74results 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 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 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 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.

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 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 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.

The invention discloses a multi-layer image display method and a display engine. The pixel row is segmented according to the rendering layer of the target pixel point on the pixel row in units of the pixel row in the target image; for each segment , extracting the source image mapping information of the segment from the source image corresponding to the rendering layer, and generating pixel information of all target pixels on the corresponding segment according to the preset interpolation algorithm and the source image mapping information of the segment. The solution of the present invention only needs to read and zoom the relevant pixel information of the actual display layer, which solves the problem caused by the existing multi-layer image display technology because the source image of each layer needs to be zoomed and then superimposed and displayed. The problems of long time consumption and low efficiency can effectively reduce the occupation of CPU resources and release memory bandwidth.