Chemical single line diagram material identification method, device and equipment and storage medium

Abstract

The invention discloses a chemical single line diagram material identification method, device and equipment and a storage medium, and the method comprises the steps: building a corresponding relation model of a serial number and a line spacing proportion value in a chemical single line diagram material table, and detecting a to-be-identified table according to the corresponding relation model; when it is detected that the serial numbers of the serial number columns of the to-be-recognized table are continuous and the line spacing is within a preset proportion range, corresponding multi-line information is merged; according to the method, the table auxiliary line of the to-be-recognized table is obtained, AI recognition is performed on the to-be-recognized table according to the table auxiliary line, and the recognition result is obtained, so that various material information in the chemical single line diagram can be accurately recognized, and compared with a traditional method, manual intervention is reduced, and the workload and error risk of manual processing are reduced; the method provides more efficient and accurate support for material information management in the chemical engineering design and construction process, has important practical application value and wide market prospect, and improves the speed and efficiency of chemical single line diagram material identification.

Description

Technical Field

[0001] This invention relates to the field of chemical image recognition technology, and in particular to a method, apparatus, equipment and storage medium for identifying chemical single-line diagram materials. Background Technology

[0002] In the construction process of chemical engineering, single-line diagrams are important technical documents containing rich material information. However, existing table recognition technologies have many problems in processing material information in single-line diagrams. Traditional table recognition cannot accurately integrate relevant information when dealing with multiple merged rows in a single-line diagram, leading to chaotic material information recognition. At the same time, it is difficult to correctly segment columns for tables without dividing lines or with inaccurate division, which affects the accuracy of material information extraction. These problems seriously restrict the automation and accuracy of material information recognition in single-line diagrams, increasing the workload and error risk of manual processing. Summary of the Invention

[0003] The main objective of this invention is to provide a method, apparatus, device, and storage medium for material identification in chemical single-line diagrams. This invention aims to solve the technical problems of traditional table-based identification methods, which fail to accurately integrate relevant information when dealing with multiple merged rows in chemical single-line diagrams, leading to chaotic material information identification. Furthermore, for tables without dividing lines or with inaccurate division, it is difficult to correctly segment columns, thus affecting the accuracy of material information extraction. This restricts the automation and accuracy of material information identification in chemical single-line diagrams, and increases the workload and error risk of manual processing.

[0004] In a first aspect, the present invention provides a method for identifying materials using a single-line diagram in chemical engineering, the method comprising the following steps: Establish a correspondence model between the serial number and the row spacing ratio in the material table of a chemical single-line diagram, and detect the table to be identified based on the correspondence model. When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within the preset ratio range, the corresponding multiple rows of information are merged. Obtain the table auxiliary lines of the chemical single-line diagram to be identified, and perform AI recognition on the table to be identified based on the table auxiliary lines to obtain the recognition result.

[0005] Optionally, the step of establishing a correspondence model between the serial number and the row spacing ratio in the chemical single-line diagram material table, and detecting the table to be identified based on the correspondence model, includes: Obtain a preset number of chemical single-line graph material tables, and construct a training dataset based on the chemical single-line graph material tables; The training dataset is trained using a machine learning algorithm to obtain a correspondence model between the serial number and the row spacing ratio, and the table to be identified is detected based on the correspondence model.

[0006] Optionally, obtaining a preset number of chemical single-line graph material tables and constructing a training dataset based on the chemical single-line graph material tables includes: Obtain a preset number of chemical single-line diagram material tables, and preprocess the chemical single-line diagram material tables to obtain preprocessed tables; The text information in the preprocessed table is extracted using character recognition technology, and the serial number column in the preprocessed table is located to generate a training dataset.

[0007] Optionally, the step of training the training dataset using a machine learning algorithm to obtain a correspondence model between serial numbers and row spacing ratios, and detecting the table to be identified based on the correspondence model, includes: The training dataset is trained according to the machine learning algorithm, the sequence number and line spacing ratio in the training dataset are learned and modeled, the line spacing ratio threshold is determined, and the correspondence between the sequence number and the line spacing ratio is obtained. Based on the correspondence, establish a correspondence model between each serial number and the ratio of each row spacing, and detect the table to be identified based on the correspondence model.

[0008] Optionally, when it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, the corresponding multiple rows of information are merged, including: When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, multiple rows of information with consecutive serial numbers and row spacing within the preset ratio range are merged into a complete material information record.

