215results about How to "Resolve interruption" patented technology

The invention relates to a video portrait tracking method based on human face identification in a complex scenario. The method comprises the steps that 1) acquired video data are decoded to acquire a series of video frame images; 2) human face detection and object tracking methods are used to process a series of acquired video frame images to acquire all possible human face trajectories; 3) the similarity between each human face trajectory acquired in step 2) and a subsequent human face approaching trajectory is determined, and human face trajectories with the similarity higher than a set threshold are merged; and 4) a merged human face trajectory is output. Compared with the prior art, the method provided by the invention has the advantages of fast processing speed, high detection precision, low false alarm rate and the like. The problems of human face trajectory disruption and confusion, which are caused by human head turning and blocking, can be solved.

The invention discloses a system based on wireless multi-link bandwidth aggregation, and belongs to the field of wireless data communication. The system disclosed by the invention comprises two devices: a wireless transmission terminal and an aggregation server. According to a real-time state of each wireless link, a weight value of each link is subjected to dynamic regulation, so that normal transmission of data and stability of a bearer service are effectively ensured. The system is not limited to a category of an operator, a wireless network mode and the like, can be suitable for binding of the same or different bandwidth interfaces, and can sufficiently utilize bandwidth resources of each link. According to the system disclosed by the invention, by binding multiple links, a transmission rate of each link can be reasonably regulated, and effects of increasing a service bandwidth and reinforcing bearing stability are taken.

The invention discloses an intelligent resilient framework (IRF) system-based method and an intelligent resilient framework (IRF) system-based system for upgrading an box type equipment. The method comprises the following steps of: upgrading a current version of each equipment in a first group by using a new version, and sharing the service flow of each equipment in the first group to a second group; and upgrading a current version of each equipment in the second group by using the new version, and sharing the service flow of each equipment in the second group to the first group. Through the method and the system, the upgrading of incompatible versions in the box type equipment is realized, so that service is uninterrupted.

The invention discloses a roaming link switching method, a mobile terminal, a network module and a storage medium. The roaming link switching method comprises the following steps: searching availablewireless access points in the current environment to obtain to-be-connected wireless access points; separately establishing links with the to-be-connected wireless access points, selecting a link fromthe links to serve as a main link to perform data communication, and using the rest links as standby links; then when it is detected that the communication quality of the main link does not satisfy preset conditions according to the main link and the standby links and that roaming link switching needs to be executed, selecting a link from the standby links, and switching the selected link into the main link to perform the data communication. Therefore, the problems of link interruption and packet loss in the roaming link switching process of the existing wireless local area network system aresolved, the data communication of a user terminal is not interrupted in the roaming link switching process, reliable network service is provided for the user, and the network experience of the user is improved.

The invention discloses a terminal switching method in a multi-network mode, comprising the following steps: setting a buffer area in a data buffering module of a terminal and setting an identifier for recording node marks in a terminal management/control module; a signaling processing module of a terminal Communicating with the network to stop data transmission between the data processing module and the server, the terminal management/control module caches the data related to data transmission in the cache area, and uses the identification to record the node flag; and the terminal registers to another network. The invention also discloses that the terminal switching device under the multi-network mode includes: a data processing module, a signaling processing module, a data cache module, and a terminal management/control module. Through the present invention, the signaling processing mode of opening a cache area and related suspend, resume, etc. is adopted to ensure no loss in data transmission, especially the management of the buffer area ensures the integrity of the data.

The invention discloses a multi-starting-point incremental three-dimensional reconstruction method based on large-scale disordered images. The method comprises the steps of performing image matching, constructing a matching relationship chart, selecting a reconstruction starting point, detecting an edge, performing multi-starting-point three-dimensional reconstruction and splicing a submodel. By means of a clustering strategy and an iteration drift strategy, an image pair which is suitable as starting points is adaptively selected from a disordered image set and reconstruction is performed simultaneously; before reconstruction, determining an optimal reconstruction sub-process of each image according to a layered shortest path first algorithm, and determining a reconstruction boundary; and obtaining a whole three-dimensional reconstruction model according to sub-models obtained in different sub-processes and splicing the common parts contained in the sub-models. The three-dimensional reconstruction method according to the invention has advantages of settling a problem of suspension in reconstruction process when a current scene comprises a plurality of non-overlapped parts, obtaining a whole three-dimensional model which is covered by the image set, preventing three-dimensional structure error caused by matching and transmission of error images, realizing parallel processing in reconstruction sub-process, and improving image reconstruction efficiency.

