21 results about "Annotation Type" patented technology

The invention discloses a training data set processing method and a freight rate file analysis method. Relates to the field of data processing, and the method comprises the steps: obtaining air transportation price bill data, obtaining a freight rate data interface document used by an aviation institution, and extracting basic transportation price bill data from the freight rate data interface document; the air transportation price bill data is disassembled to obtain M groups of freight rate data, the M groups of freight rate data are labeled to obtain labeled freight rate data, and the labeled freight rate data at least comprises a rule type, a general rule type and an annotation type; a data set used for model training is generated according to the marked freight rate data and the basic transportation price bill data, and the model data set comprises a document set and a data set associated with the document. Through the method and the device, the technical problems of high cost, low efficiency and limited accuracy in manual marking of the training data set in the related technology are solved.

The embodiment of the invention discloses a data annotation method and device, equipment and a medium. According to the scheme, the method comprises the steps of obtaining to-be-annotated data; obtaining to-be-labeled labeling type information for the to-be-labeled data; the annotation type information represents the type of annotation information which can be annotated by the to-be-annotated data; determining an annotation processing module corresponding to the annotation type information according to the annotation type information; the annotation processing module has a data annotation function of annotating annotation information corresponding to the annotation type information; and carrying out annotation processing on the to-be-annotated data by utilizing the annotation processing module to obtain annotated data containing annotation information.

A computer-implemented method for annotating a 3D modeled object representing a CAD mechanical part or assembly of parts, two or more views of the 3D modeled object being displayed and at least one of the views being compliant with a view of a technical drawing. The method includes obtaining a 3D scene and displaying an orientation-free view of the 3D modeled object, at least one fixed view of the 3D modeled object. The method also includes creating, upon user action, an annotation by selecting at least one geometrical element on at least one of the views of the 3D modeled object and an annotation type. The method further includes displaying the created annotation on at least the orientation-free view.

The application belongs to the field of artificial intelligence and educational interactive technology, and discloses an intelligent Q matrix generation method, system and terminal fusing question semantic information, which uses BiLSTM+Transformer to obtain mathematical knowledge points in a data set, and uses CRF to correct the annotation type of each word to output the mathematical knowledge points; uses BiGRU+Attention mechanism to extract the relationship of the knowledge points, and outputs the relationship between each knowledge point pair; uses the relationship between the knowledge points to design a redundant knowledge point reduction algorithm, reduces the redundant knowledge points through the reduction algorithm, and generates a Q matrix using the knowledge point set after reduction. The application can achieve a relatively high knowledge point extraction rate by using the Transformer+BiLSTM+CRF method, and can output the relationship between the knowledge point pairs more accurately by using the BiGRU+Attention method.

The invention provides a compiling method, a compiling device and computer equipment. Comprising the steps of generating a byte code file and first type information according to a source code; determining second type information according to the byte code file; combining the first type of information and the second type of information in a compiling period to obtain a first composite structure type; at least two data structures are determined according to the first composite structure type, and when the first composite structure type is matched with one or more of the at least two data structures, the phi node is constructed by compiling the first composite structure and one or more of the at least two data structures. According to the compiling method, the compiling device and the computer equipment provided by the invention, the composite structure type is obtained in a mode of mutual complementation of the labeling type information and the sampling type information, and multiple paths are established for constructing the phi node, so that the probability of de-optimization is reduced.

The invention discloses a remote sensing image labeling system and method. A basic support module of the labeling system provides visualization and editing functions of image and vector data and project management and system database management functions; the standard labeling process module provides a unified labeling process for various labeling types; the sample batch generation and conversion module provides various forms of sample batch generation and conversion functions; the labeling method comprises the following steps: firstly, creating a labeling project, setting a classification system and a standard labeling process mode, determining a standard labeling process according to the selection of the classification system and the selection of the labeling mode, and completing project construction and initializing the project by the system according to the selection; the system can carry out semi-automatic labeling, and sample batch generation and format conversion can be realized. According to the invention, sample data required by remote sensing image deep learning is efficiently manufactured in a standardized manner, all sample types of current remote sensing deep learning are covered, and the system and the method are a full-stack system and method.

