75results about How to "Avoid duplicate fetches" patented technology

The invention discloses a pedestrian target tracking method based on depth learning, which combines depth learning and correlation filtering to track the target, and effectively improves the accuracyof tracking on the premise of ensuring real-time tracking. Aiming at the problem of large change of target posture in tracking process, the deep convolution feature based on pedestrian attribute is applied to tracking. Aiming at occlusion problem, cosine similarity method is used to judge occlusion in order to avoid the introduction of dirty data caused by occlusion. In order to improve the efficiency and solve the problem of using deep convolution features in correlation filters, a bilinear interpolation method is proposed to eliminate quantization errors and avoid repeated feature extraction, which greatly improves the efficiency. Aiming at the problem of high-speed motion of target, a preselection algorithm is proposed, which can not only search the global image, but also can be used asa strong negative sample to join the training, so as to improve the distinguishing ability of the correlation filter.

The invention discloses a first-stage license plate detection and recognition method based on deep learning. The first-stage license plate detection and recognition method based on deep learning comprises the steps: S1, inputting a to-be-detected and recognized color image, and enabling the color image to be zoomed into an image with the size of 300 * 300 through an image preprocessor; s2, inputting the zoomed image into a detection module and a recognition module for license plate detection and license plate character recognition; and S3, regarding the output result of the combined detectionmodule and the recognition module as the license plate detection and recognition output result of the whole system. The beneficial effects of the invention are as follows: first-stage license plate detection and recognition system integrating license plate detection and license plate character recognition into one model has the main advantages that the advantages are as follows: according to the license plate detection and license plate character recognition algorithm based on deep learning provided by the invention, higher detection accuracy and character recognition accuracy can be kept in acomplex natural scene, so that the stability and the accuracy which exceed those of other related systems are shown.

The embodiment of the invention discloses a data processing method, a terminal device and a computer readable storage medium. The method comprises the steps of obtaining target data from a database, and storing the target data into a data buffer pool of a buffer server; receiving an operation request for the first target data of the user terminal; granting the operation authority of the data buffer pool to the user terminal corresponding to the operation request, wherein the operation authority cannot be granted to two or more user terminals; after the operation corresponding to the operationrequest is executed, recycling the operation authority. The operation authority is added to the data buffer pool, when the data in the data buffer pool needs to be operated, the operation authority needs to be obtained at first, and the operation authority cannot be obtained by two or more user terminals every time. According to the data processing method, the terminal device and the computer readable storage medium, the access efficiency can be effectively improved, the data processing time is shortened, and resource waste is lowered.

The invention discloses a positioning method. The positioning method comprises the following steps of obtaining a target image; obtaining the candidate key information of a plurality of candidate keyframes coarsely matched with the target image from a pre-constructed map, wherein the map is composed of the key information of the plurality of key frames; generating a region-of-interest (ROI) of the target image, and performing the region-of-interest matching on the target image based on the candidate key information to obtain a region-of-interest matching pair of the target image for at leastone candidate key frame; matching the feature points in the region-of-interest matching pair based on the candidate key information to obtain a feature point matching pair of the target image for theat least one candidate key frame; and calculating the relative pose of the target image and the at least one candidate key frame according to the feature point matching pair.

The present discloses a method and apparatus for obtaining billing information. The method includes the following steps that: user mails are obtained from a mail server end according to the mailbox login information of a user; a billing mail is identified from the user mails, wherein the billing mail contains electronic billing information; and the billing information is extracted from the billing mail. According to the method for obtaining the billing information of the invention, in a whole billing information obtaining process, the mailbox login information of the user is not uploaded to a billing server and is only stored on a user terminal, and therefore, user information leakage caused by attacks to the database of the billing server performed by criminals can be effectively prevented, and the information security of the user can be effectively ensured.

The invention discloses a paging data synchronization method, a paging data synchronization device and a paging data synchronization system. The paging data synchronization method includes sending a homepage data synchronization request to a network server, wherein the homepage data synchronization request includes a data updating timestamp; receiving homepage data returned by the network server, wherein the homepage data are acquired by the network server from a data storage device according to the data updating timestamp; assigning a data synchronization timestamp of a first piece of data information in the homepage data to a data refreshing timestamp, and assigning a data synchronization timestamp of a last piece of data information in the homepage data to the data updating timestamp; responding to a sliding operation instruction of a user, sending a next-page data synchronization request to the network server, wherein the next-page data synchronization request includes a data updating timestamp; receiving next-page data returned by the network server, wherein the next-page data are acquired by the network server from the data storage device according to the data updating timestamp; assigning a data synchronization timestamp of a last piece of data information in the next-page data to the data updating timestamp.

