equipment anomaly detection model, the privacy security of data is guaranteed, and the privacy security of the aircraft manufacturing equipment data is further enhanced. On the basis of guaranteeing the data privacy security, the model for detecting the abnormality of the aircraft manufacturing equipment is trained, and model training is carried out on the data of the aircraft manufacturing equipment under the condition that the data are not out of local equipment through a federal learning architecture, so that the privacy security of the data is guaranteed. According to the method, a contract theory incentive mechanism is adopted, reasonable reward is provided for the aircraft manufacturing equipment to participate in the federated learning training process, the aircraft manufacturing equipment is incentive to participate in the federated learning training, and the accuracy of the aircraft manufacturing equipment anomaly detection model is improved.

The invention discloses a distributed compression random aggregation algorithm for robust federal learning under Byzantine attack, and aims at potential loopholes after Byzantine attack on a large-scale distributed cluster and the influence of multi-agent information communication on communication bandwidth and computing resources. An algorithm for resisting attacks and reducing communication loss is provided, and robust stability of the system, efficient communication and privacy security of the client are realized. According to the algorithm, a loss function and a penalty term are combined, and elasticity to Byzantine attacks is kept; a parameter compression operator is introduced, so that the communication cost is reduced; weight and feature dimension reduction is quantized, the data dimension or size is reduced, and algorithm convergence is accelerated; and the uploaded and issued parameters are subjected to lossy compression, so that data encryption is realized, and privacy and security are ensured. The method has the advantages of being high in anti-interference capacity, low in communication cost and high in privacy safety, can be widely applied to multiple fields, and is particularly suitable for application systems with high data privacy requirements and limited communication resources.

The invention provides a business processing method and device, equipment, a medium and a program product, which can be applied to the field of artificial intelligence and the field of financial science and technology. The method comprises the following steps: detecting the position of a customer to obtain a first position of the customer; the directional sound wave generator is controlled to send first directional sound waves carrying service handling notification information to the first position, the first directional sound waves comprise first ultrasonic waves of at least two phases, the first ultrasonic waves of the at least two phases are overlapped in a first area corresponding to the first position, and a sound signal which can be heard by a user is generated; the business handling notification information comprises a second position for handling the target business for the client; when it is detected that the customer reaches the second position, the directional sound wave generator is controlled to send second directional sound waves carrying the business processing information to a second area corresponding to the second position, and the second directional sound waves are used for indicating the customer to execute the target business handling process.

The application discloses a double-head action recognition and prediction method and system based on skeleton sequence and space-time speed feature fusion, relates to the technical field of computer vision, and comprises the following steps: acquiring video data; extracting the spatial position information of each human body skeleton key point on continuous frames from the video data to construct a human body skeleton key point sequence; calculating the motion change information of adjacent frames based on the human body skeleton key point sequence, and fusing the spatial position information of the skeleton key point to construct space-time-dynamics fusion features; and taking the space-time-dynamics fusion features as input, identifying the current action category and predicting the future action category based on a time series modeling network model. The application supports the identification and early warning of multiple abnormal behavior types, and can adapt to different application scenarios.

This application relates to a transmission path planning method and apparatus for data sharing and exchange in a controlled network environment, belonging to the field of network communication and data transmission technology. The method includes: calculating the comprehensive adaptability score of the service links of the controlled network subdomains to the current service based on a security-performance dynamic coupling model, and using this score as the service link coupling feature; aggregating the service link coupling features of all subdomains to generate a global feature vector and inputting it into a policy generation model, outputting a basic path policy for cross-domain data sharing and exchange, and storing it in a historical policy database; when services are updated or new subdomains are added, performing path migration adaptation between the new service features or the service link coupling features of the new subdomains and the basic path policy, outputting a target path policy and comparing it with the historical best value for similarity, triggering policy fine-tuning based on the comparison result to obtain the optimal path policy. This method can achieve efficient, secure, and flexible data sharing and exchange in a controlled network environment.