This invention relates to a continuous image annotation method, comprising the following steps: S10. Determining the annotation type of the continuous image and obtaining a variable group corresponding to the annotation type; S20. Extracting multiple frames of images from the continuous image at intervals, and the annotation engineer annotating each object to be annotated in each frame of the image according to the determined annotation type, and determining the number of annotations; S30. Generating a preprocessing model for each variable in the variable group based on the number of annotations; S40. Obtaining the state of the above variables in the remaining frames of the continuous image through the preprocessing model; S50. Describing the state of the label together with all the variable states in each frame of the image, thus completing the annotation of the continuous image. This invention only requires annotating a small number of frames in the continuous image, and then quickly generating a preprocessing model. The preprocessing model will automatically complete the annotation type of the object to be annotated in the remaining frames of the image. Afterwards, the annotation engineer confirms the annotation, requiring no adjustment or only minor adjustments, greatly improving the annotation efficiency.

The invention discloses an annotation test point generation method and device and electronic equipment. The method comprises the following steps: acquiring incremental code information; annotation information in the incremental code information is recognized, an annotation list is generated according to the annotation information, and the annotation list at least comprises annotation names and parameter names and parameter types of the annotation information; a test point sorting list is determined according to the annotation list, annotation test points are generated according to the test point sorting list, and the test point sorting list is used for reflecting the upper and lower boundaries of the type of the annotation information, the upper and lower boundaries of the business meaning and the positive and negative equivalence classes. According to the method and the device, the technical problems that the annotation logic can be tested through a reflection mechanism and an interceptor mechanism due to the fact that related technologies focus on the execution stage of the annotation test, but the test point generation efficiency is low in the test preparation stage, especially in the face of large-scale code change and diversified annotation types are solved.

In particular, the present disclosure relates to a computer-implemented method for annotating a 3D modeled object representing a CAD machine part or part assembly, two or more views of the 3D modeled object being displayed, and at least one of the views complying with view specifications of a technical drawing. The method includes obtaining a 3D scene and displaying an undirected view of the 3D modeled object and at least one fixed view of the 3D modeled object. The method further includes creating an annotation by selecting at least one geometric element and an annotation type on at least one of the views of the 3D modeled object in response to a user operation. The method further includes displaying the created annotation at least on the undirected view.

The application discloses a kind of PDF document mark content extraction method and system, belong to PDF document mark content extraction technical field, including starting PDF document mark content extraction process, initialization extraction parameter and target document configuration;All pages of target PDF document are traversed, and various types of annotation data contained in page are obtained;According to the classification preprocessing of the annotation data obtained according to the annotation type, the application realizes the one-key automatic extraction of different PDF document mark content, is suitable for multiple annotation types, improves information extraction efficiency;Effectively guarantee the accuracy of mark content restoration and the real restoration of layout, support document archiving and information summarizing demand;Clear structure, processing is flexible, with high scalability, can be conveniently applied to OA, education, knowledge management and various kinds of scenes.

Embodiments described herein relate to a process for generation of annotation-accessible library spectral content. A system can comprise a memory that stores, and a processor that executes, computer executable components. The computer executable components can comprise an identifying component that identifies chemical compound data describing a chemical compound, and a generating component that generates an identifier from the chemical compound data based on an annotation type of the compound data as compared to an annotation ranking schema.

The invention relates to the technical field of machine learning, in particular to a labeling and training method based on an integrated platform, and the method comprises the steps: creating a labeling task in a labeling task management module, and synchronously generating an associated training task framework; the training task framework pre-binds an annotation type and an algorithm model type, and inherits a material classification system of an annotation task; and in response to the annotation completion instruction, automatically converting the annotation data and the original image into a target algorithm format corresponding to the algorithm model type, and injecting the target algorithm format into a data storage library of the algorithm model type. According to the labeling and training method based on the integrated platform, material labeling and model training are achieved on the same platform, tedious processing conversion is not needed, the labeling and model training efficiency is improved, and then the development efficiency and performance of an artificial intelligence model are improved.

Devices, systems, and techniques for generating a document transcription of a document image. In at least one embodiment, one or more neural networks generate a document transcription of a document image according to a configurable combination of annotation types input into the one or more neural networks. The document transcription can contain respective annotations of the annotation types for corresponding parts of the content included in the document transcription.