The invention discloses an image file processing method and an image file processing system, and aims to solve the technical problem that a network image file and a local image file cannot be shared by a main program module and a viewer program module of an image browsing software. According to the image file processing method and the image file processing system, network image file information, a network image file, local image file information and an abbreviation file of a local image file are shared through a local sharing storage space by the main program module and the viewer program module of the image browsing software. The local sharing storage space adopts a memory circular queue manner, and reading and writing are carried out simultaneously by a reading indicator and a writing indicator. According to the image file processing method and the image file processing system, the network and / or local image file information and the image file are prevented from repeatedly being acquired, network resource is saved and user waiting time is reduced, and therefore a user can browse and manage the local and network image in a convenient, real-time and efficient mode.

The invention discloses a method and a device of data obtaining. The method is applied to interconnection and interworking of a plurality of content distribution networks (CDNs). The method comprises that when a downstream CDN obtains content from at least one upstream CDN, a content identification list is inquired according to content identifications of the content, wherein the content identifications of all the content obtained by the downstream CDN are stored in the content identification list, and each piece of content only corresponds to one content identification; and when the content identifications of the content are not found in the content identification list of the downstream CDN, the content is obtained from the upstream CDNs. The method and the device of data obtaining can resolve the problem that downstream CDNs in correlation technique repeatedly obtain and cache the same content.

The present invention provides a communication method and system and a message server. The method comprises: receiving the request message published by a message publishing party; storing the request message into a first receiving message theme area corresponding to an information subscription party according to the identification information of the message theme area of the receiving information subscription party to allow the information subscription party to obtain the request message from the receiving information subscription party; receiving the receipt information published by the information subscription party, wherein the receipt information is generated by the information subscription party according to the request message and includes the identification information of a returning theme area; and storing the receipt message into the corresponding returning theme area according to the identification information of the returning theme area to allow the message publishing party to obtain the receipt message.

The invention provides a method and a system for optimizing a starting sequence of power generators. The method comprises the steps as follows: determining father node information of nodes in shortest weighted paths and weight values of the shortest weighted paths; determining comprehensive priority indexes of target power generators respectively according to the weight values and the starting capacity and the climbing rates of the target power generators; taking the target power generator with the minimal comprehensive priority index as a starting target; determining the shortest weighted path of the starting target through the father node information; solving a system comprising the starting target and the shortest weighted path corresponding to the starting target as well as an electric network by the primal-dual interior point optimal power flow; judging if the optimal power flow of the system is converged; and if so, incorporating the starting target and the shortest weighted path corresponding to the starting target into the electric network. According to the method, the optimal power transmission path is prevented from being repeatedly acquired, and the efficiency of acquiring the starting sequence of the power generators after blackout is greatly improved.

The invention belongs to the technical field of mobile internet, and provides a data processing method, device and system for graph code scanning. The method comprises the steps of acquiring graph code information and time information, and a storage mode; according to the type of the graph code information, acquiring decoding information and check information which are matched with the graph code information; according to the decoding information and the graph code information, acquiring bar code data; according to a pre-acquired software requirement format, storing the bar code data, and uploading the bar code data to mobile office software and a cloud database; and performing local storage of the bar code data or transmitting the bar code data to a short-range communication interface address. The method can further receive commodity basic information which is fed back by a service provider address and corresponding to the bar code data, edit the commodity basic information corresponding to the bar code data, and update the commodity basic information to the service provider address. The data processing method, the data processing device and the data processing system for the graph code scanning provided by the invention can improve the data processing efficiency in the process of graph code scanning, and can quickly realize data transmission and sharing and reduce the equipment cost.

The invention discloses a method and device for monitoring data quality, and relates to the technical field of computers. A specific embodiment of the method comprises the following steps: taking source data and / or output data of a data calculation task as to-be-verified data; obtaining a verification rule of the to-be-verified data from a preset data verification rule set; executing a data verification task on the to-be-verified data based on the verification rule, wherein the data verification task and the data calculation task are carried out asynchronously. According to the embodiment of the invention, the data quality can be ensured, and the possibility of data calculation task errors is greatly reduced; the workload and complexity of data calculation and verification are reduced; data verification rules can be modified conveniently, and personalized customization of a data verification strategy is supported.