This invention discloses a multi-dimensional privacy information perception and collection method. It constructs a formal description model of multi-dimensional privacy information, characterizing privacy information from multiple perspectives, including its composition, attributes, permissions, and control. Based on active learning and transfer learning algorithms, it builds a metadata intelligent perception, identification, and labeling method adaptable to various application scenarios and privacy protection strategies. This patent's advantages in adapting to multiple scenarios and information types allow it to adapt to the impact of changing application scenarios on privacy properties, reduce the concealment of privacy information caused by multiple information types, and improve the accuracy of privacy information perception. It focuses on perceiving multiple types of privacy information, comprehensively perceiving different privacy types such as text, images, and sound. Compared to existing single-type privacy perception methods, it is more beneficial for subsequent privacy strategy formulation and privacy control implementation, improving user privacy security. It has strong privacy perception capabilities, can process different types of data in parallel, and offers accurate privacy positioning and fast data processing.

The invention discloses a message sending method, and belongs to the technical field of communication. The method comprises the following steps: acquiring a user sending a to-be-sent message, a to-be-sent object of the to-be-sent message, and a historical chatting record of the user and the to-be-sent object; obtaining a sensitivity score of the to-be-sent message according to the degree of familiarity between the user and the to-be-sent object and the matching degree between the to-be-sent message and the historical chat record of the to-be-sent object; and determining whether to send the to-be-sent message to the to-be-sent object based on the sensitivity score. Through the method, sensitive messages can be prevented from being sent by mistake.

The application provides a front-end reasoning deployment method and system based on a TensorFlow lightweight model, comprising: in a client, creating a background thread independent of a main thread; in the background thread, performing a model loading step, the model loading step comprising: judging whether a new version of a deep learning model needs to be obtained from a server through a local cache verification mechanism, and obtaining or using the deep learning model cached locally according to the judgment result; the main thread and the background thread interact through an asynchronous communication mechanism to perform reasoning calculation based on the deep learning model in the background thread, and send the reasoning result from the background thread to the main thread. The application realizes low-delay real-time reasoning: the reasoning process is completely completed in the local client of the user, and data does not need to be uploaded to the server, eliminating the delay caused by network transmission and greatly improving the real-time performance of artificial intelligence functions.

This invention discloses a multimodal intelligent interactive portable terminal, relating to the field of portable intelligent terminal technology. It includes a hardware security sensing unit, a buttonless voice interaction unit, a covert alarm unit, a tiered alarm unit, a local AI processing unit, a peripheral expansion unit, and an optional hardware isolation unit. It integrates security monitoring, voice interaction, covert alarm, local intelligent assistance, and multi-device linkage functions, supporting offline local operation and hardware-level privacy protection. It features strong alarm concealment and high interactivity. This invention is comprehensive in function, highly secure, and highly portable, suitable for various scenarios such as personal daily life, business communication, and security protection, meeting users' diverse intelligent interaction and security protection needs.

The present invention discloses a method and device for encrypting battery data based on blockchain, the method comprising: obtaining battery parameters corresponding to the current block in the blockchain; determining the battery hash value of the current block according to the battery parameters and a first preset encryption algorithm; generating an anti-counterfeiting data table of the current block according to the battery hash value of the current block, the timestamp of the current block and the encrypted business data of a predetermined target block, wherein the target block is the previous adjacent block of the current block in the blockchain. It can be seen that the implementation of the present invention can determine the battery hash value of the current block according to the battery parameters of the current block and the first preset encryption algorithm, thereby performing secondary encryption on the battery hash value according to the encrypted business data of the target block and the timestamp of the current block, and then generating an anti-counterfeiting data table of the current block, which can reduce the difficulty of data encryption and improve the degree of data privacy security.

The embodiment of the invention provides a data processing method and device, a networking architecture and a computer readable storage medium. The method comprises the following steps: receiving uplink data; the uplink data comprises type identification information; classifying the uplink data according to the type identification information; and under the condition that the uplink data is classified into the sensed object data, the sensed object data is shunted to a preset local shunting device so as to perform data processing on the sensed object data through the local shunting device, and the sensed object data after data processing is sent to a second local server. According to the embodiment of the invention, the physical link is greatly shortened, the time delay is reduced, and the data transmission efficiency is improved.