[0009] Optionally, the step of obtaining the table auxiliary lines of the chemical single-line diagram to be identified, and performing AI recognition on the table to be identified based on the table auxiliary lines to obtain the recognition result, includes: Obtain detailed information from the single-line diagram of the chemical plant to be identified, and automatically generate auxiliary lines in the table based on the detailed information; The column structure of the table to be identified is determined based on the table auxiliary lines, and the table auxiliary lines are added to the single-line diagram of the chemical product to be identified to obtain the target image to be identified. The target image to be identified is input into the AI ​​recognition system, so that the AI ​​recognition system can perform column segmentation of the table in the target image according to the table auxiliary lines, extract the material information of each column, and generate recognition results.

[0010] Optionally, obtaining detailed information of the chemical single-line diagram to be identified, and automatically generating table auxiliary lines based on the detailed information, includes: Image preprocessing is performed on the chemical single-line diagram to be identified to obtain a preprocessed table image. Image analysis is then performed on the preprocessed table image according to a preset image analysis technique to obtain detailed information about the chemical single-line diagram to be identified. Based on the detailed information, table auxiliary lines are automatically drawn according to preset drawing rules.

[0011] Secondly, to achieve the above objectives, the present invention also proposes a chemical single-line diagram material identification device, the chemical single-line diagram material identification device comprising: The model building module is used to establish a correspondence model between the serial number and the row spacing ratio in the material table of the chemical single-line diagram, and to detect the table to be identified based on the correspondence model. The information merging module is used to merge multiple rows of information when it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range. The information recognition module is used to acquire the table auxiliary lines of the chemical single-line diagram to be recognized, and to perform AI recognition on the table to be recognized based on the table auxiliary lines to obtain the recognition result.

[0012] Thirdly, to achieve the above objectives, the present invention also proposes a chemical single-line diagram material identification device, the chemical single-line diagram material identification device comprising: a memory, a processor, and a chemical single-line diagram material identification program stored in the memory and executable on the processor, the chemical single-line diagram material identification program being configured to implement the steps of the chemical single-line diagram material identification method as described above.

[0013] Fourthly, to achieve the above objectives, the present invention also proposes a storage medium storing a chemical single-line diagram material identification program, wherein the chemical single-line diagram material identification program, when executed by a processor, implements the steps of the chemical single-line diagram material identification method as described above.

[0014] The proposed method for material identification in chemical single-line diagrams establishes a correspondence model between the serial number and the row spacing ratio in the material table of the chemical single-line diagram. Based on this model, the method detects the table to be identified. When the serial numbers in the column of the table to be identified are consecutive and the row spacing is within a preset ratio range, the corresponding multiple rows of information are merged. The method then obtains the table auxiliary lines of the chemical single-line diagram to be identified and performs AI recognition on the table based on these auxiliary lines to obtain the recognition result. This method effectively solves the problems of inaccurate merging of multiple rows and inaccurate column segmentation in the material identification of chemical single-line diagrams. It accurately identifies various material information in the chemical single-line diagram. Compared with traditional methods, it reduces manual intervention, lowers the workload and error risk of manual processing, and provides more efficient and accurate support for material information management in chemical engineering design and construction. It has significant practical application value and broad market prospects, improving the speed and efficiency of material identification in chemical single-line diagrams. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention; Figure 2 This is a flowchart illustrating the first embodiment of the chemical single-line diagram material identification method of the present invention; Figure 3 This is a flowchart illustrating the second embodiment of the chemical single-line diagram material identification method of the present invention; Figure 4 This is a flowchart illustrating the third embodiment of the chemical single-line diagram material identification method of the present invention; Figure 5 This is a flowchart illustrating the fourth embodiment of the chemical single-line diagram material identification method of the present invention; Figure 6 This is a functional block diagram of the first embodiment of the chemical single-line diagram material identification device of the present invention.

[0016] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0018] The solution of this invention mainly involves: establishing a correspondence model between the serial number and the row spacing ratio in the material table of a chemical single-line diagram; detecting the table to be identified based on the correspondence model; merging the corresponding multiple rows of information when the serial numbers in the column of the table to be identified are consecutive and the row spacing is within a preset ratio range; obtaining the table auxiliary lines of the chemical single-line diagram to be identified; and performing AI recognition on the table to be identified based on the table auxiliary lines to obtain the recognition result. This effectively solves the problems of inaccurate merging of multiple rows and inaccurate column segmentation in the material recognition of chemical single-line diagrams, accurately identifying various material information in the chemical single-line diagram. Compared with traditional methods, it reduces manual intervention and lowers costs. The increased workload and error risk associated with manual processing can provide more efficient and accurate support for material information management in the design and construction process of chemical engineering. It has significant practical application value and broad market prospects, improving the speed and efficiency of material identification in chemical single-line diagrams. It solves the technical problems of traditional table recognition in chemical single-line diagrams, which cannot accurately integrate relevant information when faced with multiple rows being merged, leading to chaotic material information identification. For tables without dividing lines or with inaccurate division, it is also difficult to correctly divide columns, thus affecting the accuracy of material information extraction. This restricts the automation and accuracy of material information identification in chemical single-line diagrams and increases the workload and error risk of manual processing.

