KR20230132179A - Device For Creating Drawings, Method For Creating Drawings And Recording Medium - Google Patents

Abstract

The present invention provides a drawing generating device and drawing generating method that obtain information about a drawing and automatically process it using an image containing at least one of numbers, signs, diagrams, and symbols.

Description

Drawing creation device, drawing creation method and recording medium {Device For Creating Drawings, Method For Creating Drawings And Recording Medium}

The present invention relates to a drawing generating device and a drawing generating method, and more specifically, to a drawing generating device, drawing generating method, and recording medium that recognize numbers, signs, shapes, and symbols and process them into drawings.

Artificial intelligence is being studied very actively today due to the increase in data quantity, advanced algorithms, and improvements in computing power and storage. This artificial intelligence is currently being used in various fields such as national defense, healthcare, education, games, and security.

Among the applications of artificial intelligence, technology that recognizes numbers or letters using captured images is being used in a variety of ways. As an example, registration number 10-2181861 discloses information on a vehicle license plate detection and recognition system and method.

Meanwhile, in industrial sites such as glass, steel plate, and plywood manufacturing, it is essential to create drawings containing necessary processing information. Conventionally, when an ordering company provided randomly drawn drawings, the ordering company needed to standardize the drawings provided through CAD work. If such drawing processing is performed manually, it can be complicated, prone to errors, and can take a lot of time and money.

As the scope of use of artificial intelligence expands, attempts to quickly apply it in the manufacturing industry are continuing. In particular, there is a need to research methods that can reduce the manufacturing cost and manufacturing time of industrial materials by automatically recognizing arbitrarily written numbers or letters and arbitrarily drawn graphical figures and drawings.

Registration No. 10-2181861 Patent Publication

In order to solve the above-described problems, the present invention provides a drawing generating device and drawing generating method for obtaining and automatically processing information about a drawing using an image containing at least one of numbers, signs, shapes, and symbols. The purpose is to

In order to achieve the above-described object, the drawing generating method according to the present invention is a drawing generating method performed by a processor, a first image obtaining a first image having a numerical part containing numbers and a classification part containing symbols. acquisition phase; An image recognition step of recognizing the numerical part and the classification part from the first image; and a data generation step of obtaining drawing information about the type, size and shape of the processed figure from the recognized numerical part and the classification part. You can.

Additionally, the image recognition step may classify the numbers in the numerical part and the symbols in the classification part based on data learned by the first artificial intelligence model.

In addition, the first image further includes a quantity unit, and the drawing generation method may further include a management step of uploading the obtained drawing information and the quantity described in the quantity unit to an inventory management system.

In addition, the first image further includes a processing unit, and the drawing generating method may further include a management step of uploading the obtained drawing information and processing items described in the processing unit to an inventory management system.

Additionally, the drawing generation method may further include a storage step of storing the obtained drawing information in a memory in the form of a drawing file.

In addition, the data generation step includes a type creation step of determining the type of the machined figure according to the symbol of the classification unit, and a shape of determining the size and shape of the machined figure according to the determined type of the machined figure and the numerical part. It may include a creation step.

Additionally, in the type creation step, the type of the processed figure can be determined by finding a match with the symbol of the classification unit in the symbol list.

In addition, the drawing generation method is, when at least one of the numerical part and the classification part cannot be recognized in the image recognition step, or the type of the processed figure cannot be obtained from the classification unit in the data generation step, A notification step of providing notification to correct the first image may be further included.

Meanwhile, in the drawing generating device including a processor, the processor acquires a first image having a numeric part including numbers and a classification portion including symbols, and generates the numerical portion and the classification portion from the first image. It is possible to recognize and obtain drawing information about the type, size, and shape of the processed figure from the recognized numerical part and classification part.

Meanwhile, the drawing generating method according to the present invention is performed by a processor, comprising: a second image acquisition step of acquiring a second image of a drawing including a drawing shape; A road shape information acquisition step of acquiring road shape information about the road shape on the second image; and a first final shape acquisition step of acquiring the first final shape using the temporary shape information.

Additionally, in the step of obtaining the street shape information, a plurality of lines forming the outline of the street shape may be recognized based on the color of the pixels of the second image.

In addition, the method of generating a drawing may further include a correction step of removing the distortion element when the gauze shape has a distortion element according to the ghat shape information obtained in the ghat shape information acquisition step.

