29 results about "Page footer" patented technology

The embodiment of the invention discloses a medicine production quality management specification GMP file generation method and related products. The method comprises the following steps: obtaining a content page of a target GMP file drawn by a dragger by using a content template, metadata information of the target GMP file and approval information for the content page; obtaining an endorsement page template and a header and footer template; generating an endorsement page based on the approval information and the endorsement page template; based on the metadata information and the header and footer template, generating a header and a footer; and synthesizing the target GMP file based on the content page, the endorsement page, the page header and the page footer. According to the embodiment of the invention, the signing efficiency of the GMP file can be improved.

The invention provides a structural decomposition and information identification method of multi-modal data, a medium and equipment, and the method comprises the steps: obtaining to-be-processed multi-modal literature data, the multi-modal literature data being documents or pictures, the types of the documents including word documents and PDF (Portable Document Format) documents; the method comprises the following steps: converting to-be-processed multi-modal literature data into to-be-processed literature data in an image form, preprocessing the to-be-processed literature data to obtain input image data, inputting the input image data into a field fine-tuning DETR model, identifying a logic region category of the input image data through the field fine-tuning DETR model, obtaining a region category identification result, and outputting the region category identification result. The logic region category comprises a title region, an author region, an abstract region, a text region, an illustration region, a table region, a formula region, a footer region and a reference region; and carrying out differential information extraction on each region category identification result to obtain information corresponding to each region category identification result so as to realize accurate identification of information in the multi-modal data.

The invention provides a text search method and device, computer equipment and a computer readable storage medium, and the method comprises the steps: obtaining a target PDF document and a text search request of the target PDF document, the document search request comprising a to-be-searched text; determining a document type of the target PDF document based on a document type identification rule; determining a document conversion model according to the document type of the target PDF document, and performing document conversion processing on the target PDF document based on the determined document conversion model to obtain a first text corresponding to the target PDF document; determining a header text and / or a footer text in the first text, and deleting the header text and / or the footer text to obtain a second text; and determining a target text matched with the to-be-searched text in the second text, and outputting text information and position information of the target text. According to the method and the device, the text searching and positioning precision of the PDF document is improved.

The invention discloses a PDF (Portable Document Format) file layout identification method based on mask processing, and aims to realize accurate identification of various layout elements such as titles, texts, page headers, page footers, tables and pictures in PDF files. According to the method, characters and coordinates in a PDF file are analyzed through MuPDF, different color masks are adopted to replace texts and punctuation marks respectively, original content of a non-text area is reserved, a page picture subjected to mask processing is generated, a training data set is constructed based on the page picture, training is conducted through a target detection model (such as YOLO), and a high-precision layout recognition model is obtained. According to the method, the accuracy and the automation level of PDF layout recognition are effectively improved, and the method has a wide application prospect.

The application relates to a contract document difference detection method and device based on an ICR character matrix. The contract document difference detection method based on the ICR character matrix comprises the following steps: extracting text data of an original contract document and a comparison contract document by using ICR technology; dividing the contract document into four parts of a header, a body, a footer and table text, and respectively splicing into long strings; and using a two-section difference detection algorithm to sequentially query difference points from the body and the table text, generating a text comparison result, and returning position information of the difference point type, related strings and related characters. Compared with existing document comparison tools, the application can perform difference detection on scanned copies or even photo-form contract electronic files; when designing comparison rules, the characteristics of structural texts such as headers, footers and tables of the contract are considered, and differences such as text line breaks and page changes that do not affect semantics are ignored, so that a difference comparison result meeting human expectations can be fed back.

The invention provides a method for extracting professional domain entity pair information based on same-category unstructured documents. Comprising the following steps: counting expression characteristics of each professional domain entity pair to be extracted in a document, determining extraction rules of texts and information, and summarizing the rules in a template; dividing lines, watermarks, directories or invalid pages, page headers and page footers in front of a preamble page in the document are removed, and text content is extracted according to a text range and comprises text content and table content; a filtering type three-layer combination extraction model is constructed to achieve entity information extraction, the three-layer combination extraction model relates to rule-based extraction, table-based extraction, large model question and answer-based extraction schemes and sequential combination thereof, result accuracy is judged through accuracy verification, and the accuracy is fed back to a scheme layer for strategy optimization. According to the method, the problems that the generalization ability of the existing rule-based information extraction is insufficient, and machine learning and deep learning methods are lack of labeling data in professional fields can be solved.

