177results about How to "Understand technical means" patented technology

The invention provides a target tracking method, a device, a system and a computer readable storage medium. The method comprises the steps of obtaining current to-be-tested frames of a plurality of cameras arranged in a monitoring area; performing target detection on the current to-be-detected frame of each camera in the plurality of cameras in sequence to obtain a detection frame set corresponding to each camera; and performing target tracking according to the detection frame set corresponding to each camera, and determining a global target trajectory according to a tracking result. By utilizing the method, the computing resources of target tracking based on multiple cameras can be reduced.

The invention provides a task scheduling method, device and system and a computer readable storage medium, and the method comprises the steps: splitting a to-be-scheduled scheduling task into a plurality of subtasks, arranging the plurality of subtasks into a plurality of stages, and configuring a task dependency relationship among the plurality of subtasks and a stage dependency relationship among the plurality of stages; in response to an execution instruction of the current stage, obtaining all subtasks of the current stage, and issuing to-be-executed subtasks meeting preset execution conditions to an execution node, wherein the preset execution conditions are determined according to the stage dependency relationship, the task dependency relationship and the task execution state; and updating the execution state of the to-be-executed subtask according to an execution result fed back by the execution node. By means of the method, the execution process of the subtasks of the scheduling task can be more effectively managed, task scheduling under various complex service scenes can be better met, and the task scheduling performance is effectively improved.

The invention discloses a communication method for Bluetooth headset equipment and the Bluetooth headset equipment. The communication method comprises the steps that a first headset serves as Bluetooth main equipment to establish Bluetooth connection with data source equipment; the first headset and a second headset establish wireless connection; according to a preset period, in a first time period of each preset period, the first headset and the second headset exchange control information used for Bluetooth communication with the data source equipment; and in a second period of time of each preset period, one of the first headset and the second headset serves as a main headset to send a Bluetooth transmission request to the data source equipment, and the first headset and the second headset receive Bluetooth data sent by the data source equipment in response to the Bluetooth transmission request at the same time based on Bluetooth connection, wherein the first headset and the secondheadset are switched to the main headset in an alternate switching mode or a switching indication mode based on the control information according to the preset period. According to the technical scheme, better balance between the main headset and the auxiliary headset can be achieved, and the main headset and the auxiliary headset can be switched or backed up at any time.

The invention provides a safe cross-domain login method and system and a server. The method comprises the steps that a first application server receives a login request which is sent by a first application client and is used for requesting to log in a target webpage provided by a second application server, and obtains user identity information corresponding to the first application client according to the login request, thereby generating a request message according to the user identity information and sending the request message to a second application server; the second application server generates a temporary trust token based on the request message and sends the temporary trust token to the first application client; the first application client sends a skip request carrying the temporary trust token to a target webpage provided by a second application server; and the second application server verifies the temporary trust token and carries out login crediting on the first application client after the temporary trust token passes the verification. By means of the method, non-inductive cross-site access can be achieved without secondary login, better internet experience is provided for users, the jump link is protected based on the temporary trust token, and the method is safer.

The invention provides a method and device for detecting a specific facial syndrome, and a computer readable storage medium. The method comprises the following steps: acquiring a face image of a to-be-detected object; determining at least one key area image of the face image; inputting the face image into a trained global detection model to obtain a global detection result; inputting at least onekey area image into a trained local detection model to obtain at least one local detection result; and according to the global detection result and the at least one local detection result, determininga target detection probability that the to-be-detected object suffers from the specific facial syndrome. By means of the method, rapid detection of the specific facial syndrome can be automatically completed without complex measuring equipment, and the initial detection difficulty of a patient with the specific facial syndrome is remarkably reduced, and the detection accuracy is improved.

A management server and a management method thereof are provided. The management server is connected to a switch on a trunk port. The switch is connected to a plurality of cloud appliances in different virtual local area networks (VLANs). The management severs operates a management system and a translation system for managing cloud appliances. The translation system is operated to modify the receiving data packet by converting the source MAC address or the source IP address and removing the VLAN tag, and to modify the transmitting data packet by converting the destination MAC address or the destination IP address and adding a VLAN tag.

The invention discloses a pairing method used between binaural Bluetooth earphones and the binaural Bluetooth earphones, and the pairing method comprises the steps that: in a preset triggering state,a first earphone in the binaural Bluetooth earphones broadcasts a first pairing request on a predefined radio frequency channel, and carries out the scanning on the predefined radio frequency channel;and when the first earphone receives a first pairing response returned by a second earphone in the binaural Bluetooth earphone in response to receiving the first pairing request through scanning, thefirst earphone and the second earphone perform pairing confirmation. According to the technical scheme, a steady earphone pairing scheme can be provided for an existing TWS Bluetooth technical scheme.