The invention discloses a database processing method and a database processing device. The database processing method comprises the following steps: a proxy client receives a processing request sent by a business application, wherein the processing request is used to request processing data in a database; the proxy client determines a data server side for executing the processing request, wherein the business application accesses to the database through the proxy client and the data server side; the proxy client sends the processing request to the determined data server side; the determined data service terminal processes data in the database according to the processing request; the determined data service terminal sends a processing result to the proxy client; and the proxy client sends the received processing result to the business application. According to the database processing method disclosed by the invention, the problem of business service interruption caused by database access logic abnormity in the related art is solved, and therefore the effect of ensuring the normal performance of the business service is achieved.

The embodiment of the invention discloses a protection method of a cross-device linear multiplex section, gateways and a controller. The protection method comprises the following steps of receiving first negotiation information sent by a first gateway by the controller, wherein the first negotiation information is sent when the fault of a link circuit between the first gateway and a second gateway is detected by the first gateway; and sending second negotiation information to the first gateway and the second gateway by the controller, and thus changing the second gateway into a standby state according to the second negotiation information, wherein the second negotiation information is used for bridging business data in a work link circuit to a protection link circuit for transmission, the work link circuit is arranged between the second gateway and the controller, and the protection link circuit is arranged between the first gateway and the controller. The protection method, the gateways and the controller, disclosed by the invention, are suitable for the scene of accessing an SDH (Synchronous Digital Hierarchy) network into a public network in a double designation way.

The invention discloses a method and a device of bidirectional forwarding detection (BFD) oscillation damping. When a chain circuit is frequently disconnected intermittently, a line card processing board computes a penalty value through an oscillation damping algorithm and enables a BFD session to be in a damping state according to the corresponding condition; and after the fault of the chain circuit is eliminated, an LP enables the BFD session to be out of the damping state according to the corresponding condition. The method and the device of the invention solve the problem of service interruption caused by continuous extrusion of a SCTP buffer area when a certain chain circuit is frequently disconnected intermittently, so that services can be transmitted through other normal chain circuits, thereby avoiding the interruption of the whole service.

The invention discloses a network protection method, a network protection method system, a controller and equipment, which are used for solving the problem of service interruption in the prior art and improving the network reliability. The network protection method comprises the following steps that the controller receives first warning information reported by the equipment in a network, and determines that cable fracture occurs in a first link; the controller issues link flow table configuration with a link diagnosis mechanism to the equipment until first feedback information sent by the equipment is received or second warning information sent by the equipment is received after preset time, wherein the first feedback information is reported to the controller after the equipment creates a second link according to the received link flow table configuration, the controller determines information that the second link is available, and the second warning information is information that the second link is unavailable and is determined by the controller; after receiving the first feedback information, the controller sends a flow table instruction, wherein the flow table instruction is used for controlling the equipment to replace the first link with the second link, to remove the first link and to add the second link into an original protection group.

The invention discloses a switchover method of single-service multi-path transmission in a heterogeneous network. The method is based on a heterogeneous network integration system, and single-service of users is transmitted in multi-path mode by a network supporting seamless switchover and a network which does not support the seamless switchover. In a transmission process, when a transmission link of the network which does not support the seamless switchover is switched over, service transmitted through the network which does not support the seamless switchover is switched over to the network supporting the seamless switchover. When the transmission link of the network which does not support the seamless switchover finishes the switchover, multi-path transmission is recovered. The switchover method of the single-service multi-path transmission in the heterogeneous network is provided and aims at solving the problems of service interruption and data service losing of the multi-path transmission in a switchover process in the prior art. Real-time performance and integration of single-service in the switchover process are guaranteed by switching over the service transmitted by the network which does not support the seamless switchover to the network supporting the seamless switchover.

