41results about How to "Troubleshoot missing technical issues" patented technology

The invention discloses a train data cache method, device and system, wherein the method comprises the steps of detecting the network state of a train during the driving of the train; when detecting that the network state is an abnormal state, storing wireless data to a cache region, and synchronizing the cached wireless data in the cache region to a temporary file, wherein the storage location of the temporary file is predetermined physical space in a magnetic disk; when the network state recovers the normal state from the abnormal state, sending the wireless data cached in the cache region to a ground server. By adoption of the method, the device and the system, provided by the invention, the technical problem of losing of train running data after transmission network outage between the train and a ground in the prior art is solved.

The invention discloses a method for generating an abstract based on an attention mechanism to generate a text abstract, which totally comprises two stages: the first stage is a sentence sorting process, the second stage is an abstract generation process, and the input of the abstract generation process is N sentences which are most relevant to an article theme and are obtained through the first stage. In the first stage, a supervised sorting method is provided for articles with titles, firstly, the similarity between each sentence and the title is calculated, then the articles are sorted according to the similarity, and finally N sentences with the highest similarity are selected. For the second stage, the invention proposes a new method for calculating the attention distribution betweenthe encoder and the decoder, i.e. the decoder should pay attention to different parts in the encoder at different times. Through the sorting method and the abstract generation method, the problem of article information attenuation caused by the fact that a part of text must be directly cut off to serve as input of the abstract generation model when the article is too long is relieved.

The invention discloses a multi-channel audio processing method and a multi-channel audio processing device. The multi-channel audio processing method comprises the following steps: determining to-be-output output channels of multi-channel audio; when the number of the output channels is smaller than the number of the sound channels of the multi-channel audio, acquiring multiple sound channel data of the multi-channel audio and azimuth data, corresponding to the output channels, of the multiple sound channel data; correcting the azimuth data, and thus corrected data is obtained for supplementing the lost data in the azimuth data; and determining the output data of the output channels based on the corrected data and the multiple sound channel data. With the adoption of the method and the device, the technical problem that information loss is caused when channel-decreasing treatment is carried out on the multi-channel audio file is solved.

The invention discloses a method, a device and a system for expanding and shrinking capacity of a database cluster. The method comprises the steps that when the database cluster is expanded, determining the number of target expansion nodes needed by nodes to be expanded in the database cluster; and synchronizing the capacity expansion information of the to-be-expanded node to a proxy device through a copying function of the to-be-expanded node, and synchronizing the capacity expansion information to target capacity expansion nodes corresponding to the target capacity expansion node number. Thetechnical problem of data loss caused in the process of horizontally expanding and shrinking the capacity of the storage database in the prior art is solved.

The invention discloses a network merging method and device, a storage medium and an electronic device. The method comprises: acquiring a plurality of to-be-merged sub-networks, each node in the sub-networks correspondingly representing an account, each edge, connected with two nodes, in the sub-networks representing that two nodes have an incidence relation, and each edge in the sub-networks having a weight; merging the plurality of sub-networks into a target network, the nodes in the target network being a union set of the nodes in the plurality of sub-networks, and the edges in the target network being a union set of the edges in the plurality of sub-networks; and configuring a target weight for each edge in the target network according to the weight of each edge in the plurality of sub-networks. According to the invention, the technical problem that the weight information of the network merging edge is lost in the prior art is solved.

The present invention discloses a video cloud storage system-based storage mode upgrading method, a video cloud storage system-based storage mode upgrading device and a system. The upgrading method includes the following steps that: when an upgrading instruction for upgrading a video cloud storage system is received, the index data of a first management node is backed up, so that backup data are obtained; the configuration file of one or more newly-added third management nodes in the video cloud storage system is configured according to cluster feature information, the backup data are importedto the third management node; the configuration files of the first management node and a second management node are configured through using the cluster feature information; and information synchronization is performed on the configuration file of the first management node, the configuration file of the second management node and the configuration file of the third management node. With the videocloud storage system-based storage mode upgrading method and the video cloud storage system-based storage mode upgrading device of the invention adopted, the technical problem of video storage data loss caused by a condition that storage work is required to be interrupted in the upgrading of the operating mode of the video cloud storage system can be solved.

The present invention provides a face tracking method with classification recognition function, including step (1) the sequence to be matched in the current frame is {F1C, F2C...FjC}; the sequence in the previous frame is {F1L, F2L.. ....FjL}; (2) Take out the FmC of the current frame in turn to calculate its LBP coded map; (3) After getting the LBP coded map, compare it with the LBP coded map of the previous frame to find out the FnL with the highest matching degree with FmC (4) judge whether it is the same person according to the degree of matching, center position distance relationship; if the face images of the same person are classified into the same class, if not add Fj+1L=FmC in the FL sequence; (5) repeat (two )-(4), until the sequence to be matched in the current frame is traversed; (6) get back to the (1) step to process the next frame. The invention adopts the LBP feature of the rotating unified mode to classify and judge the identity of the person, so that the faces of the same person are classified into one category, which solves the problem of information loss after the person is blocked in the traditional face tracking method, and expands the human face tracking method. The features of the face tracking system enhance its application value.