The invention provides a service progress monitoring method, device and system and a computer readable storage medium. The method comprises the following steps: determining at least one similar historical service of a monitored service from a historical service set through employing a similarity model, and calculating the prediction progress information of the monitored service according to the atleast one similar historical service; acquiring actual progress information of the monitored service by utilizing the program burying points; and the progress of the monitored service is monitored bycomparing the actual progress information with the predicted progress information. By means of the method, the prediction progress information of the monitored service to be executed can be predictedaccording to the historical service, and then whether the actual progress of the monitored service is abnormal or not can be judged more accurately.

A direct-mode communication system comprises a first client direct-mode communication apparatus and a backhaul network. The first client direct-mode communication apparatus transmits a register signal to the backhaul network to register with the backhaul network. After the first client direct-mode communication apparatus is registered, the backhaul network determines a location relationship between the first client direct-mode communication apparatus and a second client direct-mode communication apparatus, and transmits a wi-fi connection signal to the first client direct-mode communication apparatus according to the location relationship. The first client direct-mode communication apparatus creates a wi-fi direct-mode communication connection with the second client direct-mode communication apparatus according to the wi-fi connection signal.

The invention provides an access control method, device and system and a computer readable storage medium. The method comprises the steps: receiving an access request of a user, the access request comprising user information and a target service resource; determining a resource permission tree of the user according to the user information, and constructing the resource permission tree of each userin advance according to the plurality of service resources, the authorization relationship among the plurality of users and the dependency relationship among the plurality of service resources; a complex permission judgment problem is simplified into a tree structure traversal search problem, so that the permission judgment process is greatly simplified.

The invention provides a character segmentation method and device and a computer readable storage medium. The character segmentation method comprises the following steps: acquiring a character regionimage, and converting the character region image into a grayscale image; performing clustering analysis on the grayscale image by using a fuzzy C-means clustering algorithm, and performing binarization processing on the grayscale image according to a clustering analysis result to obtain a binary image; obtaining at least one character positioning block from the binary image by using a projection method; and performing character segmentation on the character region image according to the position information of the character positioning block. By utilizing the method and the device, a more accurate character segmentation effect can be realized for the character region image with poor image quality.

The invention provides an access docking device and system and a method and device for applying the access docking device. The access docking device is deployed in a Hadoop computing server, and comprises a compatible interface layer used for compatibly realizing a file system interface of Hadoop so as to realize access docking with a Hadoop computing service component; the operation implementation layer is used for providing a first interface function for the compatible interface layer so as to implement file operation required by the Hadoop computing service component under a file system interface; and the storage access layer is used for providing a second interface function for the operation implementation layer so as to convert file operation into access operation for object storage in the distributed storage. By utilizing the access docking device, decoupling separation of the Hadoop computing service and the storage service can be realized, and the object storage in the distributed storage can be directly accessed.

The invention provides a permission application method, component and device and a computer readable storage medium, and the method comprises the steps of receiving a permission application request of an external application, the permission application request comprising a permission name of a target permission; if the target permission is not authorized, creating a transparent permission application container according to a current stack top page (Activity) component; applying for the target permission from the Android system by using the permission application container, and receiving a permission application result from the Android system; and calling a corresponding callback function according to the permission application result to realize successful callback or failed callback of the target permission. By means of the method, the permission application process can be greatly simplified, and code redundancy is reduced.

The invention discloses a distributed energy conversion method and system. The distributed energy conversion method comprises the steps that first fluid flowing in a first circulation loop in a circulating mode is refrigerated by using a working medium of a heat pump; a heat pump is used for compressing the working medium to further raise the temperature of the working medium for heating second fluid flowing in a second circulation loop in a circulating mode; the heated second fluid is conveyed to be used for heating an input pressure working medium of an air engine to enable the input pressure working medium to vaporize into pressure working medium gas for actuating the air engine, and after the input pressure working medium is heated and cooled, the second fluid is reheated by the working medium of the heat pump for reheating the input pressure working medium of the air engine, so that the second fluid is cyclically heated and cooled; and the refrigerated first fluid is conveyed to be used for condensing an output pressure working medium of the air engine, and after the output pressure working medium of the air engine is condensed and heated up, the first fluid is re-refrigeratedby the working medium of the heat pump for condensing the output pressure working medium of the air engine, so that the first fluid is cyclically condensed and heated up.