Additionally, the drawing generating method may further include generating a drawing file using the first final shape.

Meanwhile, in the drawing generating device including a processor, the processor acquires a second image of a road surface including a road shape, obtains road shape information regarding the road shape on the second image, and The first final shape can be obtained using the temporary shape information.

Meanwhile, the drawing generating method according to the present invention is performed by a processor, and includes a third image acquisition step of acquiring a third image of a drawing including a drawing shape and a symbol; A road shape information acquisition step of acquiring road shape information about the road shape on the third image; A symbol information acquisition step of acquiring symbol information about the symbol on the third image; and a second final shape acquisition step of obtaining a second final shape by applying the symbol information to the random shape information.

In addition, the step of obtaining the street shape information includes recognizing a plurality of lines forming the outline of the street shape based on the color of the pixel of the third image; Recognizing a plurality of mantissa numbers corresponding to the plurality of lines; and obtaining the pseudo-shape information based on the recognized plurality of lines and the recognized plurality of pseudo-digits.

In addition, the drawing generating method may further include a correction step of removing the distortion element when the ghat shape has a distortion element according to the ghat shape information obtained in the ghat shape information acquisition step.

In addition, the step of obtaining symbol information may include specifying a region of interest including the road shape and the symbol on the third image; Classifying the symbol within the region of interest using a second artificial intelligence model; and obtaining the symbol information for the classified symbols.

Additionally, the drawing generating method may further include generating a drawing file using the second final shape.

Meanwhile, in the drawing generating device including a processor, the processor acquires a third image of the road surface including a road shape and a symbol, and obtains road shape information regarding the road shape on the third image, , symbol information regarding the symbol on the third image may be obtained, and the symbol information may be applied to the temporary shape information to obtain a second final shape.

Meanwhile, a program for performing the drawing generation method according to an embodiment of the present invention may be recorded on a computer-readable recording medium.

The drawing generating device, drawing generating method, and recording medium according to the present invention can obtain drawing information about the type, size, and shape of the processed drawing by recognizing numbers and symbols, and recognize the drawing included in the drawing. You can obtain road shape information about the road shape, and by recognizing the road shape and symbols included in the road shape, you can obtain the road shape information and symbol information, automatically processing the drawing and reducing the manufacturing cost and manufacturing of materials. It can provide the effect of reducing time.

1 is a flowchart showing a method for generating a drawing using a first image according to an embodiment of the present invention.
Figure 2 is a diagram showing a processed figure generated based on drawing information.
Figure 3 is a diagram showing examples of a numerical part, a classification part, a quantity part, and a processing part.
Figure 4 is a diagram showing numbers and symbols classified by the first artificial intelligence model.
Figure 5 is a diagram showing examples of types of processed figures according to types of symbols.
Figure 6 is a diagram showing an example of a processed figure recognized as a square.
Figure 7 is a diagram showing an example of a processed figure recognized as a triangle.
Figure 8 is a diagram showing an example of a processed figure recognized as a circle.
Figure 9 is a flowchart showing a drawing generation method using a second image according to an embodiment of the present invention.
Figure 10 is a diagram showing the analysis area divided in the second image of the roadmap.
Figure 11 is a diagram showing the process of forming the first final shape based on the shape information.
Figure 12 is a flowchart showing a drawing generation method using a third image according to an embodiment of the present invention.
Figure 13 is a diagram showing a third image of a road plan including a plurality of road shapes.
Figure 14 is a diagram showing the analysis area divided in the third image of the roadmap.
Figure 15 is a diagram showing symbols recognized within a region of interest.
Figure 16 is a diagram showing the process of forming a second final shape from an intermediate shape.
Figure 17 is a diagram showing the second final shape.
Figure 18 is a diagram showing an example of a distortion element.
Figure 19 is a diagram showing a standardized state of a rough drawing.
Figure 20 is a diagram showing a standardized state of a rough drawing.
Figure 21 is a diagram showing a case where a symbol violates the symbol convention.
Figure 22 is a block diagram showing a drawing generating device according to an embodiment of the present invention.

Hereinafter, a drawing generating device, drawing generating method, and recording medium according to an embodiment of the present invention will be described.

