47 results about "Infobox" patented technology

The invention discloses a two-dimensional code including variable information. The two-dimensional code comprises a two-dimensional code body, a position detection graphic area and a function graphic area, wherein the position detection graphic area and the function graphic area are attached to the two-dimensional code body; the position detection graphic area is arranged at three corners of the two-dimensional code body, and comprises a positioning block, a partition block and an information box; the positioning block is arranged at the middle part of the partitioning block; the partitioning block and the positioning block are different in color; the partitioning block is arranged in the information box; the edges of the information box are parallel to the edges of the partitioning block; and plain code information is coated in the information box, and the plain code is at least one of characters, graphs, numbers and websites. The two-dimensional code including variable information provided by the invention can set the variable plain code information according to actual requirements in the position detection graphic area, the function graphic area structure of the two-dimensional code is damaged, the space of the two-dimensional code is fully utilized, and the application value of the two-dimensional code is greatly expanded.

The invention belongs to the technical field of Chinese knowledge base application, in particular to a book-oriented reading field knowledge map construction method. The method includes general knowledge map construction, namely acquiring knowledge on the internet, and integrating a general knowledge map; field knowledge map construction, namely combining with the general knowledge map and expending relative concepts and entities of books in an iterative manner, and extracting entity relation by combining with an entity Infobox table and traditional relation; and intelligent reading recommendation, namely marking core entities in electronic books from longer entities to shorter entities, and establishing links between the entities and a book knowledge map so as to realizing the intelligent reading recommendation. According to the method, by means of establishing the book-oriented reading field knowledge maps, the entities in the books can be explained or knowledge can be recommended, depth of knowledge is increased, facilitation, intelligentization and humanization of electronic reading are realized, and better user experiences are provided.

The invention relates to an invoice information identification method and device, computer equipment, a computer readable storage medium and a computer program product, and the method comprises the steps: inputting a to-be-identified invoice image to a vertex positioning model after a server obtains the to-be-identified invoice image, inputting the to-be-identified invoice image to the vertex positioning model, and obtaining the to-be-identified invoice image; obtaining vertex coordinate information of an invoice information area of the to-be-identified invoice image; extracting the invoice information area according to the vertex coordinate information of the invoice information area; correcting the invoice information area based on the vertex coordinate information of the invoice information area; inputting the corrected invoice information area into an information box detection model to obtain coordinate information of each key text information box in the invoice information area; and extracting the key text information in the key text information box according to the coordinate information of the key text information box. According to the invoice identification method provided by the invention, accurate identification of key information in the invoice can be realized.

The invention provides an oil and gas field equipment production state judgment system and method based on big data monitoring. The system comprises a storage unit, a video inspection control unit and a video analysis and identification unit; after the video inspection control unit sends an inspection instruction each time, the video inspection control unit adds an information frame into the database, including attribute information of the preset point, such as pumping unit information; the video analysis and identification unit is used for acquiring attribute information of the preset point in the database, determining the position of the oil pumping unit in the video picture through image analysis and comparison, judging the working state of the oil pumping unit according to the movement track of the observation point, processing the identified result and returning the processed result to the database; and the video inspection control unit reads the result returned by the video analysis and identification unit. By means of the technology, the working state of the oil pumping unit can be recognized, whether the oil pumping unit idles or not can be known in time without manual monitoring, and the idling oil pumping unit can be started and stopped in time.

The invention discloses a deep learning-based structured information extraction method. The method comprises the following steps of: 1) constructing large-scale flag data by adoption of remote monitoring: constructing an extractor by using the existing Wikipedia to provide remote supervision, wherein a signature and a Wikipedia information box comprise entity-related structured facts, the facts are mentioned in a free text part of an entity webpage, and sentences which express the facts in the information box are used as training data; 2) integrating prior knowledge into a structured information extraction model, wherein the prior knowledge comprises type and phrase information, and labels of words in natural language sentences are related to the foregoing words and continuous words thereof; 3) utilizing the past and future input features by using a bidirectional hidden LSTM layer, and inputting each sentence into a bidirectional LSTM model in a sequence form; and 4) finally outputting a marker sequence, wherein the marker sequence comprises true or false markers and has a length same as the length of the input word sequence.

The embodiment of the invention relates to a video detection method and device. The method comprises the steps of monitoring a video played in a terminal and judging whether a target video frame containing a video information box exists in the video or not; obtaining position information of an additional information box in the target video frame if the target video frame containing the video information box exists in the video, wherein the video information box and the additional information box are contained in the target video frame at the same time; and extracting content information in the additional information box according to the position information of the additional information box. The additional information box is determined by detecting the video information box loaded in the video and the position information of the additional information box is obtained, so when it is detected that the content information in the additional information box is not preset content information, effective measures can be taken and illegal content information displayed in the additional information box can be timely prevented.