The invention discloses a storage mode upgrade method, device and system based on a video cloud storage system. Wherein, the upgrade method includes: in the case of receiving an upgrade instruction for upgrading the video cloud storage system, backing up the index data of the first management node to obtain the backup data; configuring the video cloud storage system using the cluster feature information configuration files of one or more newly-added third management nodes in the database, and import the backup data into the third management node; use the cluster characteristic information to configure the first management node and the second management node A configuration file of a node; performing information synchronization on the configuration file of the first management node, the configuration file of the second management node, and the configuration file of the third management node. The invention solves the technical problem that the video storage data is lost due to the need to interrupt the storage work when upgrading the working mode of the video cloud storage system.

The invention discloses a media flow optimization method and device, a storage medium and a processor. Wherein, the method includes: obtaining frames in the media stream; filtering frames whose definition is less than a dynamic threshold to obtain clear frames; comparing the similarity between the clear frames within the sliding window within a predetermined value, and determining the two frames with the largest difference; The quality of the two largest frames is evaluated, and the optimal frame among the two frames with the largest difference in each sliding window is determined according to the quality evaluation results. The invention solves the technical problem that the key frame is lost during the transmission process of the video stream.

The invention discloses a network merging method and device, a storage medium and an electronic device. Wherein, the method includes: obtaining a plurality of subnetworks to be merged, wherein each node in the subnetwork corresponds to an account number, and each edge connecting two nodes in the subnetwork indicates that there is an association relationship between the two nodes, Each edge in the sub-network has a weight; multiple sub-networks are merged into a target network, where the nodes in the target network are the union of nodes in multiple sub-networks, and the edges in the target network are edges in multiple sub-networks The union of ; configures target weights for each edge in the target network based on the weight each edge in the multiple subnetworks has. The invention solves the technical problem of loss of weight information of network merging edges in the related art.

The embodiment of the invention provides a method for selecting an outlet interface. The method comprises the following steps: obtaining the multicast rate; comparing the bandwidth of each member interface in a trunk interface with the multicast rate to obtain a first set, wherein the first set comprises all the member interfaces with bandwidth more than or equal to the multicast rate in the trunk interface and the trunk interface is the multicast outlet interface and comprises a plurality of member interfaces; and determining the multicast outlet interface according to the first set. Besides, the embodiment of the invention also provides corresponding equipment. By adopting the technical scheme provided by the embodiment of the invention, the technical problem of loss of the multicast flow can be solved.

The invention discloses a train data cache method, device and system. Wherein, the method includes: detecting the network status of the train during the running of the train; when detecting that the network status is abnormal, storing the wireless data in the buffer area, and synchronizing the wireless data buffered in the buffer area to the temporary In the file, the storage location of the temporary file is the predetermined physical space in the disk; when the network status returns from an abnormal state to a normal state, the wireless data cached in the cache area is sent to the ground server. The invention solves the technical problem that the running data of the vehicle is lost after the existing train and the ground transmission network are disconnected.

The invention relates to the field of energy storage and transmission, and discloses an energy storage temperature and humidity collection method, system, equipment and storage medium. The method includes: the temperature and humidity acquisition system sends a query command to a preset temperature and humidity meter cluster based on the MODBUS‑RTU protocol, and receives N collection data fed back by the temperature and humidity meter cluster, wherein the temperature and humidity meter cluster includes: N distributed temperature and humidity meter clusters Hygrometer, where N is a positive integer; read the collection string in the collection data, and decode the collection string according to the preset decoding algorithm to obtain temperature and humidity data; the information collection system receives the temperature and humidity data and judges the temperature and humidity data Whether the preset temperature and humidity safety threshold is exceeded; if the temperature and humidity safety threshold is exceeded, the safety warning information will be sent to the preset management port of the temperature and humidity meter corresponding to the temperature and humidity data.

The invention discloses a multi-channel audio processing method and a multi-channel audio processing device. The multi-channel audio processing method comprises the following steps: determining to-be-output output channels of multi-channel audio; when the number of the output channels is smaller than the number of the sound channels of the multi-channel audio, acquiring multiple sound channel data of the multi-channel audio and azimuth data, corresponding to the output channels, of the multiple sound channel data; correcting the azimuth data, and thus corrected data is obtained for supplementing the lost data in the azimuth data; and determining the output data of the output channels based on the corrected data and the multiple sound channel data. With the adoption of the method and the device, the technical problem that information loss is caused when channel-decreasing treatment is carried out on the multi-channel audio file is solved.