JP6833627B2 - Raster format drawing structuring system and its structuring method - Google Patents

Description

An embodiment of the present invention relates to a raster format drawing structuring system for structuring a raster format drawing and outputting a structured drawing, and a method for structuring the raster format drawing.

Plants such as factories and power plants have a huge number of drawings relating to plant systems, systems or circuits such as distribution systems, operating equipment and monitoring equipment. Many of these drawings are managed as electronic files by scanning paper or paper with a scanner. Electronic files read by these papers or scanners are referred to herein as unstructured data. Unstructured data refers to data in which the connections and relationships of symbols, wiring, parts, etc. drawn in the drawing are less clear than the data finally generated.

Originally, the drawings related to these plants are often drawn on a computer using CAD software or the like, but there are cases where the drawings are delivered only as paper drawings or the CAD drawing data is lost. In this way, when CAD drawing data does not exist, in order to revise the drawing due to the update of the power distribution system or equipment, it was created by drawing again with CAD software while looking at the paper drawing, etc., or by scanning with a scanner. Since an electronic file (for example, TIF or PDF) is a drawing in raster format (that is, a collection of dots constituting each pixel), there is a method of editing a part of the drawing using a dedicated raster image editing software. ..

In the former method of manually recreating CAD data from scratch, it takes a lot of cost and time to deal with a huge amount of plant drawings. When using CAD software, it is often the case that drawing is performed while adding connection information and attribute information of equipment and wiring, so that the drawing data of the CAD software is often structured data.

In the latter method of editing an image using dedicated raster image editing software, it is sufficient to edit only the part to be changed on the drawing. However, in the case of this method, the files before and after editing do not have attribute information and connection information in the equipment, wiring, etc. depicted in the drawing. It is possible to manually or semi-automatically add attribute information and the like one by one, but in this case, it takes time and cost as in the case of creating CAD. Furthermore, in the case of this method, since the circuit or system drawn on the drawing does not have attribute information and connection information, the electronic file generated by this method should be searched for the circuit or system. It cannot be applied to various simulations.

If the electronic file has the attribute information and connection information of equipment and wiring, it is possible to construct a system for planning isolation work in a plant, for example. In isolation work, when performing construction, maintenance, inspection, or repair, in order to ensure the safety of workers and minimize the impact on other equipment and systems, the target distribution system and equipment should be selected. It refers to the work of electrically isolating from the plant system or system and stopping (power failure).

Japanese Unexamined Patent Publication No. 2005-4726

In the background technology described above, a paper drawing or a paper drawing scanned into a raster format electronic file such as TIF or PDF is edited, added, and attribute information and connection information are added to the circuit or system. In order to utilize it for simulations and the like, it is necessary to manually draw a huge amount of drawings one by one using CAD, and to manually or semi-automatically add attribute information and the like. These require a great deal of time and cost and cannot cover all of the huge amount of drawings that the plant has.

An embodiment of the present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a raster format drawing structuring system capable of automatically structuring a raster format drawing and a method for structuring the raster format drawing.

The raster format drawing structuring system according to the embodiment of the present invention is a raster format drawing structuring system that converts a raster format drawing into a vector format drawing and structures the raster format drawing into the vector format drawing. In the vector format drawing, referring to the vector conversion function unit for vector conversion, the structured rule definition database that defines the rules for structuring the structural elements in the vector format drawing, and the structured rule definition database. It is characterized by having a structured functional unit for structuring the structural element by adding attribute information and connection information to the structural element.

The method for structuring a raster format drawing in the embodiment of the present invention is a method for structuring a raster format drawing in which the raster format drawing is converted into a vector format drawing and structured, and the wiring and circuit elements in the vector format drawing. , Prepare a structured rule definition database that defines rules for structuring structural elements such as tables, vector-convert the raster format drawing to the vector format drawing, and refer to the structured rule definition database. It is characterized in that attribute information and connection information are added to the structural element in the vector format drawing to structure the structural element.

According to an embodiment of the present invention, a raster format drawing can be automatically structured.