[0019] Reference Figure 1 , Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.

[0020] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU; a communication bus 1002; a user interface 1003; a network interface 1004; and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0021] Those skilled in the art will understand that Figure 1 The device structure shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0022] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating device, a network communication module, a user interface module, and a chemical single-line diagram material identification program.

[0023] The device of this invention calls the chemical single-line diagram material identification program stored in the memory 1005 through the processor 1001, and performs the following operations: Establish a correspondence model between the serial number and the row spacing ratio in the material table of a chemical single-line diagram, and detect the table to be identified based on the correspondence model. When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within the preset ratio range, the corresponding multiple rows of information are merged. Obtain the table auxiliary lines of the chemical single-line diagram to be identified, and perform AI recognition on the table to be identified based on the table auxiliary lines to obtain the recognition result.

[0024] The device of the present invention, through processor 1001 calling the chemical single-line diagram material identification program stored in memory 1005, also performs the following operations: Obtain a preset number of chemical single-line graph material tables, and construct a training dataset based on the chemical single-line graph material tables; The training dataset is trained using a machine learning algorithm to obtain a correspondence model between the serial number and the row spacing ratio, and the table to be identified is detected based on the correspondence model.

[0025] The device of the present invention, through processor 1001 calling the chemical single-line diagram material identification program stored in memory 1005, also performs the following operations: Obtain a preset number of chemical single-line diagram material tables, and preprocess the chemical single-line diagram material tables to obtain preprocessed tables; The text information in the preprocessed table is extracted using character recognition technology, and the serial number column in the preprocessed table is located to generate a training dataset.

[0026] The device of the present invention, through processor 1001 calling the chemical single-line diagram material identification program stored in memory 1005, also performs the following operations: The training dataset is trained according to the machine learning algorithm, the sequence number and line spacing ratio in the training dataset are learned and modeled, the line spacing ratio threshold is determined, and the correspondence between the sequence number and the line spacing ratio is obtained. Based on the correspondence, establish a correspondence model between each serial number and the ratio of each row spacing, and detect the table to be identified based on the correspondence model.

[0027] The device of the present invention, through processor 1001 calling the chemical single-line diagram material identification program stored in memory 1005, also performs the following operations: When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, multiple rows of information with consecutive serial numbers and row spacing within the preset ratio range are merged into a complete material information record.

[0028] The device of the present invention, through processor 1001 calling the chemical single-line diagram material identification program stored in memory 1005, also performs the following operations: Obtain detailed information from the single-line diagram of the chemical plant to be identified, and automatically generate auxiliary lines in the table based on the detailed information; The column structure of the table to be identified is determined based on the table auxiliary lines, and the table auxiliary lines are added to the single-line diagram of the chemical product to be identified to obtain the target image to be identified. The target image to be identified is input into the AI ​​recognition system, so that the AI ​​recognition system can perform column segmentation of the table in the target image according to the table auxiliary lines, extract the material information of each column, and generate recognition results.

[0029] The device of the present invention, through processor 1001 calling the chemical single-line diagram material identification program stored in memory 1005, also performs the following operations: Image preprocessing is performed on the chemical single-line diagram to be identified to obtain a preprocessed table image. Image analysis is then performed on the preprocessed table image according to a preset image analysis technique to obtain detailed information about the chemical single-line diagram to be identified. Based on the detailed information, table auxiliary lines are automatically drawn according to preset drawing rules.

[0030] This embodiment, through the above-described scheme, establishes a correspondence model between the serial number and the row spacing ratio in the material table of a chemical single-line diagram. Based on this model, it detects the table to be identified. When the serial numbers in the column of the table to be identified are consecutive and the row spacing is within a preset ratio range, the corresponding multiple rows of information are merged. The auxiliary lines of the chemical single-line diagram to be identified are obtained, and AI recognition is performed on the table based on these auxiliary lines to obtain the recognition result. This effectively solves the problems of inaccurate merging of multiple rows and inaccurate column segmentation in the material identification of chemical single-line diagrams. It accurately identifies various material information in chemical single-line diagrams. Compared with traditional methods, it reduces manual intervention, lowers the workload and error risk of manual processing, and provides more efficient and accurate support for material information management in the design and construction process of chemical engineering. It has significant practical application value and broad market prospects, improving the speed and efficiency of material identification in chemical single-line diagrams.