Those skilled in the art will be able to develop various devices that embody the principles of the invention and are included within the spirit and scope of the invention, although not explicitly described or shown herein. In addition, it should be understood that all conditional terms and examples listed in this specification are, in principle, expressly intended only for the purpose of enabling the inventive concept to be understood, and are not limited to the examples and states specifically listed as such.

The above-described purpose, features and advantages will become more clear through the following detailed description of the invention in conjunction with the accompanying drawings, and accordingly, a person skilled in the art will be able to easily implement the technical idea of the invention.

Embodiments described herein will be explained with reference to cross-sectional views and/or perspective views, which are ideal illustrations of the present invention. The thicknesses of films and regions shown in these drawings are exaggerated for effective explanation of technical content. The form of the illustration may be modified depending on manufacturing technology and/or tolerance. In addition, the processed drawings, horizontal shapes, and numerical values shown in the drawings are illustrative for technical explanation, and the length of the processed drawings and horizontal shapes may not necessarily be proportional to the numbers. Accordingly, embodiments of the present invention are not limited to the specific form shown, but also include changes in form produced according to the manufacturing process.

While describing various embodiments, components that perform the same function will be given the same names and same reference numbers for convenience even if the embodiments are different. Furthermore, the configuration and operation already described in other embodiments will be omitted for convenience.

Below, the configuration of a drawing generating device (hereinafter referred to as “electronic device”) that performs a drawing generating method according to an embodiment of the present invention will be described.

Figure 22 is a block diagram showing a drawing generating device according to an embodiment of the present invention.

As an embodiment of the present invention, the electronic device 1000 may be called a terminal, a device, an electronic device, etc. Electronic devices 1000 include smartphones, tablet PCs, PCs, smart TVs, mobile phones, personal digital assistants (PDAs), laptops, media players, micro servers, navigation, kiosks, MP3 players, digital cameras, home appliances, and other computing devices. It may be a device, but is not limited thereto. Additionally, the electronic device 1000 may be a wearable device such as a watch, glasses, hair band, or ring equipped with a display function and data processing function. However, it is not limited thereto, and the electronic device 1000 may include all types of devices that can process data and provide processed data.

For example, as shown in FIG. 22, the electronic device 1000 according to one embodiment includes a memory 1100, an output unit 1200, a processor 1300, a communication unit 1500, and a photographing unit 1600. ) may also be included. Not all of the illustrated components are essential components of the electronic device 1000, and the electronic device 1000 may be implemented with more components, or the electronic device 1000 may be implemented with fewer components. It may be possible.

The memory 1100 may store programs for processing and control of the processor 1300, and may also store information input to the electronic device 1000 or information output from the electronic device 1000.

The memory 1100 may be a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (for example, SD or XD memory, etc.), or RAM. (RAM, Random Access Memory) SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), PROM (Programmable Read-Only Memory), magnetic memory, magnetic disk , and may include at least one type of storage medium among optical disks. A recording medium to be described later may be the memory 1100.

The output unit 1200 may output an audio signal, a video signal, or a vibration signal, and the output unit 1200 may include a display unit and a sound output unit.

The display unit displays and outputs information processed by the electronic device 1000. Specifically, the display unit may output an image captured by a camera. The display unit may display a user interface for executing an operation related to the response in response to the user's input.

The audio output unit outputs audio data received from the communication unit 1500 or stored in the memory 1100. Additionally, the sound output unit outputs sound signals related to functions performed in the electronic device 1000 (eg, call signal reception sound, message reception sound, notification sound).

The processor 1300 typically controls the overall operation of the electronic device 1000. For example, the processor 1300 can generally control the output unit 1200, the communication unit 1500, the photographing means 1600, etc. by executing programs stored in the memory 1100.

The communication unit 1500 may include one or more components that allow the electronic device 1000 to communicate with other devices (not shown) and servers (not shown). The other device (not shown) may be a computing device such as the electronic device 1000 or a sensing device, but is not limited thereto. For example, the communication unit 1500 may include a short-range communication unit (not shown) and a mobile communication unit (not shown).

The short-range wireless communication unit 1510 includes a Bluetooth communication unit, a Bluetooth Low Energy (BLE) communication unit, a Near Field Communication unit, a WLAN (Wi-Fi) communication unit, a Zigbee communication unit, and an infrared ( It may include, but is not limited to, an IrDA (infrared Data Association) communication unit, WFD (Wi-Fi Direct) communication unit, UWB (ultra wideband) communication unit, and Ant+ communication unit.