The block diagram which shows the structured system of the raster type drawing which concerns on one Embodiment of this invention. A block diagram showing various functions executed by the structured function unit of FIG. 1 together with a database. Explanatory drawing which shows an example of a vector format drawing. The main flowchart which shows the procedure which the structured system of FIG. 1 outputs a structured drawing. The first sub-flow chart derived from the main flow chart of FIG. A second sub-flow chart derived from the main flow chart of FIG. A third sub-flow chart derived from the main flowchart of FIG.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a structured system of raster type drawings according to an embodiment of the present invention. Further, FIG. 2 is a block diagram showing various functions executed by the structured function unit of FIG. 1 together with a database. The raster format drawing structuring system 10 shown in FIG. 1 inputs, for example, a raster format drawing 1 relating to a plant, converts the raster format drawing 1 into a vector format drawing, and then structures and outputs a structured drawing 2. Is.

Here, the drawings relating to the plant include a plant building, a layout drawing, a P & ID (piping instrumentation drawing), an ECWD (deployment connection diagram), an IBD (interlock block diagram), a single connection diagram, a soft logic diagram, and the like. A raster format drawing is a drawing in which wiring, circuit elements, tables, characters, etc. drawn in the drawing are all aggregates of dots (pixels), and is a paper drawing or a TIF obtained by scanning this paper drawing with a scanner. It is an electronic file drawing such as or PDF. Therefore, this raster format drawing is an unstructured drawing because attribute information and connection information are not added to the wiring, circuit elements, etc. on the drawing.

A structured drawing (that is, a structured drawing) 2 is an image in which at least structural elements other than characters and alphanumeric characters such as wiring, circuit elements, and tables are vector format images, and characters and characters and structural elements are included in the structural elements. It is a drawing in which alphanumeric characters are defined in association with each other as character information, and attribute information and connection information are added to structural elements. Here, the connection information is wiring information, piping information, and the like. The vector format drawing or image is a drawing or image composed of the coordinates of the start point and the end point of the line segment and a function representing the line segment.

For example, in the vector format drawing shown in FIG. 3, as the attribute information of the circuit element, reference numeral 3 is a relay of a contact, and this relay has terminals 4 and 5. The connection information of this circuit element (relay) is such that the terminal 4 is connected to the wiring 6 and the terminal 5 is connected to the wiring 7. Further, in FIG. 3, reference numeral 8 also indicates wiring, and reference numeral 9 indicates a board frame.

By the way, the structured system 10 shown in FIG. 1 includes a noise removal / correction function unit 11, a vector conversion function unit 12, a character / alphanumeric character recognition function unit 13, an erroneous conversion / erroneous recognition check / correction function unit 14, and a structuring rule. It is configured to include a definition DB (DB is an abbreviation for a database; the same applies hereinafter) 15, a circuit element / attribute DB 16, a data structure definition DB 17, a structured function unit 18, and a structured check / correct function unit 19. Further, the structured system 10 includes an erroneous conversion / erroneous recognition data storage unit 20, an optimum conversion / recognition learning unit 21, a structured data storage unit 22, and an optimum structured learning unit 23. The structured system 10 may be implemented in a personal computer, a server, or the like, or may be stored in a cloud server.

The noise removal / correction function unit 11 removes and corrects noise such as spots, stains, distortions, and chips in the raster format drawing 1. The noise removal / correction by the noise removal / correction function unit 11 is performed by, for example, image processing based on a preset set value.

The vector conversion function unit 12 vector-converts the raster format drawing 1 into a vector format drawing. The vector conversion function unit 12 vector-converts the raster format drawing 1 into a vector format drawing including the characters and alphanumeric characters in the raster format drawing 1, and converts the characters and alphanumeric characters in the vector format drawing into characters and alphanumeric characters. The recognition function unit 13 may recognize it. Alternatively, the vector conversion function unit 12 extracts in advance the image in the raster format drawing 1 that meets the preset conditions for characters and alphanumeric characters before vector conversion of the raster format drawing 1, or the vector conversion target. It may be specified outside and vector-converted other than the characters and alphanumeric characters in the raster format drawing 1. As the above setting conditions, for example, an image within 10 pixels in the raster format drawing 1 is regarded as characters and alphanumeric characters. Further, a vector-converted image other than characters and alphanumeric characters in the raster format drawing 1 is referred to as a vector format image.