The application discloses a kind of based on layout partition and sequence constraint's archive image automatic division method and system, belong to archive image processing technical field.The method obtains the archive image and division rule according to page sequence, determines title area, text area and footer area using layout analysis result, and carries out optical character recognition and text normalization respectively;According to the matching candidate piece class of the exclusion relationship of the text condition chain in area, cross-area containing relationship and merged area text, and combine previous page division state, last page number, consecutive page number and same volume occurrence number Sequence constraints determine each page piece name and division boundary;Finally, the corresponding relationship of page and target piece class and determination basis are output.The application can reduce the interference of whole page unrelated text, and keep the page sequence consistency of continuous archive material, suitable for automatic division and review of scanned archives.

This specification relates to file format-based transparent encryption tailored for big data. In some aspects, a method includes receiving, by one or more computing devices, a write request including a table with one or more columns to be stored in a storage device, wherein each column includes a number of pages; generating a column key for each column and a page key for each page including sensitive information; encrypting (i) each page including sensitive information with a corresponding page key and (ii) each column with a corresponding column key; generating wrapped keys for the column keys and page keys and storing the wrapped keys into a key file; storing the encrypted columns into a data file of the storage device and storing the wrapped keys in a separate key file; and storing a reference to the key file in a file footer of the data file.

The invention relates to the technical field of information processing, and provides a file sensitive information processing method and system, a computer and a storage medium, and the file sensitive information processing method comprises the steps: obtaining a first task data set, and converting a to-be-processed file into a plurality of to-be-processed file pictures; based on the second task data set, judging whether header and footer detection is started or not, and if the header and footer detection is started, detecting and generating a plurality of area detection results; whether seal detection is started or not is judged based on the third task data set, and if seal detection is started, a plurality of seal detection results are generated through detection; judging whether text detection is started or not based on the fourth task data set, and if the text detection is started, detecting and generating a plurality of sensitive text detection results; and judging whether fuzzy processing is started or not based on the fifth task data set, and if the fuzzy processing is started, generating a fuzzy file. By adopting the method, the problems of low blind state processing efficiency, inaccurate blind state and artificial leakage risk can be avoided.

The invention discloses a PDF (Portable Document Format) document title identification method, which comprises the following steps of: acquiring a character list of a PDF document, and identifying page headers, page footers, sidebar characters and watermark characters in the character list; dividing characters in the PDF document into character paragraphs to obtain a character paragraph list; dividing the character paragraphs into a plurality of character paragraph groups with the same characteristics, and screening out candidate title groups according to the character paragraph groups; and determining a title level of the candidate title group based on the candidate title group. According to the method, the PDF document title recognition efficiency and accuracy are improved, and meanwhile, the satisfaction degree of a user on title recognition is improved.

The invention relates to the technical field of document merging, in particular to a document merging system and method, equipment and a medium, the system comprises an interaction and configuration module, a document analysis and routing module, a document merging module and a verification and repair module, and the verification and repair module is integrated in the system to be in communication connection with the document merging module. And the verification and repair module is used for verifying and repairing the page header and the page footer of the output document, so that the segments, the page header and the page footer of all the documents in the output document are kept in the original format before merging. By means of the mode, the multi-format documents can be intelligently, automatically and losslessly combined in batches, and it is ensured that the format and header and footer completeness of each document are reserved.

The invention provides a text data denoising method for a mining industry large model. The method comprises the following steps: determining a typesetting state of a home page in a knowledge document; wherein the knowledge document is a document in a mine industry knowledge base; determining page header coordinates of the knowledge document, page footer coordinates of a home page and page footer coordinates of a non-home page in the knowledge document; cutting the home page according to the header coordinates and the footer coordinates of the home page in combination with the typesetting state of the home page, so that the cut home page only contains the text; and cutting the non-home page according to the header coordinates and the footer coordinates of the non-home page, so that the cut non-home page only contains the text. According to the method, noise data in the mine industry knowledge base can be eliminated, the quality and accuracy of the coal industry knowledge base are improved, the quality of the coal industry high-quality text data set is improved, and the influence of the noise data on the model performance is reduced.