The mobile communication unit transmits and receives wireless signals to at least one of a base station, an external terminal, and a server on a mobile communication network. Here, the wireless signal may include various types of data according to voice call signals, video call signals, or text/multimedia message transmission and reception.

The photographing means 1600 can obtain an image frame, such as a still image or video, through an image sensor in a video call mode or a photographing mode. Images captured through the image sensor may be processed through the processor 1300 or a separate image processing unit (not shown). However, the photographing means 1600 may include various types of image acquisition devices and is not limited to the types.

Next, we will look at the drawing generation method according to an embodiment of the present invention.

Before detailed explanation, similar terms are distinguished.

‘Numbers’ and ‘symbols’ are used in the following meanings. 'Numbers in the dictionary meaning' and 'symbols in the dictionary meaning' appear in many instances in the specification and drawings, but in the present invention, 'numbers' are used in steps related to the first image and the processed figure 300, and the numerical part It means what is written in (31), and 'symbol' means what is used in the process related to the first image and the processed figure 300 and is written in the classification unit 32.

'Casual number (11)', 'pseudonym (120)' and 'tentative symbol' refer to temporary drawings that appear on the second and third images, and the second image, second image and first final image. It is used in processes related to the figure 150 and the second final figure 200.

The meaning of each term will be explained below along with the drawings.

1 is a flowchart showing a method for generating a drawing using a first image according to an embodiment of the present invention. Figure 2 is a diagram showing a processed figure generated based on drawing information. Figure 3 is a diagram showing examples of a numerical part, a classification part, a quantity part, and a processing part.

Referring to FIG. 1, the drawing generating method according to an embodiment of the present invention is performed by a processor 1300, and includes a numeric part 31 containing numbers and a classification unit 32 containing symbols. ) A first image acquisition step (S1100) of acquiring a first image having ; An image recognition step (S1200) of recognizing the numerical part 31 and the classification unit 32 from the first image; and the processed figure 300 from the recognized numerical part 31 and the classification unit 32. It may include a data generation step (S1300) of acquiring drawing information about type, size, and shape.

First, a first image acquisition step (S1100) of acquiring a first image having a numerical part 31 including numbers and a classification part 32 including symbols may be performed.

Referring to FIG. 2, numbers indicate information regarding the size of the machined figure 300. For example, a number may refer to the length of a polygon's side, the size of an angle, the diameter of a circle, or the radius of a circle.

The symbol means that information about the type of the processed figure 300 is written. For example, a symbol can mean either a triangle, a square, or a circle. Specific examples of symbols will be described later.

The numeric part 31 may include numbers. The numerical part 31 may consist of only numbers or may include '°'. Numbers and symbols can be written by hand or by electronic devices.

The first image has a numerical part 31 and a classification part 32, and can be created by taking an image or created by an electronic device.

The method by which the processor 1300 acquires the first image is not limited. The first image may be obtained by taking pictures with a drawing generating device. Additionally, the first image may be acquired through wired or wireless transmission from an external device or downloading from online.

Subsequently, an image recognition step (S1200) of recognizing the numerical part 31 and the classification part 32 from the first image may be performed.

The processor 1300 can recognize the numerical part 31 and the classification part 32 in the first image. The numerical part 31 is related to the size and shape of the machined figure 300 that will appear in the drawing to be created, and the classification portion 32 may be related to the type of the machined figure 300. At this time, the numerical unit 31 and the classification unit 32 may each be recognized by the first artificial intelligence model. Specific details will be described later.

Subsequently, a data generation step (S1300) of obtaining drawing information about the type, size, and shape of the machined figure 300 from the recognized numerical part 31 and the classification part 32 may be performed.

The processed figure 300 means formed from numbers and symbols recognized from the numeric part 31 and the classification part 32.

Processor 1300 may generate data. That is, the processor 1300 can extract the type, size, and shape of the processed figure 300 in data format from the numerical part 31 and the classification part 32 on the first image. The data generation step (S1300) is a process of converting numbers and symbols existing as images in the first image into data to generate data so that it can be used in the electronic device 1000.

Referring again to FIG. 2, the numerical part 31 means '200' and '100', and the classification part 32 means 'X'. That is, the processor 1300 can recognize that the ' and vertical length. The unit of the number may be mm, cm, or m, but is not limited thereto.