The character / alphanumeric character recognition function unit 13 recognizes characters and alphanumeric characters from the raster format drawing 1 or the vector format drawing. That is, the character / alphanumeric character recognition function unit 13 recognizes the characters and alphanumeric characters in the raster format drawing 1, and also recognizes the characters and alphanumeric characters in the vector format drawing vector-converted by the vector conversion function unit 12. In each embodiment, the character / alphanumeric character recognition function unit 13 may not be provided.

The erroneous conversion / erroneous recognition check / correction function unit 14 confirms whether or not there is an erroneous conversion in the vector format drawing vector-converted by the vector conversion function unit 12, and if there is an erroneous conversion, the erroneous conversion location. To fix. The erroneous conversion is, for example, an event that the wiring in the vector format drawing is not connected, or the circular shape of the circuit element in the vector format drawing is a polygonal shape. Further, the erroneous conversion / erroneous recognition check / correction function unit 14 confirms whether or not the characters and alphanumeric characters recognized by the character / alphanumeric character recognition function unit 13 have erroneous recognition, and if there is erroneous recognition. To fix. In each embodiment, when the character / alphanumeric character recognition function unit 13 is not provided, the erroneous conversion / erroneous recognition check / correction function unit 14 uses the characters and English recognized by the character / alphanumeric character recognition function unit 13. It is configured as an erroneous conversion check / correction function unit that does not have a function of confirming whether or not there is erroneous recognition of numbers and only has a function of confirming whether or not there is erroneous conversion in the vector format drawing.

The structured rule definition DB15 is a database that defines rules for structuring structural elements (for example, wiring, drawing frame, board frame, circuit element, table structure, operation switch development drawing, crossover line, etc.) in a vector format drawing. The characteristics of the structural elements are defined and registered. For example, in the vector format drawing, the characteristic of the wiring is a solid line having a predetermined length or more (see reference numerals 6, 7, and 8 in FIG. 3), and the feature of the board frame is a two-dot chain line (see reference numeral 9 in FIG. 3). , For example, a relay of a circuit element is characterized by a specific character representing the relay (see "49X", "30X", "52X" in FIG. 3) and a shape near the character (circle in FIG. 3 and less than a predetermined length). See the solid line of).

The circuit element / attribute DB 16 is a part of the structured rule definition DB 15, and is a database that specifically defines circuit elements. That is, in this circuit element / attribute DB 16, the characteristics (characters and shapes) of the circuit element on the vector format drawing are registered in association with the attribute information and connection information of the circuit element. The circuit element / attribute DB 16 may be provided separately from the structured rule definition DB 15.

The data structure definition DB 17 is a database that defines a hierarchy (layer) of a data structure as a reference when classifying and recording data of structured structural elements. For example, structured wiring data and circuit element data (including attribute information and connection information of the circuit element) are classified into recording media (not shown) in a hierarchical structure based on the data structure hierarchy of the data structure definition DB17. Is recorded.

The structured function unit 18 refers to the structured rule definition DB 15 (including the circuit element / attribute DB 16), and the structural element such as wiring and circuit elements in the vector format drawing vector-converted by the vector conversion function unit 12. The characters and alphanumeric characters recognized by the character / alphanumeric character recognition function unit 13 are associated with each other, and attribute information and connection information are further added to structure the structural elements. Further, the structured function unit 18 classifies and records the data of the structured structural elements in a recording medium (not shown) based on the hierarchy of the data structure of the data structure definition DB 17. As shown in FIG. 2, the structured function unit 18 has a wiring extraction function, a diagram frame / board frame extraction function, a circuit element extraction function, a table structure extraction function, an operation switch development diagram extraction function, a crossing line extraction function, and the like. It has a drawing name, system number, and separate volume number extraction function.

The wiring extraction function of the structured function unit 18 is as follows. First, the structuring function unit 18 searches for and extracts a solid line having a predetermined length or more in the vector format drawing. Next, when the extracted solid line matches the characteristic (solid line) of the wiring registered in the structured rule definition DB 15, the structured function unit 18 registers the extracted solid line in the structured rule definition DB 15. Define and structure as. After that, the structured function unit 18 classifies and records the data of the structured wiring into a recording medium based on the hierarchy of the data structure defined in the data structure definition DB 17, and records the data in the wiring with an identification number (line). Number) is attached.