[0031] Based on the above hardware structure, an embodiment of the chemical single-line diagram material identification method of the present invention is proposed.

[0032] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the chemical single-line diagram material identification method of the present invention.

[0033] In the first embodiment, the chemical single-line diagram material identification method includes the following steps: Step S10: Establish a correspondence model between the serial number and the row spacing ratio in the chemical single-line diagram material table, and detect the table to be identified based on the correspondence model.

[0034] It should be noted that the correspondence model is a model that corresponds between different serial numbers and different row spacing ratios. After establishing the correspondence model between serial numbers and row spacing ratios in the chemical single-line diagram material table, the table to be identified can be detected through the correspondence model.

[0035] Step S20: When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within the preset ratio range, the corresponding multiple rows of information are merged.

[0036] It should be understood that when it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within the preset ratio range, the corresponding multiple rows of information can be merged. This can solve the problem that traditional table recognition cannot accurately integrate relevant information when facing the situation of multiple rows being merged in a chemical single-line diagram, resulting in chaotic material information recognition.

[0037] Step S30: Obtain the table auxiliary lines of the chemical single-line diagram to be identified, and perform AI recognition on the table to be identified based on the table auxiliary lines to obtain the recognition result.

[0038] It is understandable that after obtaining the table auxiliary lines of the chemical single-line diagram to be identified, the table to be identified can be AI-recognized based on the table auxiliary lines, thereby obtaining the table AI recognition result, which can solve the problem of column segmentation errors caused by the absence of dividing lines or inaccurate segmentation of the table.

[0039] This embodiment, through the above-described scheme, establishes a correspondence model between the serial number and the row spacing ratio in the material table of a chemical single-line diagram. Based on this model, it detects the table to be identified. When the serial numbers in the column of the table to be identified are consecutive and the row spacing is within a preset ratio range, the corresponding multiple rows of information are merged. The auxiliary lines of the chemical single-line diagram to be identified are obtained, and AI recognition is performed on the table based on these auxiliary lines to obtain the recognition result. This effectively solves the problems of inaccurate merging of multiple rows and inaccurate column segmentation in the material identification of chemical single-line diagrams. It accurately identifies various material information in chemical single-line diagrams. Compared with traditional methods, it reduces manual intervention, lowers the workload and error risk of manual processing, and provides more efficient and accurate support for material information management in the design and construction process of chemical engineering. It has significant practical application value and broad market prospects, improving the speed and efficiency of material identification in chemical single-line diagrams.

[0040] Furthermore, Figure 3 This is a flowchart illustrating the second embodiment of the chemical single-line diagram material identification method of the present invention, as shown below. Figure 3 As shown, based on the first embodiment, a second embodiment of the chemical single-line diagram material identification method of the present invention is proposed. In this embodiment, step S10 specifically includes the following steps: Step S11: Obtain a preset number of chemical single-line diagram material tables, and construct a training dataset based on the chemical single-line diagram material tables.

[0041] It should be noted that by collecting and obtaining a pre-defined number of chemical single-line diagram material tables, a training dataset for model training can be constructed based on the chemical single-line diagram material tables.

[0042] Furthermore, step S11 specifically includes the following steps: Obtain a preset number of chemical single-line diagram material tables, and preprocess the chemical single-line diagram material tables to obtain preprocessed tables; The text information in the preprocessed table is extracted using character recognition technology, and the serial number column in the preprocessed table is located to generate a training dataset.

[0043] It is understandable that by obtaining a preset number of chemical single-line diagram material tables, the chemical single-line diagram material tables can be preprocessed to obtain a preprocessed table. The text information in the preprocessed table can be extracted by character recognition technology, and a training dataset can be generated by locating the serial number column in the preprocessed table.

[0044] In the specific implementation, chemical single-line diagrams and material tables from multiple design institutes are collected and organized to construct a training dataset. Specifically, image processing techniques are used to preprocess the table images, including grayscale conversion, noise reduction, and binarization, to improve image quality and recognition accuracy. Then, character recognition technology is used to extract text information from the table and locate the serial number column, thereby generating a training dataset based on the processed data.

[0045] Step S12: Train the training dataset according to the machine learning algorithm to obtain the correspondence model between the serial number and the row spacing ratio, and detect the table to be identified according to the correspondence model.

[0046] It should be understood that the training dataset can be trained using machine learning algorithms to obtain a correspondence model between the serial number and the row spacing ratio, thereby detecting the table to be identified based on the correspondence model.