For example, the processor 1300 may use the first image containing '200 The processor 1300 may generate the machined drawing 300 using drawing information.

Next, let's look at the quantity unit 40.

Referring again to FIG. 2, the first image according to an embodiment of the present invention further includes a quantity unit 40, and the drawing generation method includes the obtained drawing information and the quantity 42 described in the quantity unit 40. ), may further include a management step of uploading to the inventory management system.

The quantity unit 40 may indicate the number of drawings to be created. Additionally, the quantity part may indicate the number of materials to be ordered. Processor 1300 may upload drawing information and quantity 42 to an inventory management system. The quantity unit 40 may include a quantity symbol 41 and a quantity 42. The quantity 42 means the number of drawings to be created, and the quantity symbol 41 can distinguish the quantity 42 from the numerical part 31 and the classification part 32.

According to FIGS. 2 and 3, the quantity unit 40 may be composed of a quantity symbol 41 of '=' and a quantity 42 of '1'. The processor 1300 recognizes the notation '200 Can be uploaded to inventory management system. At this time, the quantity symbol 41 is not limited to '=', and may have various forms such as '→', '/', 'long', 'EA', and '-'.

Next, we will look at the processing unit 50.

Referring to FIG. 3, the first image according to an embodiment of the present invention further includes a processing unit 50, and the drawing generating method includes the obtained drawing information and processing details described in the processing unit 50, A management step of uploading to an inventory management system may be further included.

The processing unit 50 may indicate processing details of materials to be ordered. Processing details refer to the processing details ordered by the orderer. Processing items may be classified into first processing items (51), second processing items (52), and/or third processing items (53). The processing unit 50 may include at least one of the first processing item 51, the second processing item 52, and the third processing item 53.

Taking glass as an example, the first processing item 51 may mean the thickness of the material (glass). In Figure 3(a), the first processing item 51 includes '5T', which means that the thickness of the material is 5mm. In Figure 3(b), the first processing item 51 includes '8T', which means that the thickness of the material is 8mm. However, the unit is not limited to mm, and the material is not limited to glass.

The second processing item 52 may refer to the color of the material (glass). 3(a) to 3(c), the second processing items 52 include 'transparent', 'blue', and 'green', respectively, which means that the colors of the materials are transparent, blue, and green, respectively. .

The third processing item 53 may represent the appearance of the material (glass). In Figure 3(c), the third processing item 53 includes 'chamfering', which means chamfering the material. In Figure 4(d), the third processing item 53 includes 'perforation', which means perforating the material. However, it is not limited to 'chamfering' and 'perforation', and various processing details can be described.

The first artificial intelligence model, which will be described later, can learn data about the first processing item (51), the second processing item (52), and the third processing item (53), and the learned first artificial intelligence model is a processing unit. (50) can be recognized and classified.

The processor 1300 may upload drawing information and processing details of the processing unit 50 to the inventory management system. For example, in the case of FIG. 3(d), the processor 1300 may obtain drawing information consisting of a square with a width of 200 cm and a height of 100 cm. Additionally, the processor 1300 may recognize the quantity as '5'. Additionally, the processor 1300 may recognize the first processing information 51 as '5T'. Additionally, the processor 1300 may recognize the second processing information 52 as 'green'. Additionally, the processor 1300 may recognize the third processing information 53 as 'perforation'.

That is, the processor 1300 can upload data about a square with a width of 200 cm and a height of 100 cm, quantity '5', thickness '5T', color 'green', and 'perforation' to the inventory management system.

In the present invention, when an orderer provides a first image with briefly and arbitrarily depicted contents, the drawing generation device can automatically convert the contents written on the first image into data and upload it to the inventory management system, generating drawing data. there is. This can have the effect of reducing material manufacturing costs, manufacturing time, etc.

Next, we look at the image recognition step (S1200).

Figure 4 is a diagram showing numbers and symbols classified by the first artificial intelligence model. Figure 5 is a diagram showing examples of types of processed figures according to types of symbols.

Referring to FIG. 4, the image recognition step (S1200) according to an embodiment of the present invention is based on data learned by the first artificial intelligence model. ) can be classified by the above symbols.

