Extracting and analyzing information from engineering drawings
A method and system for extracting information from a drawing. The method includes classifying nodes in the drawing, extracting attributes from the nodes, determining whether there are errors in the node attributes, and removing the nodes from the drawing. The method also includes identifying edges in the drawing, extracting attributes from the edges, and determining whether there are errors in the edge attributes. The system includes at least one processing component, at least one memory component, an identification component, an extraction component, and a correction component. The identification component is configured to classify nodes in the drawing, remove the nodes from the drawing, and identify edges in the drawing. The extraction component is configured to extract attributes from the nodes and edges. The correction component is configured to determine whether there are errors in the extracted attributes.
1. A system, comprising:
a memory; and
a processor communicatively coupled to the memory, wherein the processor is configured to:
identify, in a drawing, regions that contain nodes;
classify the identified regions based on the nodes;
extract node attributes from the identified regions;
determine whether there are errors in the extracted node attributes;
when the determination is complete, remove the node attributes and the nodes from the drawing;
when the node attributes and the nodes have been removed from the drawing, identify edges in the drawing;
extract edge attributes from the edges; and
determine whether there are errors in the extracted edge attributes.
2. The system of claim 1 , further comprising an attribute database.
3. The system of claim 2 , wherein the attribute database includes definitions of node symbols.
4. The system of claim 2 , wherein the attribute database includes domain information.
5. The system of claim 1 , wherein the extracted node attributes and edge attributes comprise text annotations.
6. The system of claim 1 , wherein the classification uses a region-based convolutional neural network (R-CNN).
7. The system of claim 6 , wherein the classification is rotation-invariant.
8. The system of claim 6 , wherein the classification is scale-invariant.
9. A method, comprising:
identifying, in a drawing, regions that contain nodes;
classifying the identified regions based on the nodes;
extracting node attributes from the identified regions;
determining whether there are errors in the node attributes;
when the determination is complete, removing the node attributes and the nodes from the drawing;
when the node attributes and the nodes have been removed from the drawing, identifying edges in the drawing;
extracting edge attributes from the edges; and
determining whether there are errors in the edge attributes.
10. The method of claim 9 , wherein the classifying uses a region-based convolutional neural network (R-CNN).
11. The method of claim 9 , wherein the identifying the edges comprises carrying out an edge detection operator and a reinforcement learning technique.
12. The method of claim 9 , wherein the classifying is scale-invariant.
13. The method of claim 9 , wherein the extracting the node attributes and the edge attributes comprises comparing the node attributes and the edge attributes to information in an attribute database.
14. The method of claim 13 , wherein the attribute database includes data from a set of published standards.
15. The method of claim 9 , wherein the determining whether there are errors in the node attributes and the edge attributes is based on domain information.
16. A computer program product for extracting information from a drawing, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method, the method comprising:
identifying, in the drawing, regions that contain nodes;
classifying the identified regions based on the nodes;
extracting node attributes from the identified regions;
determining whether there are errors in the node attributes;
when the determination is complete, removing the node attributes and the nodes from the drawing;
when the node attributes and the nodes have been removed from the drawing, identifying edges in the drawing;
extracting edge attributes from the edges; and
determining whether there are errors in the edge attributes.
17. The computer program product of claim 16 , wherein the classifying uses a region-based convolutional neural network (R-CNN).
18. The computer program product of claim 16 , wherein the identifying the edges comprises carrying out an edge detection operator and a reinforcement learning technique.
19. The computer program product of claim 16 , wherein the determining whether there are errors in the node attributes and the edge attributes is based on domain information.
20. The computer program product of claim 19 , wherein the domain information is stored in an attribute database.