The application belongs to the technical field of Internet, and particularly relates to an industrial Internet cross-domain metadata trusted fusion method and system, comprising the following steps: collecting structured metadata, time series metadata and relationship metadata of each industrial domain node, and constructing an intra-domain metadata graph based on field attributes and call relationships; performing semantic coding and time series coding on the intra-domain metadata graph, obtaining corresponding semantic representations and time series representations, and calculating inter-field consistency; identifying stable field clusters, drift field clusters and abnormal nodes according to the inter-field consistency in each industrial domain node; and performing segmented aggregation on the stable field clusters and the drift field clusters by each industrial domain node to generate anchor field summaries and drift trajectory summaries. The application combines metadata graph coding, field clustering and federal dynamic weighting, and realizes industrial cross-domain metadata private trusted fusion through abnormal compensation correction.

The application provides a privacy protection density clustering method, an electronic device and a storage medium. The method comprises the following steps: sending a plurality of ciphertext data to two cloud servers, so that each cloud server determines the object type of each received ciphertext data, the object type comprising a core object or a boundary object, and based on a privacy protection DBSCAN algorithm, each core object and a target boundary object in the neighborhood range of the core object are grouped into a temporary cluster, the target boundary object only belongs to one temporary cluster, the connection relationship between each temporary cluster and other temporary clusters is determined, and the connection relationship is fed back; according to the connection relationship fed back by the two cloud servers, the connected temporary clusters are merged, and a clustering result is obtained. The application can improve privacy security, and improve clustering efficiency and clustering quality.

The utility model provides a quilt, and relates to the technical field of smart home. The plurality of temperature sensors are arranged at a plurality of preset positions of the quilt core at intervals, and the plurality of temperature sensors are used for monitoring temperature signals of a plurality of body parts of a human body respectively; and the output interface is electrically connected with the plurality of temperature sensors, and the output interface is used for outputting the plurality of temperature signals. The multiple temperature sensors are arranged at the multiple preset positions of the quilt core at intervals so as to monitor the temperature signals of the multiple body parts of the human body, the output interface is arranged, the output interface is electrically connected with the multiple temperature sensors so as to output the multiple temperature signals, and therefore the temperature of the human body can be monitored. Real-time collection of the body surface temperature distribution characteristics of the user can be achieved in a non-contact temperature monitoring mode, and therefore the accuracy and privacy safety of sleep behavior monitoring are improved.

The intelligent beauty makeup mirror comprises a bottom shell and a surface cover, an opening is formed in the upper end of the bottom shell, the surface cover covers the opening of the bottom shell, a mirror surface is formed in at least the middle portion of the surface cover, and an observation window and red light beads are arranged on the bottom shell; and the part between the red light lamp bead and the observation window is set as follows: when the red light lamp bead emits red light to the outside of the bottom shell, the red light can pass through the observation window after being reflected by an external camera and can be observed through the mirror surface. Therefore, on the basis of the beauty makeup mirror, a camera detection function is added, and whether a camera exists in a use environment or not is detected through reflection and transmission of red light, so that privacy security is improved. The function of the detector is combined to the beauty mirror which is generally carried by women during travel, so that the use function of the beauty mirror is increased, the carried articles are not increased, and the beauty mirror is convenient to travel and play.

The application discloses a kind of distribution network multi-objective distributed optimization scheduling method based on ADMM algorithm, belong to distribution network optimization control field, consider three indexes of distribution network overall operation cost, multi-zone load rate difference and photovoltaic accommodation rate comprehensively, each index is weighted using entropy weight method, to realize the effect of distributed optimization, using ADMM algorithm, overall optimization model is split into upper and lower two layers and is studied.Minimize distribution network power purchase cost and multi-zone load rate difference as target in upper layer, minimize local energy storage operation cost and light abandonment in lower layer as target.A small amount of information is exchanged between upper and lower two layers, and each is optimized, and the optimal output of distributed photovoltaic and energy storage is obtained by iteration until the algorithm converges.The application has the advantages of scientific and reasonable method, strong applicability, good effect and the like.