It is possible to drastically improve the processing efficiency of the acquisition processing. According to a content data request, it is judged whether the content data is stored on the hard disc. When the content data is not stored, the content data is acquired from a content providing server (6) and after this, a content attribute file (47) is acquired from an attribute information providing server (7). When the content data is stored on the hard disc, the content attribute file (47) is acquired from the attribute information providing server (7) without requesting the content data. Thus, it is possible to eliminate duplicate acquisition of content data, thereby significantly reducing the acquisition processing time. This can drastically improve the processing efficiency of the acquisition processing.

The invention provides an envelope acquisition method for an intelligent sound box. The envelope acquisition method comprises the following steps of acquiring first URL data corresponding to sound source files stored in the intelligent sound box; in a correspondence relation list of cached envelopes and the URL data, searching whether the first URL data is included; if the first URL data is included in the correspondence relation list of the cached envelopes and the URL data, then matching the envelopes corresponding to the first URL data and the sound source files by an intelligent terminal. The envelope acquisition method and device provided by the invention can prevent folder envelopes from being acquired repeatedly so as to be beneficial to saving communication resources.

The invention relates to a page switching method and device, a storage medium and electronic equipment, and aims to solve the problems of white screen and resource waste caused by repeated acquisitionof display content in the process of jumping from a card-form page to a landing-page-form page. The page switching method comprises the steps that in response to a switching display operation for a first display window, controlling the height and width of the first display window to be changed, wherein the first display window is switched to a second display window; determining target display content to be displayed in the second display window, the target display content including first display content corresponding to the first display window and second display content corresponding to thesecond display window, wherein the second display content is synchronously acquired when the first display content is acquired, and the displayed content is hidden in the display process of the firstdisplay window; controlling the target display content to be displayed in the second display window.

The invention discloses a user behavior real-time processing method. The method comprises the following steps: step 1: when a user behavior occurs, judging whether user behavior information needs to be acquired or not; when the user behavior information needs to be acquired, performing the step S2; when the user behavior information does not need to be acquired, performing the step 5; step 2: acquiring the user behavior information, and organizing the user behavior information into a data stream; step 3: updating user attributes through the user behavior information acquired in the step 2, and adding the user attributes before and after updating into the data stream; step 4: analyzing the data stream generated in the step 3, and finishing a system action required to be executed; and step 5: stopping the processing of the user behavior. Through the method, the technical effects of increasing the business operation speed, reducing the development cost and reducing the system coupling degree can be achieved.

The invention relates to an extracting statistical method and system aiming at semi-structured big data, and belongs to the field of big data extracting statistics. The extracting statistical method and system solve the problem that the process of extraction and statistics of the semi-structured big data is tedious and easy to cause data redundancy. According to the extracting statistical method and system, by providing a client to let a user input operational statements of the extraction and statistics aiming at the semi-structured big data, the operation statements are synchronized to a parsing conversion module which parses the operation statements and converts parsing results into configuration rules; the client calls an application engine module to generate a job task according to the configuration rules, and submit the job task to a underlying framework; the underlying framework splits the job task into multiple subtasks, distributes the subtasks to a cluster for execution, and returns the resulting data obtained after the execution back to the client to show to the user. The extracting statistical method and system are used for improving the maintainability and the automatic visualization level of the extraction and statistics aiming at the semi-structured large data and reducing the data redundancy, and the extracting statistical method and system are simple and reliable.

The invention relates to an image processing method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a plurality of target images of a first object;performing image fusion on the plurality of target images of the first object to obtain a fused image of the first object; and performing image feature extraction on the fusion image of the first object to obtain fusion features of the first object. According to the embodiment of the invention, the image feature extraction efficiency can be improved.

The invention discloses an image processing method and device, a storage medium and electronic equipment. The method comprises the following steps: acquiring a difference region in a first image and a second image; wherein the first image and the second image are images to be compared; respectively determining a minimum enclosing rectangle of the contour corresponding to each difference area to obtain a minimum enclosing rectangle set; combining a plurality of external rectangles of which the adjacent or overlapping degree is greater than or equal to a preset threshold value in the minimum external rectangle set to obtain a target external rectangle set; the adjacent or overlapping degree is the ratio of the first area to the second area; wherein the first area is the sum of the areas of any two circumscribed rectangles in the minimum circumscribed rectangle set, and the second area is the area of the minimum circumscribed rectangle of the union set of any two circumscribed rectangles. The technical problem that the image change information extraction efficiency is low is solved.

