KR-20260064171-A - METHOD AND DEVICE FOR IDENTIFYING PIPING AND INSTRUMENTATION DIAGRAM USING GRAPH NEURAL NETWORK

KR20260064171AKR 20260064171 AKR20260064171 AKR 20260064171AKR-20260064171-A

Abstract

The present disclosure provides a method for identifying an electrical piping instrumentation diagram performed by at least one processor, comprising: acquiring an image of the electrical piping instrumentation diagram; identifying a plurality of line objects from the image of the electrical piping instrumentation diagram; identifying relationships between the plurality of line objects; generating a plurality of nodes corresponding to the plurality of line objects and a plurality of edges connecting the plurality of nodes based on the identified plurality of line objects and the relationships between the identified plurality of line objects, and classifying the plurality of line objects using a graph neural network.

Inventors

문두환
한승태
문유찬
김지법

Assignees

고려대학교 산학협력단

Dates

Publication Date: 20260507
Application Date: 20241031

Claims (13)

In a method for identifying electrical conduit instrumentation performed by at least one processor, Step of acquiring an image of the electrical wiring instrumentation diagram; A step of identifying a plurality of line objects from the above electrical wiring instrumentation diagram image; A step of identifying the relationship between the plurality of line objects mentioned above; Based on the relationship between the identified plurality of line objects and the identified plurality of line objects, a step of generating a plurality of nodes corresponding to the plurality of line objects and a plurality of edges connecting the plurality of nodes; and A method comprising the step of classifying the plurality of line objects using a graph neural network. Method for identifying electrical conduit instrumentation.
In Article 1, The plurality of nodes above include a first node corresponding to a first line object and a second node corresponding to a second line object, and The first node has a first node attribute value, and the second node has a second node attribute value different from the first attribute value. Method for identifying electrical conduit instrumentation.
In Article 2, The first node and the second node differ from each other in at least one of position, length, thickness, arrow type, whether comment symbols are connected, and proximity to text related to multiples. Method for identifying electrical conduit instrumentation.
In Article 2, The first node attribute value and the second node attribute value are different in size or identifier. Method for identifying electrical conduit instrumentation.
In Article 1, The relationship between the plurality of line objects includes at least one of whether two or more line objects are directly connected, whether they are connected upon extension, and whether they are a broken line relationship. Method for identifying electrical conduit instrumentation.
In Article 1, The plurality of edges includes a first edge connected to a first node and a second node included in the plurality of nodes, and a second edge connected to the first node and a third node. The first edge has a first edge attribute value, and the second edge has a second edge attribute value different from the first edge attribute value. Method for identifying electrical conduit instrumentation.
In Article 6, The first edge attribute value and the second edge attribute value are different in size or identifier from each other. Method for identifying electrical conduit instrumentation.
In Article 1, The step of classifying the above-mentioned plurality of line objects The method includes the step of inputting a graph comprising the plurality of nodes and the plurality of edges into at least one layer in which the plurality of nodes learn node attribute values from neighboring nodes to preserve structural data of the graph. Method for identifying electrical conduit instrumentation.
In Article 8, A step comprising performing a perturbation technique using dropout on the attribute values of the plurality of nodes. Method for identifying electrical conduit instrumentation.
In Article 1, The step of identifying the plurality of line objects includes the step of determining the thickness of the plurality of line objects, The step of determining the thickness of the above line object A step of sampling multiple pixels between the start point and the end point of the above line object; The step of moving and tracing the plurality of pixels in a certain direction; and A step comprising calculating a thickness value based on the distance moved by the plurality of pixels. Method for identifying electrical conduit instrumentation.
In Article 1, The step of identifying the above plurality of line objects is A step of obtaining an input line from the above electrical conduit instrumentation diagram; A step of searching for a start point, an end point, and an intersection point from the above input line; and The step of decomposing the input line based on the above intersection point Method for identifying electrical conduit instrumentation.
A computer program stored on a computer-readable, non-transient recording medium to execute the method for identifying any one of the electrical conduit instruments of claims 1 through 11.
Communication module; memory for storing instructions; and It includes at least one processor connected to the memory and configured to execute at least one computer-readable program included in the memory, and The above at least one program is, Acquire an image of the electrical piping instrumentation, Identifying multiple line objects from the above electrical wiring instrumentation diagram image, and Identifying the relationships between the above multiple line objects, and Based on the relationships between the identified plurality of line objects and the identified plurality of line objects, a plurality of nodes corresponding to the plurality of line objects and a plurality of edges connecting the plurality of nodes are generated, and Classifying the above multiple line objects using a Graph Neural Network Electrical wiring instrumentation identification device.