This specification relates to file format-based transparent encryption tailored for big data. In some aspects, a method includes receiving, by one or more computing devices, a write request including a table with one or more columns to be stored in a storage device, wherein each column includes a number of pages; generating a column key for each column and a page key for each page including sensitive information; encrypting (i) each page including sensitive information with a corresponding page key and (ii) each column with a corresponding column key; generating wrapped keys for the column keys and page keys and storing the wrapped keys into a key file; storing the encrypted columns into a data file of the storage device and storing the wrapped keys in a separate key file; and storing a reference to the key file in a file footer of the data file.

The invention discloses a bidding document title structure extraction method based on an OCR engine. The bidding document title structure extraction method comprises the steps that a bidding document in a pdf format is converted into an image file; performing text content recognition on the bidding document image file by utilizing OCR (Optical Character Recognition); sorting bidding document character recognition results according to the text block position information of the OCR recognition results, and obtaining bidding document text contents; performing title structure extraction on the paragraph content of the bidding document text; optimizing the title paragraphs by using header and footer rules and title features; the method solves the problems of low efficiency and strong manual dependence in the prior art, can quickly and accurately extract the title information of the document, improves the title extraction efficiency and accuracy, and is suitable for various types of documents.

The invention relates to the technical field of computers, and provides a file content management method and device, electronic equipment, a storage medium and a computer product.The method comprises the steps that a to-be-processed file is received, structure recognition and content recognition are conducted on content blocks in the to-be-processed file, and content block structure information and text information are obtained; the content block comprises at least one of a text block, an image block, a table block and a header and footer block; performing content filling on the text information according to the content block structure information to obtain at least one filling content block; and after each filling content block is rendered, outputting according to a content block layout mode of the to-be-processed file. The reliability and readability of file content management can be improved.

The embodiment of the invention provides a document header and footer identification method and device, and the method comprises the steps: obtaining an electronic document, and determining a to-be-identified document page in the electronic document, the to-be-identified document page being any one document page in the electronic document; identifying all text lines in the to-be-identified document page, and determining respective text line features of all the text lines; traversing the text line features of all the text lines, and executing the following steps: identifying the text lines belonging to the page header, and endowing the text lines belonging to the page header with structural elements with page header attributes; and / or identifying the text line belonging to the footer, and endowing the text line belonging to the footer with the structural element of the footer attribute. Therefore, identification of the page header and the page footer can be realized based on the single document page, less time is consumed, and the identification speed is improved.

The invention relates to the technical field of document processing, and discloses a layout and reading sequence identification method, system and device and a storage medium, and the method comprises the steps: generating a training data set, training a column document layout identification model, and preprocessing an input document image; the output document image comprises columns, image blocks, table blocks, page annotations, label frames of all the universal layout elements, and the positions and confidence degrees of the corresponding label frames, wherein the file image comprises the columns, the image blocks, the table blocks, the page annotations and the universal layout elements; performing post-processing optimization on the identified tab boxes, grouping non-column tab boxes in the page based on the positions of the column tab boxes to form a plurality of grouping blocks, performing local sorting on internal elements of the grouping blocks, and performing global sorting on all the grouping blocks and non-grouped independent tab boxes to generate a global sorting result; and replacing the grouping blocks in the global sorting result with a local sorting result, and endowing all elements with continuous sequence identifiers so as to output a document reading sequence. According to the invention, identification efficiency and accuracy can be improved.

The present application belongs to the technical field of PDF file analysis, and provides a PDF to Word header and footer extraction method, device and computer readable storage medium, the method comprising the following steps: step S1: pre-processing the PDF file, completing the nesting, layout processing and element sorting of characters and graphics, obtaining an ordered PDF preprocessing file for each page content; step S2: page splitting: splitting the PDF preprocessing file into odd page files and even page files; step S3: identifying and demarcating the effective area of the header and footer of each page file; step S4: in the effective area, matching the common elements of each page from the second page, and judging the header and footer type of the PDF page; step S5: judging the header and footer type of all pages; step S6: generating the header and footer information of each page according to the header and footer type of all pages, and realizing header and footer extraction. The present application can realize efficient identification of headers and footers.