The invention relates to the technical field of network live broadcast, and discloses a live broadcast service data processing method, a live broadcast service data processing device, live broadcast service data processing equipment and a medium. The data items comprise data items which need to be called by a live broadcast service and are distributed in a plurality of data sources and operation items used for executing predefined operation on the data items, the data items comprise attribute data items of an anchor user, and the operation items comprise attribute tags; according to the operation items, predefined operation corresponding to the operation items is executed on the corresponding data items in the multiple data sources, a data set corresponding to the attribute tags in the operation items is determined, and the data set comprises data item subsets corresponding to multiple anchor users; and pushing a data set corresponding to the live broadcast service to the terminal equipment. According to the invention, the standardized operation of calling the data items of the multiple data sources by the multiple live broadcast services is realized through the data definition template, the data entropy increase is avoided, the system efficiency is improved, and the system overhead is saved.

The invention provides a construction method of a large-scale remote sensing image feature point library. The method comprises the steps of determining dimensions of tiles, performing overlapping grid partition on a remote sensing image according to resolution information of the remote sensing image to achieve division of the remote sensing image, extracting feature points and descriptor data of the tiles by detection algorithms of the feature points of SIFT / SURF (scale invariant feature transform / speed up robust feature) and the like, and storing coordinate information corresponding to the feature points and descriptors in a spatial database by taking an R* tree as an index. The feature point spatial database generated by the method can be used for quick splicing templates or quick discovery and detection of variations of the remote sensing image; and compared with the prior art, the method takes feature point detection as pretreatment, avoids repeated detection and can improve processing efficiency.

The invention provides a user selection method and a device. The method includes: receiving a selection condition for user selection again; analyzing corresponding information in accordance with the selection condition, the result of a previous user selection result and / or a previous selection condition to obtain the user selection result for the implementation of the user selection again. Through the further selection based on the selection result obtained by the pervious user selection and relevant information, the embodiment of the invention does not need a requiring party to re-submit the previous selection conditions, avoids the interactive process of repeatedly obtaining relevant information, improves the selection efficiency and provides better and more efficient services to the requiring party.

The invention relates to a driver fatigue driving real-time detection method and system based on deep learning, and the method comprises the steps: S1, collecting a real-time image of a cab, and obtaining a face image through a neural network; sS, normalizing the face image to obtain a face input feature map, and inputting the face input feature map into a face key point detection deep neural network to obtain face key points; S3, selecting a part of face key points to be matched with 3D standard face key points, and identifying nodding and non-front-looking behaviors; S4, extracting an eye region according to the face key points, detecting the opening and closing state of eyes, and calculating the eye fatigue state; S5, calculating the mouth opening and closing degree according to the mouth key points, and recognizing yawning behaviors; and S6, calculating the fatigue value of the driver according to the steps S3, S4 and S5. Rich face feature information can be recognized, different behavior features of the head, the eyes and the mouth are detected respectively, multiple fatigue behaviors are detected, and the reliability and practicability of the method are improved.

The invention discloses an unmanned aerial vehicle fault management method and device and a storage medium. The method comprises that a remote control terminal sends, to an unmanned aerial vehicle, fault request information for acquiring each component of the unmanned aerial vehicle (S101); the unmanned aerial vehicle acquires fault information sent by each component in response to the fault request information, generates a fault identifier according to the fault information of each component, and sends the fault identifier to the remote control terminal, the fault identifier comprising component information and fault information (S102); and after the remote control terminal receives the fault identifier, component information and fault information corresponding to the fault identifier are displayed in a display interface of the remote control terminal (S103). And the accuracy of obtaining the fault information is improved.

The invention provides a webpage information processing method and device. A main server stores acquired network resource data corresponding to a to-be-captured theme into a queue; each slave server acquires at least one piece of network resource data in the queue, sends an acquisition request to the network resource data and receives webpage information corresponding to the network resource data,so that the webpage information corresponding to the network resource data is captured at least through each slave server, and the capturing efficiency of the webpage information is improved. The main server empties the queue after the storage time of the network resource data corresponding to the to-be-captured theme reaches the expiration time, the updated network resource data corresponding tothe to-be-captured theme is re-stored in the queue, and the network resource data in the queue is updated regularly, so that the webpage information can be captured based on the changed network resource data, the webpage information is incrementally acquired, and the webpage information is prevented from being repeatedly acquired.