The drawing frame / board frame extraction function of the structured function unit 18 is as follows. First, the structured function unit 18 searches for and extracts a two-dot chain line in a vector format drawing. Next, when the extracted two-dot chain line matches the feature (two-dot chain line) of the board frame registered in the structured rule definition DB 15, the structured function unit 18 structures the extracted two-dot chain line. It is defined and structured as a board frame registered in the rule definition DB 15. After that, the structured function unit 18 classifies and records the data of the structured board frame into a recording medium based on the hierarchy of the data structure defined in the data structure definition DB 17, and records the board when necessary. An identification number (line number) is attached to the frame. The structuring function unit 18 also structures the drawing frame in the same manner as the board frame, records the data, and assigns an identification number when necessary.

The circuit element extraction function of the structured function unit 18 is as follows. First, the structuring function unit 18 searches for and extracts a shape in the vicinity of characters and alphanumeric characters representing a specific circuit element in a vector format drawing. Next, when the shape of the extracted circuit element matches the shape of the circuit element registered in the circuit element / attribute DB 16, the structured function unit 18 sets the extracted circuit element of the shape to the circuit element / attribute. It is structured by replacing the circuit element registered in DB16. After that, the structured function unit 18 classifies and records the attribute information and connection information of the structured circuit element into a recording medium based on the hierarchy of the data structure defined in the data structure definition DB 17, and records the circuit element. Is given an identification number.

The table structure extraction function, the operation switch development diagram extraction function, the crossover line extraction function, and the drawing name / system number / separate volume number extraction function of the structuring function unit 18 are also implemented in the same manner as the wiring extraction function described above. That is, when the features of the table structure, the crossover, etc. extracted on the vector format drawing match the features of the table structure, the crossover, etc. registered in the structuring rule definition DB 15, the structuring function unit 18 has the vector format. The table structure, crossovers, etc. on the drawing are defined as the table structures, crossovers, etc. registered in the structuring rule definition DB 15, and are structured. Then, the structured function unit 18 classifies and records the data such as the structured table structure and the crossing line based on the hierarchy of the data structure of the data structure definition DB 17, and records the data in the table structure and the crossing line when necessary. Give an identification number.

The structured check / correct function unit 19 shown in FIG. 1 is structured based on the structured rule definition DB 15 with respect to the structured result (that is, the structured drawing 2) structured by the structured function unit 18. Recognize the unstructured part as improper structuring (that is, improper structuring result), and correct the recognized improper structuring part (that is, improper structuring result). Here, the parts related to improper structuring (that is, improper structuring result) are, for example, wiring is unconnected, attribute information is not given to structural elements that require attribute information, wiring, circuit elements, etc. An identification number is not given to a structural element that requires an identification number, such as. When the structure check / correction function unit 19 determines that the correction is unnecessary, the structure system 10 outputs the structure drawing 2.

In the erroneous conversion / erroneous recognition data storage unit 20, the erroneous conversion / erroneous recognition check / correction function unit 14 determines that the vector conversion by the vector conversion function unit 12 is erroneous conversion and corrects the erroneous conversion. The format drawing, the result of erroneous conversion by the vector conversion function unit 12 (that is, erroneous conversion), the conversion result corrected by the erroneous conversion / erroneous recognition check / correction function unit 14, and the erroneous conversion by the vector conversion function unit 12 The conversion setting value at the time is input and accumulated.

Further, in the erroneous conversion / erroneous recognition data storage unit 20, the erroneous conversion / erroneous recognition check / correction function unit 14 determines that the recognition of characters and alphanumeric characters by the character / alphanumeric character recognition function unit 13 is erroneous and corrects it. Regarding the misrecognition, the raster format drawing before recognition, the result of misrecognition by the character / alphanumeric character recognition function unit 13 (that is, misrecognition), and the recognition corrected by the misconversion / misrecognition check / correction function unit 14. The result and the recognition setting value when the character / alphanumeric character recognition function unit 13 erroneously recognizes are input and accumulated.