The embodiment of the present invention discloses a shift scheduling method and device, wherein the method includes: establishing a shift scheduling display interface based on time and place of use; The dragged information to be scheduled generates a schedule information box; according to the position corresponding to the first drag operation, the schedule information box is placed on the schedule display interface. In the embodiment of the present invention, during the scheduling process, the scheduling personnel can complete the scheduling by dragging and dropping, which makes the scheduling process more convenient, and can also clearly check the scheduling situation of each place of use, and improves the scheduling efficiency. Under the circumstances, the rationality of the shift scheduling of each place of use is guaranteed.

Techniques and systems are provided for processing video data and processing one or more media format files. For example, video data can be obtained and processed to generate a media format file that includes the video data and a restricted scheme information box associated with the video data. The restricted scheme information box includes a plurality of scheme type boxes that identify a plurality of restricted scheme types required for processing the video data. In another example, the media format file including the video data and the restricted scheme information box associated with the video data is obtained. The media format file can be parsed, and the video data can be processed according to the plurality of restricted scheme types identified by the plurality of scheme type boxes.

The present invention proposes a dynamic detection circuit for advertising audiences and related detection devices and methods. The dynamic detection circuit for advertising audiences detects the number of outdoor advertising audiences, including: receiving detection request information frames sent by other wireless communication devices; recording The receiving time of the probe request message box; check the frame field configuration of the probe request message box; assign the probe request message box to the In the corresponding data group, and ensure that the information box field configurations of all probe request information boxes in the same data group are the same; and calculate the outdoor advertising audience passing through a specific area within a specific period of time based on the total number of established data groups Quantity estimates. Even if the wireless communication device uses the random source address mechanism to send the detection request message frame, the aforementioned method can still effectively identify the number of wireless communication devices, and then estimate the number of outdoor advertising audiences.

The present invention relates to the field of association methods, in particular to an event association method, device and storage device. The association method includes the following steps: receiving the content information of the event input by the user, and generating a first information box; generate a second information frame; receive the user's instruction for associating between different first information frames and second information frames; record the association relationship between different first information frames and second information frames that the user chooses to associate; Displays associations to user output. The first information frame is generated from the content information, the second information frame is generated from the behavior information, and then a correlation relationship is formed between the first information frame and the second information frame. Subsequent users can clearly understand the event through intuitive correlation. The correlation between each piece of information helps the user to have a clear understanding of the event and facilitates the user to sort out the memory.

The present invention provides a method for extracting relation triples from Baidu Encyclopedia based on rules and remote supervision. The method mainly adopts the method based on rules and regular expressions to extract relation triples for structured texts in information sets such as information boxes. groups, and these triples can then be used as inputs to long-range supervision algorithms. For the unstructured text with scattered information, on the one hand, the present invention extracts a small part of relation triples by writing simple, accurate and obvious rules; As the input of the remote supervision algorithm, mark all sentences containing head and tail entities in the body text, classify them according to the relationship, train the classifier, and then apply the classifier to other sentences in the body text, thereby discovering more Triad.

Methods, apparatuses, and computer program products for signaling and storing compressed point clouds are described. Subsample entries associated with a sequence of subsamples within a sequence of samples may indicate whether the sequence of subsamples is individually encapsulated in a track without other subsamples or additional header data. Subsample entry types may be indexed at a track-level subsample description box. The point cloud compression coded bitstream component types may be signaled by including corresponding point cloud unit header information in fields related to codec-specific parameters of a track level subsample description box. The subsample information box may indicate a subsample entry index for a corresponding subsample. A flag in such an information box may indicate the presence of a sub-sample description entry index. The description index box may contain a sub-sample description entry index in the same container as the sub-sample information box. The track segment header frame may include a sub-sample description entry index applied to samples of the track segment.

The invention discloses a method for extracting Chinese entity attributes. The method includes extracting the text of Baidu Encyclopedia entry pages, using entry tags to filter the pages, using the data in the information boxes in the remaining pages to perform remote labeling to obtain training data, and analyzing the training data. Carry out word segmentation and generalization, convert the generalized training data into word vectors, and then get the classification results after passing through the classifier, and fill them into the attribute slots of the corresponding categories; this Chinese entity attribute extraction method does not require manual definition of features and Other additional resources benefit from the fact that the bidirectional LSTM model can use previous information and future information for feature learning from the positive and negative directions of the sentence, and has a higher accuracy rate in the entity attribute extraction task.