The first artificial intelligence model may be a number and character recognition model. The first artificial intelligence model can be trained to classify numbers using number images and labeling data of '0' to '9'. Additionally, the first artificial intelligence model can be trained to classify symbols using symbol images and labeling data such as 'X', '□', and '△'. At this time, the type of symbol being learned is not limited. Other characteristics of the first artificial intelligence model are the same as those of the second artificial intelligence model, which will be described later, so they will be omitted.

The processor 1300 can classify the numbers in the numeric section 31 and the symbols in the classification section 32 using the first artificial intelligence model. This corresponds to the process of recognizing numbers and symbols, which are just images, so that they can be used in the electronic device 1000.

Referring to FIG. 5, the relationship between the symbol and the type of the processed figure 300 can be seen. If the symbol is ' In case of 'ø' or '○', the type of processed figure (300) corresponds to a circle.

As in the above-described example, the processor 1300 can distinguish the type of the machined figure 300 according to the symbol of the classification unit 32. Types of symbols and processed figures 300 can be stored as a symbol list. Additionally, the above-described symbols are illustrative and are not necessarily limited thereto, and may be removed or added to the symbol list.

Next, we will look at the process of saving the drawing file.

The drawing generation method according to an embodiment of the present invention may further include a storage step (S1400) of storing the obtained drawing information in a memory in the form of a drawing file.

The processor 1300 can acquire drawing information in the data generation step (S1300) and use the acquired drawing information to store it in memory in the form of a drawing file. A drawing file may consist of at least one of STEP, STP, STL, and DWG. However, drawing files are not limited to the listed examples and may have various extensions that can be used as computer programs.

Next, we will look at the data generation step (S1300).

The data generation step (S1300) according to an embodiment of the present invention includes a type creation step (S1310) of determining the type of the machined figure 300 according to the symbol of the classification unit 32 and the determined machined figure 300. ) may include a shape creation step (S1320) of determining the size and shape of the machined figure 300, depending on the type and the numerical part 31.

First, a type creation step (S1310) of determining the type of the processed figure 300 may be performed according to the symbol of the classification unit 32.

The processor 1300 may determine the type of the machined figure 300 according to the symbol displayed in the classification unit 32. When the symbol of the classification unit 32 symbolizes a square, the type of the processed figure 300 is determined to be a square, and the numbers in the numeric part 31 can be used as data forming a square.

Additionally, when the symbol of the classification unit 32 symbolizes a triangle, the type of the processed figure 300 is determined to be a triangle, and the numbers in the numerical unit 31 can be used as data forming a triangle.

In addition, when the symbol of the classification unit 32 symbolizes a circle, the type of the processed figure 300 is determined as a circle, and the numbers of the numerical unit 31 can be used as data forming a circle. Furthermore, the same applies to polygons other than triangles and squares.

Subsequently, a shape creation step (S1320) of determining the size and shape of the machined figure 300 may be performed according to the determined type of the machined figure 300 and the numerical part 31.

Figure 6 is a diagram showing an example of a processed figure recognized as a square. Figure 7 is a diagram showing an example of a processed figure recognized as a triangle. Figure 8 is a diagram showing an example of a processed figure recognized as a circle.

The processor 1300 can recognize the numerical part 31 and determine the size and shape of the machined figure 300. When the processor 1300 recognizes the numerical unit 31 and the classification unit 32, the arrangement order of the numerical unit 31 and the classification unit 32 is not concerned. In addition, the distinction between the numerical part 31 and the distinction between the numerical part 31 and the classification part 32 may be expressed as spaces, such as ',' and '.' It can be expressed in various ways, such as: However, it is not limited anymore.

Referring to FIG. 6(a), the numeric part 31 may be made up of '200' and '100', and the classification part 32 may be made up of 'X'. At this time, unlike Figure 2, the numeric part 31 and the classification part 32 are described in a different arrangement as '200 100 It can mean.

Referring to FIG. 6(b), the numerical part 31 may be composed of '200' and '100', and the classification part 32 may be composed of '□'. '200 □ 100' may mean a square with a horizontal length of '200' and a vertical length of '100'.

Referring to FIG. 7(a), the numerical part 31 may be composed of '200', '100', and '150', and the classification part 32 may be composed of '△'. ‘200 100 150 △” may mean a triangle whose three side lengths are ‘200’, ‘100’, and ‘150’, respectively.