The optimum conversion / recognition learning unit 21 contains information on erroneous conversion of the vector conversion function unit 12 stored in the erroneous conversion / erroneous recognition data storage unit 20 (a raster format drawing before vector conversion and erroneous conversion by the vector conversion function unit 12). The result, the conversion result corrected by the erroneous conversion / erroneous recognition check / correction function unit 14, and the conversion setting value when the vector conversion function unit 12 erroneously converts) are input. Then, the optimum conversion / recognition learning unit 21 causes an erroneous conversion in the vector conversion function unit 12 based on the information regarding the erroneous conversion of the vector conversion function unit 12 input from the erroneous conversion / erroneous recognition data storage unit 20. The optimum conversion setting value that reduces the rate is calculated and learned, and the erroneous conversion setting value is adjusted. When there is a large amount of erroneous conversion data, the optimum conversion / recognition learning unit 21 may use deep learning using a multi-layer neural network. The optimum conversion / recognition learning unit 21 further feeds back the calculated optimum conversion setting value to the vector conversion function unit 12 to update the conversion setting value of the vector conversion in the vector conversion function unit 12.

In addition, the optimum conversion / recognition learning unit 21 provides information on misrecognition of the character / alphanumeric recognition function unit 13 stored in the misconversion / misrecognition data storage unit 20 (raster format drawing before misrecognition and character / English). The result of erroneous recognition by the number recognition function unit 13, the recognition result corrected by the erroneous conversion / erroneous recognition check / correction function unit 14, and the recognition setting value when the character / alphanumeric character recognition function unit 13 erroneously recognizes) Enter. Then, the optimum conversion / recognition learning unit 21 is based on the information regarding the erroneous recognition of the character / alphanumeric recognition function unit 13 input from the erroneous conversion / erroneous recognition data storage unit 20, and the character / alphanumeric recognition function unit 13 The optimum recognition setting value that reduces the occurrence rate of false recognition in is calculated and learned, and the false recognition setting value is adjusted. When there is a large amount of misrecognition data, the optimum conversion / recognition learning unit 21 may use deep learning using a multi-layer neural network. The optimum conversion / recognition learning unit 21 further feeds back the calculated optimum recognition setting value to the character / alphanumeric character recognition function unit 13, and the character / alphanumeric character recognition setting value in the character / alphanumeric character recognition function unit 13. To update.

In the structured data storage unit 22, improper structuring related to the portion corrected by the structured check / correction function unit 19 (that is, the structuring result by the structured function unit 18 corrected by the structured check / correction function unit 19) ), Vector format images before structuring, characters and alphanumeric characters, improper structuring output by the structuring function unit 18 that needs to be modified (that is, structuring results that need to be modified), and structure. The result corrected by the conversion check / correction function unit 19 and the structured setting value (structuring rule) of the structured function unit 18 when improperly structured (that is, the structured result requiring correction) is output. , Is input and accumulated.

The optimal structured learning unit 23 has information stored in the structured data storage unit 22 (vector format image before structuring, characters and alphanumeric characters, and a structured result output by the structured function unit 18 that needs to be modified. And the result corrected by the structured check / correction function unit 19 and the structured setting value of the structured function unit 18 when the structured result requiring the correction is output) are input. Then, the optimal structured learning unit 23 reduces the number of occurrences of structured results that need to be corrected by the structured function unit 18 based on the information input from the structured data storage unit 22. Calculate and learn the values to reset the structured settings. Further, the optimum structured learning unit 23 feeds back the calculated optimum structured setting value to the structured rule definition DB 15 (including the circuit element / attribute DB 16), and feeds back the calculated optimum structured setting value to the structured rule definition DB 15 (circuit element / attribute DB 16). (Including) to update the structured setting value (structuring rule).

Next, the operation of the structured system 10 configured as described above will be described with reference to FIGS. 4 to 7.
The structured system 10 inputs the raster format drawing 1 (S1). When there is noise such as spots or stains on the raster format drawing 1 that has been input, the noise removal / correction function unit 11 corrects the noise.

Next, the vector conversion function unit 12 vector-converts the raster format drawing 1 into a vector format drawing (S2), and the character / alphanumeric character recognition function unit 13 converts the raster format drawing 1 or the vector converted by the vector conversion function unit 12. Recognize characters and alphanumeric characters in the format drawing (S3).