Systems and methods are provided for efficiently building an object detection learning model for an unlabeled pool of images. A recommendation engine automatically recommends an annotation type for the images in the unlabeled pool based on previous object detection and an updated mean average precision of the model, where the mean average precision represents the performance of the model.

The disclosure notably relates to a computer-implemented method for annotating a 3D modeled object representing a CAD mechanical part or assembly of parts, two or more views of the 3D modeled object being displayed and at least one of the views being compliant with a view of a technical drawing. The method comprises obtaining a 3D scene and displaying an orientation-free view of the 3D modeled object, at least one fixed view of the 3D modeled object. The method also comprises creating, upon user action, an annotation by selecting at least one geometrical element on at least one of the views of the 3D modeled object and an annotation type. The method further comprises displaying the created annotation on at least the orientation-free view.

The invention discloses an annotation-based PDF (Portable Document Format) electronic signature and signature verification method, and belongs to the technical field of electronic signature and information security. The signature method comprises the following steps: loading a to-be-signed PDF document to obtain a first document object model; creating an annotation object, and setting a sub-type key value of the annotation object as a user-defined signature annotation type; embedding a visual signature identifier in the appearance of the annotation object, and setting a protective flag bit; constructing a signature dictionary, reserving a signature value position, and determining a byte interval of signature calculation; associating the signature dictionary to the V attribute of the annotation object; calculating an original hash abstract of the to-be-signed data according to the byte interval; performing digital signature on the original hash abstract by using a signature party private key; writing the byte interval and the signature value into a signature dictionary; and performing incremental updating operation on the PDF document to complete signature. According to the application, the electronic signature is converted into the annotation object and the user-defined identifier, so that the interactive operation of the user can be blocked on the physical level without additional encryption.

This invention discloses an adaptive optimization method for the layout of railway subgrade drawings with multiple types of annotations, comprising: S1, establishing structural semantic expression information for all annotations in the CAD drawing; S2, identifying the annotation type of the CAD drawing, reading its structural semantic expression information, and obtaining the adjustment rules for each annotation; S3, calculating the dynamic adjustment coefficient of the overlap rate of all drawing annotations; S4, calculating the repulsive force vector between any two annotations, the attractive force vector between each annotation and its annotated object, and the overlapping repulsive force vector experienced by overlapping annotations; S5, calculating the resultant force vector experienced by each annotation, and moving the position of the annotation according to the direction of the resultant force vector and the annotation adjustment rules; S6, repeating S3 to S5 until the optimized railway subgrade CAD drawing is obtained. This method ensures that the overall layout of the CAD drawing is compact and aesthetically pleasing while guaranteeing that the annotations do not overlap.

To provide a computer-implemented method for annotating a 3D modeled object representing a CAD mechanical part or assembly of parts.SOLUTION: A method for displaying two or more views of a 3D modeled object, at least one of the views being compliant with a view of technical drawings, comprises obtaining a 3D scene and displaying an unoriented view of the 3D modeled object and at least one fixed view of the 3D modeled object, creating an annotation by a user action by selecting at least one geometric elements on at least one of the views of the 3D modeled object and a type of annotation, and displaying the created annotation on at least the unoriented view.SELECTED DRAWING: Figure 1

A processor comprises circuits causing a neural network 320 to generate a document transcription 354 of a document image 321 according to a configurable combination of annotation types 325 provided wi

This invention discloses a method and system for professional annotation and drawing generation of BIM models based on grid analysis. The method includes: Step 1: Selecting a profession and its corresponding professional rule set through a profession selection module; Step 2: Obtaining all model data in the current two-dimensional view of the BIM system and performing corresponding data preprocessing according to the annotation type of the profession selected in Step 1; Step 3: The annotation generation module automatically generates annotations for specific components according to the selected profession; Step 4: Automatically generating drawings with professional annotations. This invention relates to the field of building information model data processing and engineering drawing automation technology, and can solve the problems of existing technologies that cannot perform automatic annotation based on grid analysis and cannot perform customized annotation for different professions.

Apparatuses, systems, and techniques to generate a document transcription of a document image. In at least one embodiment, one or more neural networks generate a document transcription of a document image according to a configurable combination of annotation types input to the one or more neural networks. The document transcription may include respective annotations of the annotation types for corresponding portions of content included in the document transcription.