[0047] Furthermore, step S12 specifically includes the following steps: The training dataset is trained according to the machine learning algorithm, the sequence number and line spacing ratio in the training dataset are learned and modeled, the line spacing ratio threshold is determined, and the correspondence between the sequence number and the line spacing ratio is obtained. Based on the correspondence, establish a correspondence model between each serial number and the ratio of each row spacing, and detect the table to be identified based on the correspondence model.

[0048] In the specific implementation, by combining multiple design institute drawings and through analysis, it was found that in the material tables of chemical single-line diagrams, the serial number is usually in the first column. Based on this pattern, this embodiment adopts a method of merging multiple rows according to the serial number and the row spacing ratio. Specifically, by learning and analyzing a large number of chemical single-line diagram material tables, a correspondence model between the serial number and the row spacing ratio is established. Specifically, machine learning algorithms, such as support vector machines and neural networks, can be used to learn and model the serial number and the row spacing ratio, determine a reasonable row spacing ratio threshold, and obtain the correspondence between different serial numbers and different row spacing ratios. Thus, a correspondence model between each serial number and each row spacing ratio is established based on the correspondence, and the table to be identified is detected based on the correspondence model.

[0049] This embodiment, through the above-described scheme, obtains a preset number of chemical single-line diagram material tables, constructs a training dataset based on these tables, trains the training dataset using a machine learning algorithm to obtain a correspondence model between the sequence number and the row spacing ratio, and detects the table to be identified based on this model. This solves the problem of inaccurate integration of multi-row merged information in existing chemical single-line diagram material identification, accurately identifies various material information in chemical single-line diagrams, improves the accuracy, speed, and efficiency of chemical single-line diagram material identification.

[0050] Furthermore, Figure 4 This is a flowchart illustrating the third embodiment of the chemical single-line diagram material identification method of the present invention, as shown below. Figure 4 As shown, based on the first embodiment, a third embodiment of the chemical single-line diagram material identification method of the present invention is proposed. In this embodiment, step S20 specifically includes the following steps: Step S21: When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, the multiple rows of information with consecutive serial numbers and row spacing within the preset ratio range are merged into a complete material information record.

[0051] It should be noted that when it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, multiple rows of information with consecutive serial numbers and row spacing within the preset ratio range can be merged into a complete material information record.

[0052] In practical implementation, during the actual recognition process, the input chemical single-line diagram material table image is processed. When multiple rows of information with consecutive serial numbers and row spacing within a set threshold range are detected, they are merged into a complete material information record. That is, when consecutive serial numbers in a column are detected and row spacing is within a set ratio range, the corresponding multiple rows of information are merged, thereby accurately integrating the material information from multiple rows. For example, when recognizing a table containing material information such as pipes and fittings, multiple rows of description information for the same material can be accurately merged, avoiding information omissions and incorrect integration.

[0053] This embodiment, through the above-described scheme, merges multiple rows of information with consecutive serial numbers and row spacing within a preset ratio range into a single complete material information record when the serial number column of the table to be identified is detected to be within the preset ratio range. This solves the problem of inaccurate integration of multiple merged information in existing chemical single-line diagram material identification, and can accurately identify various types of material information in chemical single-line diagrams. Compared with traditional methods, it reduces manual intervention, lowers the workload and error risk of manual processing, and improves the speed and efficiency of chemical single-line diagram material identification.

[0054] Furthermore, Figure 5 This is a flowchart illustrating the fourth embodiment of the chemical single-line diagram material identification method of the present invention, as shown below. Figure 5 As shown, based on the first embodiment, a fourth embodiment of the chemical single-line diagram material identification method of the present invention is proposed. In this embodiment, step S30 specifically includes the following steps: Step S31: Obtain detailed information of the chemical single-line diagram to be identified, and automatically generate table auxiliary lines based on the detailed information.

[0055] It should be noted that after obtaining the detailed information describing the relevant table content in the table to be identified, table auxiliary lines can be automatically generated based on the detailed information.

[0056] Furthermore, step S31 specifically includes the following steps: Image preprocessing is performed on the chemical single-line diagram to be identified to obtain a preprocessed table image. Image analysis is then performed on the preprocessed table image according to a preset image analysis technique to obtain detailed information about the chemical single-line diagram to be identified. Based on the detailed information, table auxiliary lines are automatically drawn according to preset drawing rules.

[0057] It is understandable that after image preprocessing of the chemical single-line diagram to be identified, a preprocessed table image can be obtained. Then, image analysis is performed on the preprocessed table image according to the preset image analysis technology to obtain detailed information of the chemical single-line diagram to be identified. Based on the detailed information, table auxiliary lines are automatically drawn and generated according to the preset drawing rules.