Next, the erroneous conversion / erroneous recognition check / correction function unit 14 determines whether or not the vector format drawing vector-converted by the vector conversion function unit 12 has erroneous conversion (S4). Further, the erroneous conversion / erroneous recognition check / correction function unit 14 determines whether or not the characters and alphanumeric characters recognized by the character / alphanumeric character recognition function unit 13 have erroneous recognition (S5).

When the erroneous conversion / erroneous recognition check / correction function unit 14 determines that there is no erroneous conversion in the vector conversion and there is no erroneous recognition in the recognition of characters and alphanumeric characters (S6), the structured function unit 18 performs the vector conversion. The vector format drawing converted by the function unit 12 and the characters and alphanumeric characters recognized by the character / alphanumeric character recognition function unit 13 are input (S7).

Next, the structuring function unit 18 calls and inputs information regarding the data structure of the vector format drawing to be structured from the data structure definition DB 17 (S8). Further, the structuring function unit 18 calls and inputs information regarding the rules for structuring from the structuring rule definition DB 15 (S9). Further, the structured function unit 18 calls and inputs information about the circuit element in the vector format drawing from the circuit element / attribute DB 16 (S10).

The structuring function unit 18 is based on the conversion result (vector format drawing) by the vector conversion function unit 12 and the recognition result of characters and alphanumeric characters by the character / alphanumeric character recognition function unit 13, and the structuring rule definition DB 15 and circuit elements. The structural elements in the vector format drawing are structured with reference to the attribute DB 16 (S11). Further, the structured function unit 18 classifies and records the data of the structured structural elements based on the hierarchy of the data structure of the data structure definition DB 17.

The structuring check / correction function unit 19 confirms whether or not the structuring result (structured drawing 2) structured by the structuring function unit 18 needs to be modified (S12). When the structured check / correction function unit 19 determines that the structured result does not need to be corrected (S13), the structured drawing 2 is output from the structured system 10 (S14).

In step S6, when the erroneous conversion / erroneous recognition check / correction function unit 14 determines that there is an erroneous conversion in the vector conversion by the vector conversion function unit 12, the erroneous conversion / erroneous recognition check / correction is performed as shown in FIG. The function unit 14 corrects the erroneous conversion (S21), and outputs the corrected vector format drawing to the structured function unit 18 (S22).

Further, the erroneous conversion / erroneous recognition check / correction function unit 14 corrects the raster format drawing before vector conversion, the result of erroneous conversion by the vector conversion function unit 12, and the erroneous conversion / erroneous recognition check / correction function unit 14. The converted conversion result and the conversion setting value at the time of erroneous conversion by the vector conversion function unit 12 are output to the erroneous conversion / erroneous recognition data storage unit 20 and stored (S23).

Based on the information accumulated in the erroneous conversion / erroneous recognition data storage unit 20 in step S23, the optimum conversion / recognition learning unit 21 performs the optimum conversion to reduce the number of erroneous conversions in the vector conversion function unit 12. Calculate the set value (S24). Then, the optimum conversion / recognition learning unit 21 feeds back the optimum conversion setting value to the vector conversion function unit 12 to update the conversion setting value of the vector conversion function unit 12 (S25).

Further, in step S6 of FIG. 1, when the erroneous conversion / erroneous recognition check / correction function unit 14 determines that the character / alphanumeric character recognition function unit 13 has erroneous recognition, it is shown in FIG. As described above, the erroneous conversion / erroneous recognition check / correction function unit 14 corrects the erroneous recognition (S31), and outputs the corrected characters and alphanumeric characters to the structured function unit 18 (S32).

Further, the erroneous conversion / erroneous recognition check / correction function unit 14 includes a raster format drawing before recognition, a result of erroneous recognition by the character / alphanumeric character recognition function unit 13, and an erroneous conversion / erroneous recognition check / correction function unit 14. The recognition result corrected by the above and the recognition set value when the character / alphanumeric character recognition function unit 13 erroneously recognizes are output to the erroneous conversion / erroneous recognition data storage unit 20 and stored (S33).