The invention provides a dynamic memory control method and a dynamic memory control system, which are used in memory control on a Linux operating system. The method comprises the following steps of: pre-allocating at least one first memory block which is not smaller than a memory mapping region allocating threshold to a user process in a memory mapping region of a process space; dividing the first memory block into at least two second memory blocks with the same sizes; when a process memory application of a memory being not greater than each second memory block is received, allocating the memory of each second memory block to a process corresponding to the process memory application; when the process releases the memory, returning the memory, occupied by the second memory blocks, of the process into the first memory block; and after all of the second memory blocks in the first memory block are released, returning the memory of the first memory block to the Linux operating system. The method and the system can effectively solve the problem that some services requiring repeated application and mass memory block release often cause service interruption of the operating system due to memory exhaust.

The invention discloses a method and a device for protecting a point-to-multipoint pseudo-wire (P2MP PW). The method comprises the following steps of: determining a switching node in the P2MP PW according to a PW routing table, wherein the switching node establishes a main P2MP PW and a standby P2MP PW; and switching a flow to the standby P2MP PW by using the switching node when the main P2MP PW fails. By the method and the device, the problem of specific implementation of a P2MP PW protection network in related technology is solved, and the stability and the safety of the P2MP PW are improved.

The invention discloses a heterogeneous network baton handover method supporting M2M (Machine to Machine) communication. A network is formed by a CDMA (Code Division Multiple Access) network system and an LTE (Long Term Evolution) network system. The method comprises the following steps of: when an M2M gateway needs to be carried out with handover, firstly carrying out hard handover on the CDMA network, and enabling the LTE network to bear all business data before handover; after the completion of a handover process of the CDMA network, redialing, and carrying out handover on the LTE network, wherein the CDMA network is responsible for all service flows before the handover of the LTE network; and after the completion of the handover of the LTE network, redistributing the service flows. The CDMA and the LTE adopt hard handover, and compared with soft handover, the waste of a large amount of channel resources is reduced. The baton handover method ensures continuity of business transmission in a handover process, so as to solve the problem that business is interrupted in the hard handover process.

The invention provides a processing method for I/O (input/output) data access interrupt, a system and equipment. The method includes transmitting second SCSI (small computer system interface) logon instructions to a memory array if virtualization equipment is judged to break down or first paths of a first path group break down completely, and transmitting first SCSI reservation deletion instructions carried with first path identities to the memory array via second paths so that the memory array deletes first SCSI reservation corresponding to the first path identities; transmitting second SCSI reservation setting instructions carried with second path identities to the memory array via the second paths after response information, which is transmitted by the memory array, of the second SCSI logon instruction and response information of the first SCSI deletion instruction are received so that the memory array sets second SCSI reservation corresponding to the second path identities; and processing I/O data access for the memory array via the second paths.

The invention provides an azimuth trace acquisition method and device for a flexible towed linear array sonar noise target so as to effectively solve a problem of target azimuth trace interruption anddeviation caused by flexible towed linear array formation disturbance and underwater acoustic environment complexity, and improve the accuracy of a sonar noise target tracking azimuth trace. The method comprises the following steps: step 1, performing split sub-array frequency domain beam forming on current frame data of an array element channel according to flexible towed linear array array manifold information and a previous frame target orientation; step 2, carrying out target azimuth accurate measurement on the output formed by the frequency domain wave beam according to a processing frequency band; step 3, performing association confirmation on the target angle measurement value of the current frame by using a variable tracking gate, and removing and correcting outliers; and step 4,carrying out smooth filtering processing on the multiple frames of target azimuth values to obtain a stable and continuous target azimuth trace. The device applies the method. According to the invention, the technical problem of improving the accuracy of the sonar noise target tracking azimuth trace is solved.

The invention discloses a dual-redundancy unmanned aerial vehicle brake controller and a cold standby control method. The dual-redundancy unmanned aerial vehicle brake controller mainly consists of amain control system, a standby control system and a monitor and arbitration unit, wherein the main control system comprises a main control microprocessor, a main bus transceiver and a main internal power supply; and the standby control system comprises a standby control microprocessor, a standby bus transceiver and a standby internal power supply. The cold standby control method is adopted in thebrake controller, and components and parts of standby control system are but out to reduce the electric stresses of the components and parts, so that the problem that the life of the components and parts of hot standby control systems are influenced by the electric stresses is solved; and in design of cold standby control systems, the monitor and arbitration unit is utilized to cache current PWM signal duty cycle instructions, so that the problem that interruption exists in PWM signal output in the cold standby brake control method is solved.