[0058] In the specific implementation, before recognizing the chemical single-line diagram, image preprocessing is also performed first. Image analysis techniques, such as edge detection and connected component analysis, can be used to obtain information such as the distribution of text in the table, font size, and line spacing. Based on this information, auxiliary lines of the table are automatically drawn according to preset rules.

[0059] Step S32: Determine the column structure of the table to be identified based on the table auxiliary lines, add the table auxiliary lines to the single-line diagram of the chemical to be identified, and obtain the target image to be identified.

[0060] Understandably, the drawing of auxiliary lines should ensure that they accurately reflect the column structure of the table and do not affect the recognition of text information. The column structure of the table to be recognized is determined based on the table auxiliary lines, and then the table auxiliary lines are added to the single-line diagram of the chemical to be recognized to form the target image to be recognized.

[0061] In practical implementation, to solve the problem of column segmentation errors caused by the absence of dividing lines or inaccurate segmentation in tables, this embodiment proposes to draw table auxiliary lines to help AI correctly segment columns. In specific implementation, before material recognition is performed on the chemical single-line diagram, table auxiliary lines can be automatically generated based on image features and table layout. These auxiliary lines are drawn according to information such as the distribution of text, font size, and row spacing in the table, so that they can accurately reflect the column structure of the table.

[0062] Step S33: Input the target image to be identified into the AI ​​recognition system so that the AI ​​recognition system can perform column segmentation of the table in the target image to be identified according to the table auxiliary lines, extract the material information of each column, and generate recognition results.

[0063] It should be understood that when the target image to be identified is input into the AI ​​recognition system, artificial intelligence (AI) can segment the table in the target image to be identified into columns according to the table auxiliary lines, extract the material information of each column, and generate the corresponding material data recognition result.

[0064] In practical implementation, the image with added auxiliary lines is input into the AI ​​recognition system. The AI ​​system uses the auxiliary lines to segment the table content into columns, extracting material information from each column. By using the auxiliary lines as a reference during column segmentation, the AI ​​can more accurately divide the table content into different columns, thus achieving correct extraction of material information. For example, when processing tables containing information such as prefabricated materials, codes, and quantities, it can clearly and accurately segment the information in each column, avoiding information confusion.

[0065] This embodiment, through the above-described scheme, obtains detailed information of the chemical single-line diagram to be identified, automatically generates table auxiliary lines based on the detailed information, determines the column structure of the table to be identified based on the table auxiliary lines, adds the target image to the chemical single-line diagram to be identified, and obtains the target image to be identified. The target image to be identified is then input into an AI recognition system, which performs column segmentation of the table in the target image based on the table auxiliary lines, extracts the material information of each column, and generates recognition results. This effectively solves the problems of multi-row merging and inaccurate column segmentation in the material recognition of chemical single-line diagrams, accurately identifies various types of material information in chemical single-line diagrams, and reduces manual intervention, workload, and error risk compared to traditional methods. It can provide more efficient and accurate support for material information management in the design and construction process of chemical engineering, has significant practical application value and broad market prospects, and improves the speed and efficiency of material recognition in chemical single-line diagrams.

[0066] Accordingly, the present invention further provides a chemical single-line diagram material identification device.

[0067] Reference Figure 6 , Figure 6 This is a functional block diagram of the first embodiment of the chemical single-line diagram material identification device of the present invention.

[0068] In the first embodiment of the chemical single-line diagram material identification device of the present invention, the chemical single-line diagram material identification device includes: The model building module 10 is used to establish a correspondence model between the serial number and the row spacing ratio in the chemical single-line diagram material table, and to detect the table to be identified based on the correspondence model.

[0069] The information merging module 20 is used to merge the corresponding multiple rows of information when it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range.

[0070] The information recognition module 30 is used to obtain the table auxiliary lines of the chemical single-line diagram to be recognized, and to perform AI recognition on the table to be recognized based on the table auxiliary lines to obtain the recognition result.

[0071] The model building module 10 is further configured to acquire a preset number of chemical single-line diagram material tables, construct a training dataset based on the chemical single-line diagram material tables, train the training dataset according to a machine learning algorithm to obtain a correspondence model between the serial number and the row spacing ratio, and detect the table to be identified based on the correspondence model.

[0072] The model building module 10 is also used to obtain a preset number of chemical single-line graph material tables, preprocess the chemical single-line graph material tables to obtain preprocessed tables, extract text information in the preprocessed tables using character recognition technology, locate the serial number column in the preprocessed tables, and generate a training dataset.