Based on the information accumulated in the erroneous conversion / erroneous recognition data storage unit 20 in step S33, the optimum conversion / recognition learning unit 21 reduces the occurrence rate of erroneous recognition in the character / alphanumeric character recognition function unit 13. The optimum recognition setting value is calculated (S34). Then, the optimum conversion / recognition learning unit 21 feeds back the optimum recognition setting value to the character / alphanumeric character recognition function unit 13 to update the recognition setting value of the character / alphanumeric character recognition function unit 13 (S35). ).

In step S13 of FIG. 1, when the structuring check / correction function unit 19 determines that the structuring result (structured drawing 2) by the structuring function unit 18 needs to be corrected, this is as shown in FIG. The structuring check / correction function unit 19 corrects the structuring result that requires the above-mentioned correction (S41), and outputs the corrected structuring result, that is, the structuring drawing 2 from the structuring system 10 (S42).

Further, the structured check / correction function unit 19 includes a vector format image before structuring, characters and alphanumeric characters, a structured result that needs to be corrected, and a result corrected by the structured check / correction function unit 19. The structured setting value (structured rule) of the structured function unit 18 when the structured result requiring correction is output is output to the structured data storage unit 22 and stored (S43).

Based on the information accumulated in the structured data storage unit 22 in step S43, the optimal structure learning unit 23 is optimally structured to reduce the number of occurrences of structuring that need to be modified by the structured function unit 18. Calculate the set value (S44). Then, the optimum structured learning unit 23 feeds back the optimum structured setting value to the structured rule definition DB 15 (including the circuit element / attribute DB 16), and feeds the structured rule definition DB 15 (circuit element / attribute DB 16). The structured setting value (structuring rule) of (including) is updated (S45).

Since it is configured as described above, the following effects are obtained according to the present embodiment.
When the raster format drawing 1 is input to the structured system 10, the vector conversion function unit 12 converts the raster format drawing 1 into a vector format drawing, and the character / alphanumeric function unit 13 vector-converts the raster format drawing 1 or the vector. Recognize characters and alphanumeric characters from formal drawings. Then, the structured function unit 18 refers to the structured rule definition DB 15 (including the circuit element / attribute DB 16), and displays characters on the structured element in the vector format drawing vector-converted by the vector conversion function unit 12. -The characters and alphanumeric characters recognized by the alphanumeric character recognition function unit 13 are associated with each other, and attribute information and connection information are further added to structure the structural elements. Further, the structured function unit 18 classifies and records the data of the structured structural elements based on the hierarchy of the data structure of the data structure definition DB 17. Therefore, the vector conversion function unit 12, the character / alphanumeric character recognition function unit 13, the structured rule definition DB 15 (including the circuit element / attribute DB 16), the data structure definition DB 17, and the structured function unit 18 vectorize the raster format drawing 1. It can be converted into a formal drawing and structured automatically.

Although the embodiment of the present invention has been described above, this embodiment is presented as an example and is not intended to limit the scope of the invention. This embodiment can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the gist of the invention, and the replacements and changes thereof can be made. , It is included in the scope and gist of the invention, and is also included in the scope of the invention described in the claims and the equivalent scope thereof.

For example, in the present embodiment, the case where the raster format drawing 1 is a raster format drawing related to a plant has been described, but the raster format drawing 1 may be a raster format drawing relating to a device such as a vehicle, a ship, or an aircraft. Good.

1 ... Raster format drawing, 2 ... Structured drawing, 10 ... Structured system, 11 ... Noise removal / correction function unit, 12 ... Vector conversion function unit, 13 ... Character / alphanumeric recognition function unit, 14 ... Incorrect conversion / error Recognition check / correction function unit, 15 ... Structured rule definition DB, 16 ... Circuit element / attribute DB, 17 ... Data structure definition DB, 18 ... Structured function unit, 19 ... Structured check / correction function unit, 20 ... Incorrect Conversion / misrecognition data storage unit, 21 ... Optimal conversion / recognition learning unit, 22 ... Structured data storage unit, 23 ... Optimal structured learning unit.