The invention provides a wireless earphone pairing method and system, and the method comprises the steps: respectively building wireless communication connection between an intermediate device and a first wireless earphone and between the intermediate device and a second wireless earphone, and transmitting pairing information to the first wireless earphone and the second wireless earphone based onthe built wireless communication connection; establishing a wireless communication link between the first wireless headset and the second wireless headset based on the pairing information. By using the method, pairing between the first wireless earphone and the second wireless earphone can be conveniently realized.

The invention provides a data compression method and device, a data decompression method and device and a processing method and device based on data compression and decompression, and the data compression method comprises the steps: receiving to-be-compressed data which is sparse data outputted by any layer of a neural network model; and performing compression processing on the to-be-compressed data based on a sparse compression algorithm and a bit plane compression algorithm to obtain compressed data. By utilizing the method, a relatively high compression ratio can be realized, so that the data transmission bandwidth and the storage space of an external memory can be saved, the memory access efficiency is improved, and the chip computing power is improved.

The invention provides a parallel processing method and device for a graphics pipeline and a readable storage medium, the method is applied to a driving layer of a multi-core graphics processor, the graphics pipeline adopts a TBR-based architecture, and the method comprises the following steps: receiving a plurality of drawing call instructions, grouping the plurality of drawing call instructions to generate a specified number of geometric stage requests, each geometric stage request carries a sequence identifier determined according to an original input sequence of a plurality of drawing call instructions; in response to the rendering instruction, distributing a specified number of processing requests to a plurality of cores of the multi-core graphics processor in parallel, and enabling the plurality of cores to execute geometric phase operations in parallel to generate a polygon list corresponding to each geometric phase request; and splicing the polygon list according to the sequence identifier carried by each geometric stage request to obtain a target polygon list. By means of the method, the geometric stage processing efficiency based on block rendering can be improved.

The invention provides an abnormal node recognition method, an abnormal node recognition model training method, an abnormal node recognition model training device, an abnormal node recognition model training system and a storage medium, the abnormal node recognition model training method comprises the following steps: constructing an explicit association graph based on an explicit association relationship between transaction elements, the transaction elements at least comprising a transaction account; constructing an implicit association graph based on the similarity among the plurality of transaction accounts, wherein the similarity among the plurality of transaction accounts is calculated according to the historical transaction data of the plurality of transaction accounts; fusing the explicit association graph and the implicit association graph to obtain an association graph taking the transaction account as a node; and training a graph neural network based on the features and labels of each transaction account and the association graph to obtain an abnormal node recognition model. By means of the method, the complete and high-accuracy association graph structure can be constructed, and the recognition effect of the node recognition model is improved.

The invention provides a training sample generation method and device, a model training method and device, and a character recognition method and device. The method comprises the steps: obtaining a character image, and determining each character contained in the character image; determining a weight value of each character contained in the character image by using a projection method, and labelingthe character image according to the weight value of each character contained in the character image to form a training sample; the model training method comprises the step of training a character recognition model by utilizing the training sample. The character recognition method comprises the step of performing character recognition by utilizing the character recognition model. By utilizing themethod and the device, accurate recognition of the character image of the double-half character type of the character wheel type meter can be realized, and an recognition result with bias and higherrecognition accuracy can be given.

The invention provides a write strategy adjusting method for a cache, a cache device and computing equipment, the cache comprises a plurality of cache groups mapped with a lower-level memory in a group associative mode, and the method comprises the following steps: setting a plurality of monitoring groups and a plurality of following groups in the plurality of cache groups, the monitoring groups comprise a first monitoring group configured to be in a write-back and write allocation mode and a second monitoring group configured to be in a direct write and non-write allocation mode; and monitoring instance modes executed by the first monitoring group and the second monitoring group in response to the cache access request, and adaptively adjusting a write strategy of the following group according to the instance modes. By means of the method, the write strategy can be adjusted in a self-adaptive mode according to the dynamically-changing cache access scene, and then the cache function is optimized.

The invention provides a method and device for estimating a signal arrival angle. The method comprises the following steps: acquiring a reference signal sequence and a plurality of groups of signal sample sequences by an array antenna; converting the reference signal sequence and the plurality of groups of signal sample sequences into a reference phase sequence and a plurality of groups of phase sample sequences, respectively performing Vandermonde inverse operation to determine a corresponding reference offset value and a plurality of sample offset values to form a space vector, and determining a signal arrival angle estimation value according to the inner product of the space vector and a plurality of expected space vectors. By using the method, the arrival angle can be estimated only byperforming a series of operation processes with low complexity, so that the operation amount is saved, and a more accurate estimation value is obtained.