[0073] The model building module 10 is further configured to train the training dataset according to a machine learning algorithm, learn and model the serial number and line spacing ratio in the training dataset, determine the line spacing ratio threshold, and obtain the correspondence between the serial number and the line spacing ratio; establish a correspondence model between each serial number and each line spacing ratio based on the correspondence, and detect the table to be identified based on the correspondence model.

[0074] The information merging module 20 is also used to merge multiple rows of information with consecutive serial numbers and row spacing within the preset ratio range into a complete material information record when it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within the preset ratio range.

[0075] The information recognition module 30 is further configured to acquire detailed information of the chemical single-line diagram to be recognized, automatically generate table auxiliary lines based on the detailed information, determine the column structure of the table to be recognized based on the table auxiliary lines, add the target image to be recognized to the chemical single-line diagram to be recognized, and obtain the target image to be recognized; input the target image to be recognized into the AI ​​recognition system, so that the AI ​​recognition system can perform column segmentation of the table in the target image to be recognized based on the table auxiliary lines, extract the material information of each column, and generate a recognition result.

[0076] The information recognition module 30 is also used to perform image preprocessing on the chemical single-line diagram to be recognized, obtain a preprocessed table image, perform image analysis on the preprocessed table image according to a preset image analysis technology, obtain detailed information of the chemical single-line diagram to be recognized, and automatically draw and generate table auxiliary lines according to the detailed information and preset drawing rules.

[0077] The steps for implementing each functional module of the chemical single-line diagram material identification device can be referred to in the various embodiments of the chemical single-line diagram material identification method of the present invention, and will not be repeated here.

[0078] Furthermore, this embodiment of the invention also proposes a storage medium storing a chemical single-line diagram material identification program, which, when executed by a processor, performs the following operations: Establish a correspondence model between the serial number and the row spacing ratio in the material table of a chemical single-line diagram, and detect the table to be identified based on the correspondence model. When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within the preset ratio range, the corresponding multiple rows of information are merged. Obtain the table auxiliary lines of the chemical single-line diagram to be identified, and perform AI recognition on the table to be identified based on the table auxiliary lines to obtain the recognition result.

[0079] Furthermore, when the chemical single-line diagram material identification program is executed by the processor, it also performs the following operations: Obtain a preset number of chemical single-line graph material tables, and construct a training dataset based on the chemical single-line graph material tables; The training dataset is trained using a machine learning algorithm to obtain a correspondence model between the serial number and the row spacing ratio, and the table to be identified is detected based on the correspondence model.

[0080] Furthermore, when the chemical single-line diagram material identification program is executed by the processor, it also performs the following operations: Obtain a preset number of chemical single-line diagram material tables, and preprocess the chemical single-line diagram material tables to obtain preprocessed tables; The text information in the preprocessed table is extracted using character recognition technology, and the serial number column in the preprocessed table is located to generate a training dataset.

[0081] Furthermore, when the chemical single-line diagram material identification program is executed by the processor, it also performs the following operations: The training dataset is trained according to the machine learning algorithm, the sequence number and line spacing ratio in the training dataset are learned and modeled, the line spacing ratio threshold is determined, and the correspondence between the sequence number and the line spacing ratio is obtained. Based on the correspondence, establish a correspondence model between each serial number and the ratio of each row spacing, and detect the table to be identified based on the correspondence model.

[0082] Furthermore, when the chemical single-line diagram material identification program is executed by the processor, it also performs the following operations: When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, multiple rows of information with consecutive serial numbers and row spacing within the preset ratio range are merged into a complete material information record.

[0083] Furthermore, when the chemical single-line diagram material identification program is executed by the processor, it also performs the following operations: Obtain detailed information from the single-line diagram of the chemical plant to be identified, and automatically generate auxiliary lines in the table based on the detailed information; The column structure of the table to be identified is determined based on the table auxiliary lines, and the target image to be identified is added to the single-line diagram of the chemical process to be identified to obtain the target image to be identified. The target image to be identified is input into the AI ​​recognition system, so that the AI ​​recognition system can perform column segmentation of the table in the target image according to the table auxiliary lines, extract the material information of each column, and generate recognition results.

[0084] Furthermore, when the chemical single-line diagram material identification program is executed by the processor, it also performs the following operations: Image preprocessing is performed on the chemical single-line diagram to be identified to obtain a preprocessed table image. Image analysis is then performed on the preprocessed table image according to a preset image analysis technique to obtain detailed information about the chemical single-line diagram to be identified.

[0085] Those skilled in the art will understand that all or part of the steps in the methods described above can be implemented by a program instructing related hardware. The program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium is a computer-readable storage medium, including: USB flash drive, mobile hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program code.