The invention discloses a table picture OCR (Optical Character Recognition) method and tool using an Excel template, and the method comprises the following steps: S1, obtaining a target Excel template file, and analyzing printing information in the target Excel template file, including page headers, page footers, page numbers and edge distances; s2, performing format comparison on a printing image exported from the Excel file and an image to be identified, and determining the position of printing information in the image; s3, preprocessing the to-be-identified image, deleting a printing information area, and reserving a table main body area; and S4, performing OCR identification on the preprocessed image, and extracting a table structure and text content. According to the method, the printing information (such as page header, page footer, page number and the like) in the Excel template is analyzed in advance, the image is preprocessed before OCR recognition, the printing information area is removed, and therefore the accuracy and efficiency of OCR recognition are improved.

The utility model discloses a music stand capable of automatically turning pages. The music stand comprises a stand body, the stand body comprises a music score placing plate, a music score is placed on the music score placing plate, and through holes are densely distributed in the music score placing plate. A page turning assembly is arranged in the through hole through a fixing assembly and controlled to be triggered by a control assembly. The page turning assembly comprises a page footer turning module, the page footer turning module comprises a page footer motor fixedly arranged on the fixing assembly, a page footer gear box is fixedly arranged at the output end of the page footer motor, a page footer roller is fixedly arranged on an output shaft of the page footer gear box, and the page footer roller is arranged at the edge of the music score page. According to the device, the fixing assembly is specially arranged through the characteristic that the music stand is provided with multiple holes, so that the adaptability of the device and the installation convenience of the device are improved; the device can control the page turning assembly to work through the control assembly according to the action of a user, and therefore the function of automatically turning pages without manual operation is achieved.

The invention discloses a cleaning and construction method and system for large model text training data and a related device, and belongs to the technical field of natural language processing, and the method comprises the steps: S101, obtaining original text data from different sources; s102, utilizing the discriminant word list, and using lexical element matching and regular expression methods to perform preliminary data cleaning to obtain preliminarily cleaned text data; s103, removing page headers and page footers based on lexical element matching and semantic similarity; s104, based on the text data of which the page header and the page footer are removed, combining lexical element matching and title core recognition, and dividing multi-level titles and corresponding text blocks; s105, performing text error correction on the text block to obtain a text block subjected to data cleaning; s106, adding increment information, forming structured data, and constructing a pre-training data set; according to the text data processing and cleaning method, the data processing flow can be flexibly adjusted according to actual requirements, and the method is applied to text data of different sources and different formats.

The present disclosure provides a page turning method, apparatus, computer equipment and storage medium, wherein the method comprises: obtaining a first view corresponding to a current display page of a book, a second view corresponding to a previous page of the current display page, and a third view corresponding to a subsequent page of the current display page; in response to a triggered page turning operation, determining first area information and second area information in the first view, and third area information in a target view matching the page turning direction according to a footer position point corresponding to a page turning direction and a page turning trigger point indicated by the page turning operation; using a graphics processor, based on the first view, determining a first partial view matching the first area information and a second partial view matching the second area information; and based on the target view, determining a third partial view matching the third area information; and performing splicing rendering on the first partial view, the second partial view and the third partial view to obtain a page turning view.

The invention discloses a dynamic editable header and footer generation method and system based on chapter granularity, and relates to the dynamic editable header and footer generation method based on the chapter granularity, which comprises the following steps: S1, receiving a target document, decompressing the target document, and creating a basic header and footer template; s2, extracting chapter information and segment character positions of the document; s3, copying and renaming the basic header and footer template to obtain an updated header and footer file; and S4, repackaging all the files containing the updated header and footer into a document in a docx format.