Claims (12)

A raster format drawing structuring system that converts raster format drawings into vector format drawings and structures them.
A vector conversion function unit that vector-converts the raster format drawing into the vector format drawing,
A structured rule definition database that defines rules for structuring structural elements in the vector format drawing, and
It is characterized by having a structured function unit for structuring the structural element by adding attribute information and connection information to the structural element in the vector format drawing with reference to the structured rule definition database. A structured system for raster format drawings. Further provided with a data structure definition database that defines a hierarchy of data structures as a reference when recording the data of the structured elements.
The structure of the raster format drawing according to claim 1, wherein the structuring function unit is configured to classify and record the structured data of the structural elements based on the data structure definition database. System. The structured rule definition database includes a shape that is a feature of a circuit element on a vector format drawing, and a circuit element / attribute database in which attribute information and connection information of the circuit element are registered in association with each other.
The structured function unit extracts a shape representing a specific circuit element from the vector format drawing, and when the shape of the extracted circuit element matches the shape of the circuit element registered in the circuit element / attribute database. The structuring of the raster format drawing according to claim 1 or 2, wherein the circuit element having the extracted shape is replaced with a circuit element registered in the circuit element / attribute database to be structured. system. It is characterized by having an erroneous conversion check / correction function unit that confirms whether or not there is an erroneous conversion in the vector format drawing that has been vector-converted by the vector conversion function unit and corrects if there is an erroneous conversion. The structured system for raster format drawings according to any one of claims 1 to 3. Incorrect conversion data that accumulates the erroneous conversion corrected by the erroneous conversion check / correction function unit, the conversion setting value at the time of the erroneous conversion in the vector conversion function unit, and the corrected conversion result. Has a storage section
The structured system for raster format drawings according to claim 4, further comprising an optimum conversion learning unit that adjusts the conversion setting value based on the information accumulated in the erroneous conversion data storage unit. Further provided with a character / alphanumeric character recognition function unit that recognizes characters and alphanumeric characters from the raster format drawing or the vector format drawing.
The raster format drawing according to any one of claims 1 to 3, wherein the structuring function unit is configured to associate the characters and alphanumeric characters with the structural elements in the vector format drawing. Structured system. It is confirmed whether or not there is erroneous conversion or erroneous recognition in each of the vector format drawing vector-converted by the vector conversion function unit and the characters and alphanumeric characters recognized by the character / alphanumeric character recognition function unit. The raster format drawing structuring system according to claim 6, further comprising an erroneous conversion / erroneous recognition check / correction function unit that corrects erroneous conversions and erroneous recognitions. Regarding the misconversion or misrecognition corrected by the misconversion / misrecognition check / correction function unit, the conversion setting value or characters / English at the time of the misconversion or misrecognition and the misconversion in the vector conversion function unit. It has a misconversion / misrecognition data storage unit that stores the recognition setting value at the time of the misrecognition in the number recognition function unit and the corrected conversion result or recognition result.
The seventh aspect of claim 7, wherein the optimum conversion / recognition learning unit for adjusting the conversion set value or the recognition set value is provided based on the information accumulated in the erroneous conversion / erroneous recognition data storage unit. A structured system for rasterized drawings. The structured system for a raster format drawing according to any one of claims 1 to 8, further comprising a noise removal / correction function unit that removes and corrects noise in the raster format drawing. With respect to the structuring result structured by the structuring function unit, a part that is not structured based on the structuring rule definition database is recognized as inappropriate structuring, and the recognized improper structuring is performed. The structured system of a raster format drawing according to any one of claims 1 to 9, further comprising a structured check / correction function unit for modifying the above. Regarding the improper structuring related to the location corrected by the structured check / correction function unit, the improper structuring that needs to be corrected and the structure in the structured function unit that outputs the improper structuring. It has a structured data storage unit that stores the set value and
The structure of the raster format drawing according to claim 10, further comprising an optimal structured learning unit for resetting the structured setting value based on the information accumulated in the structured data storage unit. System. A method of structuring a raster format drawing that converts a raster format drawing into a vector format drawing and structures it.
Prepare a structured rule definition database that defines rules for structuring structural elements such as wiring, circuit elements, and tables in the vector format drawing.
Vector conversion of the raster format drawing into the vector format drawing is performed.
A method for structuring a raster format drawing, which comprises adding attribute information and connection information to the structural element in the vector format drawing to structure the structural element with reference to the structured rule definition database.

Images