[0086] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0087] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0088] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for identifying materials using a single-line diagram in chemical engineering, characterized in that, The chemical single-line diagram material identification method includes: Establish a correspondence model between the serial number and the row spacing ratio in the material table of a chemical single-line diagram, and detect the table to be identified based on the correspondence model. When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within the preset ratio range, the corresponding multiple rows of information are merged. Obtain the table auxiliary lines of the chemical single-line diagram to be identified, and perform AI recognition on the table to be identified based on the table auxiliary lines to obtain the recognition result.

2. The chemical single-line diagram material identification method as described in claim 1, characterized in that, The process of establishing a correspondence model between the serial number and the row spacing ratio in the chemical engineering single-line diagram material table, and detecting the table to be identified based on the correspondence model, includes: Obtain a preset number of chemical single-line graph material tables, and construct a training dataset based on the chemical single-line graph material tables; The training dataset is trained using a machine learning algorithm to obtain a correspondence model between the serial number and the row spacing ratio, and the table to be identified is detected based on the correspondence model.

3. The chemical single-line diagram material identification method as described in claim 2, characterized in that, The step of obtaining a preset number of chemical single-line graph material tables and constructing a training dataset based on the chemical single-line graph material tables includes: Obtain a preset number of chemical single-line diagram material tables, and preprocess the chemical single-line diagram material tables to obtain preprocessed tables; The text information in the preprocessed table is extracted using character recognition technology, and the serial number column in the preprocessed table is located to generate a training dataset.

4. The chemical single-line diagram material identification method as described in claim 2, characterized in that, The step of training the training dataset using a machine learning algorithm to obtain a correspondence model between serial numbers and row spacing ratios, and then detecting the table to be identified based on the correspondence model, includes: The training dataset is trained according to the machine learning algorithm, the sequence number and line spacing ratio in the training dataset are learned and modeled, the line spacing ratio threshold is determined, and the correspondence between the sequence number and the line spacing ratio is obtained. Based on the correspondence, establish a correspondence model between each serial number and the ratio of each row spacing, and detect the table to be identified based on the correspondence model.

5. The chemical single-line diagram material identification method as described in claim 1, characterized in that, When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, the corresponding multiple rows of information are merged, including: When it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range, multiple rows of information with consecutive serial numbers and row spacing within the preset ratio range are merged into a complete material information record.

6. The chemical single-line diagram material identification method as described in claim 1, characterized in that, The process of obtaining the auxiliary lines of the table in the single-line diagram of the chemical industry to be identified, and performing AI recognition on the table based on the auxiliary lines to obtain the recognition result includes: Obtain detailed information from the single-line diagram of the chemical plant to be identified, and automatically generate auxiliary lines in the table based on the detailed information; The column structure of the table to be identified is determined based on the table auxiliary lines, and the target image to be identified is added to the single-line diagram of the chemical process to be identified to obtain the target image to be identified. The target image to be identified is input into the AI ​​recognition system, so that the AI ​​recognition system can perform column segmentation of the table in the target image according to the table auxiliary lines, extract the material information of each column, and generate recognition results.

7. The chemical single-line diagram material identification method as described in claim 6, characterized in that, The process of obtaining detailed information from the single-line diagram of the chemical plant to be identified, and automatically generating auxiliary lines for the table based on the detailed information, includes: Image preprocessing is performed on the chemical single-line diagram to be identified to obtain a preprocessed table image. Image analysis is then performed on the preprocessed table image according to a preset image analysis technique to obtain detailed information about the chemical single-line diagram to be identified. Based on the detailed information, table auxiliary lines are automatically drawn according to preset drawing rules.

8. A chemical single-line diagram material identification device, characterized in that, The chemical single-line diagram material identification device includes: The model building module is used to establish a correspondence model between the serial number and the row spacing ratio in the material table of the chemical single-line diagram, and to detect the table to be identified based on the correspondence model. The information merging module is used to merge multiple rows of information when it is detected that the serial numbers in the serial number column of the table to be identified are consecutive and the row spacing is within a preset ratio range. The information recognition module is used to acquire the table auxiliary lines of the chemical single-line diagram to be recognized, and to perform AI recognition on the table to be recognized based on the table auxiliary lines to obtain the recognition result.

9. A chemical single-line diagram material identification device, characterized in that, The chemical single-line diagram material identification device includes: a memory, a processor, and a chemical single-line diagram material identification program stored in the memory and executable on the processor, wherein the chemical single-line diagram material identification program is configured to implement the steps of the chemical single-line diagram material identification method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a chemical single-line diagram material identification program, which, when executed by a processor, implements the steps of the chemical single-line diagram material identification method as described in any one of claims 1 to 7.

Images