The invention provides an unmanned aerial vehicle high-altitude autonomous homeward voyage obstacle avoidance method and system applied to a mobile airport, and aims to solve the problem that obstacles possibly exist in a complex and changeable homeward voyage route of the mobile airport, whether the obstacles exist in front or not can be remotely judged in advance, and if the obstacles exist, an obstacle avoidance strategy needs to be autonomously selected. The obstacle avoidance strategy of the unmanned aerial vehicle during high-altitude autonomous return flight is realized, and the safety of the unmanned aerial vehicle is ensured. The high-altitude autonomous return flight obstacle avoidance of the unmanned aerial vehicle refers to that the unmanned aerial vehicle needs to return to a new coordinate position after the mobile airport moves to a new return flight point during mobile operation of the mobile airport. In the course of returning of the unmanned aerial vehicle, safe returning can be comprehensively judged according to information such as the field environment, an obstacle avoidance strategy is given, and high-altitude autonomous returning of the unmanned aerial vehicle is achieved.

The invention belongs to the technical field of Internet of Vehicles federated learning, and particularly relates to an alliance-based Internet of Vehicles federated learning method, system and device. The method comprises the following steps: firstly, converting convergence local model parameters uploaded by each vehicle into proportion vectors representing local data category distribution by using a small-scale balance auxiliary data set at a base station side; the distribution similarity among the vehicles is measured based on the divergence, the vehicles with similar distribution are clustered to form a plurality of alliances in a self-adaptive manner through a preset divergence threshold, and the number of the alliances does not need to be preset; and finally, vehicles in each alliance cooperatively train an exclusive alliance model. According to the method, the problem of model generalization under data isomerism is effectively solved, the prediction precision and convergence speed of each vehicle end model are remarkably improved while the privacy of the original data is guaranteed, and the method is suitable for an Internet of Vehicles environment.

The present application discloses a method and device for determining organ change information based on images and genes, and relates to the field of medical technology. The device includes: an acquisition unit for acquiring a target gene sequence and X medical images of a target object in a preset time period from a target data source; a determination unit for determining a model input feature based on X medical images and target gene sequences through a trunk neural network; a prediction unit for inputting the model input feature into a target branch network connected to the trunk neural network, and predicting the change information of the target organ of the target object through the target branch network based on the model input feature. The present application solves the technical problem in the prior art that the accuracy of predicting the radiotherapy effect of a patient is poor due to the lack of technology for deep integration of the patient's image features and genetic information.

The invention discloses an electric power operator behavior edge identification and AR correction method and system based on a double-flow heterogeneous lightweight network. The method comprises the following steps: acquiring a first visual angle video stream and sensor data of a six-axis inertial measurement unit through intelligent equipment; performing normalization and de-noising processing on the video frame, and segmenting the video frame into short-time video clips according to a time window; running the pre-trained double-flow heterogeneous lightweight network, performing real-time reasoning on the video clip, and outputting a current behavior category label; the behavior confidence degree output by the double-flow heterogeneous lightweight network model is compared with a preset threshold value to judge whether the event is a violation event or not; if the event is identified as a violation event, a deviation correction mechanism is triggered immediately; and encrypting and uploading the violation event log to an enterprise security management platform or other third-party platforms for users to check, count and trace. The method can effectively solve the problems of high behavior identification delay, dependence on cloud, low edge equipment operation efficiency, insufficient multi-behavior identification precision and the like in the prior art.

The invention provides a generative artificial intelligence service-oriented edge collaborative reasoning optimization method, which relates to the technical field of distributed computing, and comprises the following steps: a client sends a model request and text description to an edge server; the edge server generates an initial noise image according to the text description and the generative artificial intelligence model, and sends the generative artificial intelligence model and the initial noise image to the client; the client executes multiple rounds of reasoning processes according to the generative artificial intelligence model and the initial noise image to generate a client noise image; and the edge server continues to execute the remaining reasoning process based on the client noise image to generate a noise picture after reasoning is completed, and the noise picture after reasoning is completed is converted through the variational auto-encoder to obtain a generated picture. According to the invention, high-quality performance of service can be kept under the condition of limited computing resources and network bandwidth, and the stability and safety of user experience are remarkably improved.