A geological map intelligent recognition method based on picture chapter semantics

Abstract

The application provides a geological map intelligent identification method based on picture chapter semantics, which comprises the following steps: determining the business classification of the picture according to the picture caption sentence; and identifying the relationship of the objects in the picture according to the relationship defined by the picture caption sentence. The picture caption and the picture are analyzed as a pair of inside and outside, and the chapter analysis of the picture caption is used to correspond to the chapter semantic analysis of the picture. Through the chapter semantic analysis of the picture, the business classification and the object relationship identification of the picture are realized, the context information of the picture is recovered, the automatic identification of the literature picture is realized, the foundation for the engineering application of the picture is laid, and the requirements of the user are met.

Description

Technical Field

[0001] This invention relates to the field of exploration and development technology, and in particular to an intelligent recognition method for geological maps based on image chapter semantics. Background Technology

[0002] In the integrated application of exploration and development, for geological graphic files such as structural maps, seismic profiles, reservoir profiles, integrated logging charts, well logging curves, well-connected seismic profiles, and sedimentary facies maps in the literature, it is necessary to identify the business category and object of the map in order to establish a petrochemical business intelligent application centered on the research object. To this end, it is necessary to determine the position of the image in the literature to determine its business framework, and at the same time, identify the text in the image to determine the subject associated with the image.

[0003] In existing technologies, the business hierarchy of images is not recorded. Images in existing documents are directly converted from .rel files based on the structure of Word .docx files. The image names are just numbers, and the captions are not converted, thus losing the semantic meaning of the images. In addition, since there are no text descriptions for the images, it is impossible to determine the position of the image in the document's structure, and therefore it is impossible to assign accurate business classification information to the image.

[0004] The relationships between text in an image cannot be identified. While text in an image can be identified and its objects can be assigned meaning through an object table, the relationships between objects in the image cannot be identified. This is because even a person cannot identify the relationships between elements in an image without any external reference information, as the corpus of relationships in the image cannot be labeled. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide an intelligent geological map recognition method based on image chapter semantics to overcome or at least partially solve the above problems.

[0006] This invention provides an intelligent geological map recognition method based on image chapter semantics, comprising:

[0007] Determine the business category of the image based on the caption;

[0008] Identify the relationships between objects in an image based on the relationships defined in the captions.

[0009] Optionally, the intelligent recognition method further includes: a geological map intelligent recognition technology based on image chapter semantics, including: format conversion, image location recognition, image name recognition, image file recognition, image business classification, and image saving.

[0010] Optionally, determining the business category of an image based on its caption specifically includes:

[0011] Statement to retrieve the mirror image of an image;

[0012] The chapter structure containing the caption is taken as the chapter structure of the image;

[0013] The activity described in the caption is used as the business activity described in the image to identify the business type.

[0014] Optionally, identifying the relationships between objects in the image based on the relationships defined in the captions specifically includes:

[0015] Identify a sequence of objects in an image;

[0016] The relationships between sequences of objects in an image are determined based on a predefined semantic framework of captions.

[0017] Optionally, the format conversion specifically includes: converting PDF and doc files into docx files, which are used to determine the business type and caption name of the image based on its position in the docx file.

[0018] Optionally, the image location recognition specifically includes:

[0019] Identify the position of an image in a docx file according to the paragrath sequence;

[0020] By combining the document.xml file of the docx file with the docx module, the location of the image can be positioned.

[0021] This invention provides an intelligent geological map recognition method based on image discourse semantics, comprising: determining the business classification of the image based on the caption; and identifying the relationships between objects in the image based on the relationships defined in the caption. The caption and image are analyzed in pairs, with the discourse analysis of the caption corresponding to the discourse semantic analysis of the image.

[0022] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1This is a flowchart of a method for intelligent recognition of geological maps based on image chapter semantics according to the present invention;

[0025] Figure 2 This is a roadmap for the intelligent geological map recognition technology based on image chapter semantics, as described in this invention.

[0026] Figure 3 This is a schematic diagram illustrating the location image determination process based on the XML file in the docx file according to the present invention.

[0027] Figure 4 This is a schematic diagram of image name recognition disclosed in this invention;

[0028] Figure 5 This is a schematic diagram of the image chapter structure disclosed in this invention;

[0029] Figure 6 This is a schematic diagram illustrating image saving as disclosed in this invention.

[0030] Figure 7 This is a schematic diagram of the image text recognition disclosed in this invention;

[0031] Figure 8 This is a schematic diagram of the text file in the image disclosed in this invention;

[0032] Figure 9 This is a schematic diagram of the object dictionary disclosed in this invention;

[0033] Figure 10 This is a schematic diagram of the image object relationship recognition disclosed in this invention.

[0034] Attached label: Format conversion 1, Image location recognition 2, Image name recognition 3, Image file recognition 4, Image business classification 5, Image saving 6, Image reading 7, Image text recognition 8, Image object verification 9, Image object relationship recognition 10. Detailed Implementation

[0035] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0036] The terms "comprising" and "having," and any variations thereof, in the specification, embodiments, claims, and drawings of this invention are intended to cover non-exclusive inclusion, such as including a series of steps or units.

[0037] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0038] like Figure 1 As shown, a method for intelligent recognition of geological maps based on image chapter semantics includes:

[0039] Step 100: Determine the business category of the image based on the caption;

[0040] Step 200: Identify the relationships between objects in the image based on the relationships defined in the captions;

[0041] Step 300: Intelligent recognition technology roadmap for geological maps based on image chapter semantics.