Description

Method and Device for Identifying Piping and Instrumentation Diagram Using Graph Neural Network The present invention relates to a method and apparatus for identifying electrical piping instrumentation diagrams using a graph neural network, and more specifically, to a method for identifying lines of electrical piping instrumentation diagrams using a graph neural network. Electrical piping and instrumentation drawings contain symbols, text, and line objects, and line objects are broadly classified into lines containing symbols and solid lines. Here, solid lines are classified into eight types: thick solid lines, general solid lines, dimension lines, dimension extension lines, leader lines, specification boundary leader lines, annotation symbol leader lines, and drainage lines. Conventionally, symbols and arrows within electrical piping and instrumentation drawings were recognized based on deep learning, and eight types of lines were identified using a rule-based algorithm. However, these conventional technologies utilize rule-based algorithms, which have the problem of being unable to distinguish lines with patterns other than predefined rules and lacking noise handling capabilities. FIG. 1 is a configuration diagram of a computing device for identifying electrical wiring instrumentation according to one embodiment. FIG. 2 is a flowchart of a method for identifying electrical conduit instrumentation according to one embodiment. FIG. 3 is a diagram illustrating the structure of the result data recognized in an electrical conduit instrumentation diagram according to one embodiment. FIG. 4 is a diagram illustrating the structure of result data recognized for lines and arrows within an electrical conduit instrumentation diagram according to one embodiment. FIG. 5 is a flowchart of a method for identifying line objects according to one embodiment. FIG. 6 is an example diagram of a method for decomposing an input line according to one embodiment. FIG. 7 is an example diagram of a method for determining the thickness of a line object according to one embodiment. FIGS. 8 and 9 are exemplary diagrams that do not satisfy the broken line candidate condition according to one embodiment. FIGS. 10 to 11 are exemplary drawings that do not satisfy the broken line identification condition according to one embodiment. FIG. 12 is a flowchart of a method for classifying nodes using a graph neural network according to one embodiment. FIG. 13 is a diagram comparing the results of a line classification method in an electrical conduit instrumentation diagram according to one embodiment with the results of a conventional method. In FIG. 13, the left side is an input electrical conduit instrumentation diagram image, the middle side is a line classification result identified using a conventional rule-based algorithm, and the right side is a line classification result identified according to the electrical conduit instrumentation diagram identification method according to one embodiment. FIGS. 14 to 17 are drawings showing the results of line object classification according to the electrical conduit instrumentation identification method according to one embodiment. Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the contents described in the attached drawings. However, the present invention is not limited or restricted by exemplary embodiments. Unless otherwise defined, all terms used in this specification (including technical and scientific terms) shall be used in a meaning that is commonly understood by those skilled in the art to which this disclosure belongs, but this may vary depending on the intent of those skilled in the art, case law, the emergence of new technology, etc. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise. In certain cases, terms have been selected at the applicant's discretion, and in such cases, their meanings will be described in detail in the relevant explanatory sections. Accordingly, terms used in this disclosure should be defined not merely by their names, but based on their meanings and the content throughout this disclosure. Throughout this specification, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. Furthermore, the singular form used in this specification includes the plural form unless specifically stated otherwise. Additionally, the expression "at least one of a, b, and/or c" as used throughout this specification may encompass 'a alone', 'b alone', 'c alone', 'a and b', 'a and c', 'b and c', or 'a, b, and c all'. Meanwhile, terms such as "first and/or second" used in this specification may be used to describe various components, but they are used solely for the purpose of distingui