The invention provides a network data acquisition method and a network data acquisition system for sports events. The method comprises the steps of receiving a network data acquisition request; calculating first theme relevancy between the network data acquisition request and historical data; if the first theme relevancy of all historical data in a memorizer is smaller than target theme relevancy, extracting a uniform resource locator of at least one historical site from the memorizer; specifying an acquisition task; receiving an acquisition result of a specified network data acquisition device; analyzing the acquisition result according to a target parameter to obtain an analysis parameter; calculating second theme relevancy between the acquisition result and the network data acquisition request; and if the second theme relevancy is greater than or equal to the target theme relevancy, obtaining a first acquisition request through the acquisition request. Repeated acquisition of effective data is avoided, and whether the acquisition request reaches the target theme relevancy or not is judged in real time to obtain an ideal acquisition request.

The invention discloses a pedestrian re-identification method based on multi-channel attention features, comprising the following steps: 1) constructing a convolutional neural network model based on channel attention, and pre-training a backbone network; 2) extracting pedestrian pictures in the backbone network 3) The weighted vector is multiplied by the output feature of the backbone network, and then added with it to obtain the channel attention feature; 4) Multiple attention features are repeatedly extracted, and the sea The Ringer distance is used to regularize the feature diversity; 5) The attention feature is input into the fully connected layer and the classifier, and the training minimizes the cross entropy loss and metric loss; 6) The test set image is input into the trained model to extract features, and the metric Sorting achieves pedestrian re-identification. The invention extracts the discriminative features of pedestrians based on the attention mechanism, limits the repeated extraction of similar attention features, and effectively improves the accuracy and robustness of pedestrian re-identification.

The embodiment of the invention discloses a method for integrated detection and fine-grained identification of a remote sensing image ship target. The method includes: acquiring a training sample set; determining a feature extraction shared model integrated in a remote sensing image processing model, a location detection branch model for outputting the location information of the target object, and a type recognition branch model for outputting the type information of the target object ; Adopt the shared image features extracted from the training samples by the feature extraction shared model to alternately train the position detection branch model and the type recognition branch model, so as to complete the training of the remote sensing image processing model when the preset convergence conditions are met. . By adopting the solution of the present application, it is possible to avoid repeated extraction of the shared part of the image features during training or use of the position detection branch model and the type recognition branch model, avoid additionally increasing the task load of this part of the extraction operation, and not only greatly reduce the model training. amount of tasks.

The invention discloses a message communication method and a system. The message communication method includes the following steps. Firstly, before a sending client terminal sends information comprising images, the fact that whether an image managing server stores the images or not is judged. If negative, the images are uploaded to the image managing server and provided corresponding identification. The image managing server stores the images with identification of the images as a search main key. Secondly, the sending client terminal sends the message comprising the images to a receiving client terminal. The images in the sent message are expressed as descriptive information of the images. The descriptive Information comprises identification of the images. Thirdly, the images are obtained based on the identification of the images. The receiving client terminal displays the images at a corresponding position of the message. The message communication method and the system can use any images in message communication and save network resources.

The invention discloses a measurement parameter hybrid framing method of a missile-borne recorder. According to the method, measurement parameters are received by use of an FPGA chip, wherein digital measurement parameters directly go into the FPGA chip through a serial port for storage in FIFO2; and analog measurement parameters go into the FPGA chip through a multiplexer and an A / D converter which are sequentially connected and are finally written into an externally connected FLASH memory through the FPGA. The analog measurement parameters are acquired by use of an A / D converter with n channels, after fully packed data packets are obtained through acquisition, the data packets are sent to the FPGA; the FPGA, according to an acquisition frequency, extracts data from the fully packed data packets and carries out framing, and data after the framing is stored in FIFO1; and then, in a GPIO interruption service program, the data of the FIFO1 and the FIFO2 are respectively written into different pages of the externally connected FLASH memory. The method provided by the invention simplifies the frame format, improves the effective signal transmission bandwidth, reduces the complexity of a multipath switch control algorithm and data processing software, and has certain flexibility.