The invention discloses a Chinese-English PDF (Portable Document Format) document translation system with consistent formats. The method comprises the following steps: identifying module categories in a Chinese PDF document; extracting various types of information in each module; on the basis of the Chinese text information and the regional information, generating an English translation by adopting a length-limited machine translation method based on a large language model; for the picture, the table and the formula module covered by the English translation, writing the picture, the table and the formula module into a region corresponding to the original Chinese PDF in an English PDF document in a picture form; for text, list, chart title, article title and section title modules, the translated English text is written into a region corresponding to the Chinese original text of the English PDF document in a paragraph form according to the finally determined font size, font and color; and for footnote, page header and footer modules, directly writing the Chinese original text into the English PDF document according to the regional information. According to the method, the format of the translation and the format of the original text are accurately aligned, and the method is particularly suitable for high-fidelity translation requirements of technical documents such as academic papers and the like.

The present invention provides a method for extracting document information, including: obtaining a document to be parsed; using a document structure parsing model to parse different elements in the document and give identification bounding boxes for the elements; for the parsed header, footer, QR code, illustration, and trademark parts, processing the images within their bounding boxes into blank images; for the parsed seal part, if the seal and printed text overlap, using an algorithm to remove the seal part and retain the text part covered by the seal, and replacing the text part after removing the seal with the original image at the seal position; extracting the printed text and handwritten text from the processed document image, and recognizing the printed text and handwritten text in the document image; combining the positions of the original table, printed text, and handwritten text in the document image to assemble the recognized text together; based on a generative language large model, designing prompt words to generate the required structured data to be extracted and outputting it in a fixed format.

The application discloses a cross-page table intelligent processing and retrieval enhancement system and method, relates to the technical field of document information processing, and comprises the following steps: performing character recognition and table area detection on input documents page by page, identifying candidate table block areas, removing header and footer texts, and generating a purified table block record set; extracting column position structures and table header token sets based on the purified table block record set, constructing cross-page continuous operators, generating cross-page determination thresholds, forming a cross-page connection relationship set, merging cross-page tables through the cross-page connection relationship set, restoring row and column structures, generating logical table records, constructing retrieval enhancement index records, and completing retrieval enhancement storage; and matching natural language queries by using the retrieval enhancement index records, performing hit verification, and outputting logical table structured representation texts and cross-page connection group page number ranges. The application realizes cross-page table structure recognition and natural language retrieval enhancement.

The invention belongs to the technical field of PDF file analysis, and provides a PDF-to-Word header and footer extraction method and device and a computer readable storage medium, and the method comprises the following steps: S1, carrying out pre-processing on a PDF file, completing nesting of characters and graphs, layout processing and element sorting, and obtaining a PDF pre-processing file with ordered contents on each page; s2, page splitting: splitting the PDF preprocessing file into odd-number page files and even-number page files; s3, identifying and delimiting effective areas of page headers and page footers of each page of file; s4, in the effective area, matching the public elements of each page from the second page, and judging the header and footer types of the PDF page; s5, judging header and footer types of all the pages; and S6, generating header and footer information of each page according to the header and footer types of all the pages, and realizing header and footer extraction. According to the method, the page header and the page footer can be identified efficiently.

To properly extract and exclude unnecessary parts such as headers and footers even when there is an error in the reading position during scanning or an OCR misjudgment.SOLUTION: An image processing device includes extraction means 901 that extracts objects from the image of each page of a document, analysis means 902 that uses an image of one page of the document as a reference page and other pages as analysis pages, and analyzes each object on the reference page and each object on the analysis page, and determination means 903 that determines whether each object on the reference page is a necessary part or an unnecessary part based on the result of the analysis means 902.SELECTED DRAWING: Figure 1

The invention belongs to the technical field of PDF analysis, and provides a paragraph alignment method and device in analysis of pdf and a medium. According to the method, an open source framework PDFBOX of Apache is utilized to read a PDF file, then characters are sorted, converted and typeset, information such as paragraphs, titles, page headers and page footers is determined, row-level splitting is performed based on paragraph contents, a fault-tolerant space is set to analyze and judge row-level contents, paragraph alignment is determined, accurate alignment of the converted document can be kept, and the accuracy of the file alignment is improved. And the method has the characteristics of high quality and easiness in editing.