The invention provides a signal arrival angle estimation method and device and a computer readable storage medium. The method comprises the steps: carrying out preprocessing of a plurality of paths ofsignals received by a non-omnidirectional array antenna to obtain a receiving phase sequence; performing phase compensation on the received phase sequence to obtain a compensation phase sequence corresponding to each of the plurality of candidate arrival angles; executing phase difference operation according to the compensation phase sequence corresponding to each candidate arrival angle to obtain a phase difference sequence corresponding to each candidate arrival angle; and performing weighted summation according to the absolute value of the phase difference sequence corresponding to each candidate arrival angle in combination with an antenna pattern of the array antenna, and estimating a signal arrival angle from the plurality of candidate arrival angles according to a summation result.With the method, the estimation error caused by the jitter of the received signal can be reduced.

The invention provides a data quantification method and device based on a neural network and a computer readable storage medium. The method comprises the steps of determining the first output activation of a target layer of the neural network according to any one preset frame in a to-be-detected frame sequence, and carrying out the quantification of the first output activation based on a current quantification parameter; iteratively updating the current quantization parameter according to the first output activation of the target layer and the target quantization bit width; and determining second output activation of the target layer according to the next frame of any preset frame in the to-be-tested frame sequence, and performing quantization operation on the second output activation based on the updated current quantization parameter. By means of the method, quantization errors can be reduced, and a better data quantization effect is achieved.

The invention provides a business processing method, device and system and a computer readable storage medium, and the method comprises the steps: obtaining a to-be-processed business, and decomposingthe to-be-processed business into a plurality of sub-businesses according to a preset processing flow; creating a plurality of thread groups corresponding to the plurality of sub-businesses, and configuring a plurality of queues among the plurality of thread groups according to the preset processing flow; and processing the corresponding sub-business through each thread group in the plurality ofthread groups, and transmitting data among the plurality of thread groups by utilizing the plurality of queues. By utilizing the method, the business to be processed can be decomposed into a pluralityof sub-businesses, and each sub-business is subjected to independent multi-thread operation, so that the processing efficiency of complex businesses is improved.

The invention provides a multifunctional household cabinet. The cabinet comprises a cabinet body, a clothes drying cavity, an air conditioner assembly and a washing assembly are at least integrated in the cabinet body, an air inlet and an air outlet are formed in the wall of the cabinet body, the air inlet is in fluid communication with the air conditioner cavity, the air outlet is in fluid communication with the clothes drying cavity, and a communicating and ventilating opening is formed between the clothes drying cavity and the air conditioner cavity; a first air duct for airing clothes is defined between the ventilation opening and the air outlet of the clothes drying cavity; the air conditioner cavity defines a second air duct between the air inlet and the communicating and ventilating opening, and the air conditioner assembly is configured to drive airflow to enter the second air duct from the air inlet and then to be discharged to the clothes drying cavity from the communicating and ventilating opening; and the washing assembly is arranged below the clothes drying cavity, and a clothes taking and placing opening which is communicated with the washing assembly and can be opened and closed is formed in the bottom of the clothes drying cavity. By means of the cabinet body, occupied space can be saved, and resource waste is remarkably reduced.

The invention provides a neural network quantification method and device, and a computer readable storage medium. The method comprises the steps of determining distribution data of activation output of a target network layer of a neural network according to a correction data set; determining the target quantization range of the target network layer according to the distribution data; and performing fixed-point quantization on the target network layer according to the target quantization range and the target quantization bit width. By utilizing the method, the precision loss in neural network quantization can be reduced, and a better quantization effect is achieved.

The invention provides a clock synchronization method and system. The system is composed of a core clock server, a plurality of regional clock servers and a plurality of cloud virtual machines, and the method comprises the steps that the core clock server performs first clock synchronization on the regional clock servers, and adjusts a synchronization strategy of the first clock synchronization according to feedback information of the regional clock servers; and the regional clock server performs second clock synchronization on the cloud virtual machine in the corresponding region, and adjustsa synchronization strategy of the second clock synchronization according to feedback information of the cloud virtual machine in the corresponding region. By utilizing the method, the stability of clock synchronization of the large-scale cloud virtual machine in the cloud environment can be improved.