[0042] Determining the business category of an image through captions specifically includes:

[0043] Images are treated as another way of expressing sentences. In the NLP process, the images are not processed first. Instead, the image captions, which are mirror objects of the images, are processed first. The chapter structure in which the image captions are located is taken as the chapter structure of the images, and the activities described by the image captions are taken as the business activities described by the images, in order to identify the business type of the images.

[0044] Identifying the relationships between objects in an image through the relationships defined by the captions specifically includes: identifying the sequence of objects in the image and determining the relationships between objects in the image based on the semantic framework of the predefined captions.

[0045] like Figure 2 As shown, the intelligent geological map recognition technology roadmap based on image chapter semantics includes format conversion 1, image location recognition 2, image name recognition 3, image file recognition 4, image business classification 5, image saving 6, image reading 7, image text recognition 8, image object verification 9, and image object relationship recognition 10. It is divided into two main modules: one is chapter analysis, which is used to determine the business type of the image, and the other is object and object relationship recognition.

[0046] Format conversion 1 refers to converting PDF or doc files to docx files. The purpose is to determine the business type and caption name of the images by using their position in the docx file.

[0047] PDFs are converted to docx files using the converter in pdf2docx.

[0048] For doc files, the win32com module can be used for conversion.

[0049] After format conversion, all files are uniformly docx files.

[0050] Image location recognition refers to identifying the position of an image within a docx file according to the paragraph sequence. Since the python-docx module does not provide image locations, it combines the document.xml file of the docx file with the docx module to locate the image position. The conversion result is all docx information indexed by XML paragraphs, including the location information of all text, images, and tables, such as... Figure 3 As shown, the mixed sorting is based on the XML index, and the text index is based on the docx module index.

[0051] Image recognition 3 determines the image's position by analyzing features of the context sentence, such as the absence of a period or the presence of images and numbers in the first five characters. For example... Figure 4 As shown, if there is no name, name it according to the sequence number; if the name of the figure is identified in the preceding or following text, name it with the figure name.

[0052] Image file recognition 4 refers to identifying image files saved in the correct order within a docx file. This is done by obtaining the corresponding image file using its sequence number within the docx file and then saving it with the appropriate name. This is achieved using the `document` class of the docx module.

[0053] Image category 5 refers to the hierarchical structure of the chapter / article where the image is located, which is also considered a category. The hierarchical relationship of the headings is read using the docx module; the string from heading1 to the nearest heading1 is the image's category. If the category name differs from the chapter name, a name conversion is required. The identified hierarchy is as follows: Figure 5 As shown.

[0054] Image saving 6 refers to saving the image to a specified file directory, such as... Figure 6 As shown, the chapter recognition of images in the PDF file is now complete. All images containing business information are stored in a designated directory. Next, we will proceed to recognize the objects in the images within the image directory.

[0055] Image reading 7 refers to accessing all images in the image directory to identify object relationships within the images.

[0056] Image text recognition (8) refers to recognizing text within images, achieved by calling Baidu's open-source OCR module. The image recognition results are displayed as follows: Figure 7 As shown, the corresponding txt file for each image is saved in the same folder as the image, such as... Figure 8 As shown.

[0057] Image object verification 9 refers to filtering the recognized text using a known object dictionary. The filtered text represents the objects; the remaining text is not relevant to the business logic. The object dictionary is as follows: Figure 9 As shown.

[0058] Image object relationship identification refers to determining the relationships between image objects based on the relationships defined in the image captions. For Figure 7 The diagram shows that all wells are adjacent to each other with the same position. Therefore, the identified objects are considered to be adjacent, as shown in the diagram. Figure 10 As shown.

[0059] In summary, by identifying the relationships between objects in an image, we have completed the business classification and identification of images and the recognition of object relationships in the image.

[0060] Beneficial effects: By performing semantic analysis on image texts, business classification and object relationship recognition of images were achieved, the contextual information of images was restored, and automatic recognition of document images was realized, laying the foundation for the engineering application of images and meeting user requirements.

[0061] The above specific embodiments further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent recognition of geological maps based on image chapter semantics, characterized in that, The intelligent recognition method includes: Determine the business category of the image based on the caption; Identify the relationships between objects in an image based on the relationships defined in the captions.

2. The intelligent recognition method for geological maps based on image chapter semantics according to claim 1, characterized in that, The intelligent recognition method also includes: a geological map intelligent recognition technology route based on image chapter semantics, including: format conversion, image location recognition, image name recognition, image file recognition, image business classification, and image saving.

3. The intelligent recognition method for geological maps based on image chapter semantics according to claim 1, characterized in that, The process of determining the business category of an image based on the caption text specifically includes: Statement to retrieve the mirror image of an image; The chapter structure containing the caption is taken as the chapter structure of the image; The activity described in the caption is used as the business activity described in the image to identify the business type.

4. The intelligent recognition method for geological maps based on image chapter semantics according to claim 1, characterized in that, The process of identifying the relationships between objects in an image based on the relationships defined in the captions specifically includes: Identify a sequence of objects in an image; The relationships between sequences of objects in an image are determined based on a predefined semantic framework of captions.

5. The intelligent recognition method for geological maps based on image chapter semantics according to claim 2, characterized in that, The format conversion specifically includes converting PDF and doc files into docx files, which is used to determine the business type and caption name of the image based on its position in the docx file.

6. The intelligent recognition method for geological maps based on image chapter semantics according to claim 2, characterized in that, The image location recognition specifically includes: Identify the position of an image in a docx file according to the paragrath sequence; By combining the document.xml file of the docx file with the docx module, the location of the image can be positioned.

Images