Referring to FIG. 7(b), the numerical part 31 may be composed of '200', '100', and '45°', and the classification part 32 may be composed of '△'. At this time, ',' distinguishes the numerical part 31 from each other, and distinguishes the numerical part 31 and the classification part 32 from each other, but the means of distinction is not limited to ','. '200, 100, 45°' may mean a triangle where the lengths of the two sides are '200' and '100', respectively, and the angle between them is '45°'.

Referring to FIG. 7(c), the numerical part 31 may be composed of '200', '30°', and '60°', and the classification part 32 may be composed of '△'. '△ 200 30°60°' may mean a triangle where the length of one side is '200' and the angles at both ends are '30°' and '60°', respectively.

Referring to FIG. 8(a), the numerical part 31 may be made of '100' and the classification part 32 may be made of 'R'. '100 R' may mean a circle with a radius of '100'.

Referring to FIG. 8(b), the numerical part 31 may be made of '200' and the classification part 32 may be made of 'ø'. 'ø 200' may mean a circle with a diameter of '200'. The classification unit 32 can recognize the numerical unit 31 as a radius or a diameter depending on preference.

Referring to FIG. 8(c), the numeric part 31 may be made up of '100' and the classification part 32 may be made up of '○'. '100 ○' may mean a circle with a radius of '100'.

Referring to FIG. 8(d), the numeric part 31 may be made of '100' and '50', and the classification part 32 may be made of '○'. '○ 100 50' may mean an ellipse with a major axis length of '100' and a minor axis length of '50'.

As various examples shown in FIGS. 6 to 8, the processor 1300 determines the type of the machined figure 300 using the classification unit 32 and determines the type of the machined figure 300 using the numerical part 31. The size and shape can be determined.

In addition, the processor 1300 is not limited to the arrangement of the numerical unit 31 and the classification unit 32 and can obtain drawing information by recognizing the numerical unit 31 and the classification unit 32. In addition, the processor 1300 is not limited to the type of the classification unit 32 and can determine the type of the processed figure 300 by recognizing the classification unit 32.

This can be achieved by the processor 1300 recognizing the numerical part 31 and the classification part 32, respectively. In other words, this creates the effect that correspondence is possible even when different users have different arrangements of the numeric part 31 and the classification unit 32 or write the symbols of the classification unit 32 differently.

Next, we look at the type creation stage.

Referring again to FIGS. 5 to 8, in the type creation step according to an embodiment of the present invention, the type of the processed figure 300 can be determined by finding a match in the symbol list that matches the symbol of the classification unit 32. .

The first artificial intelligence model of the processor 1300 classifies the symbols in the classification unit 32, and the processor 1300 can determine the type of the processed figure 300 by finding a match in the symbol list that matches the classified symbol. At this time, the symbol list can be added, deleted, and changed. That is, classification of symbols is performed by the first artificial intelligence model, and determining the type of the processed figure 300 can be performed by the processor 1300.

Next, we look at the notification step.

The drawing generation method according to an embodiment of the present invention cannot recognize at least one of the numerical part 31 and the classification part 32 in the image recognition step (S1200), or the data generation step (S1300) If the type of the processed figure 300 cannot be obtained from the classification unit 32, a notification step of providing a notification to correct the first image may be further included.

A case in which at least one of the numerical part 31 and the classification part 32 cannot be recognized may include a case where there is only the numerical part 31 or only the classification part 32, and the numerical part 31 and the classification part 32 are not recognized. This may include a case where part 32 is not written in the correct format, and may include a case where the symbol matching rate is less than 0.5 when the first artificial intelligence model classifies the symbol. However, it is not limited to this. In this case, a notification may be provided so that the first image including the numerical part 31 and the classification part 32 can be corrected.

Furthermore, cases where the type of the processed figure 300 cannot be obtained from the classification unit 32 may include cases where there is no corresponding symbol in the symbol list. However, it is not limited to this. In this case, a notification may be provided so that the first image including the numerical part 31 and the classification part 32 can be corrected. At this time, the notification may be provided through the output unit 1200.

In this way, the notification step has the effect of increasing the accuracy of drawings created according to the present invention.

Next, a drawing generating device according to an embodiment of the present invention includes a processor 1300, which has a numerical part 31 including numbers and a classification part 32 including symbols. Acquire 1 image, recognize the numerical portion 31 and the classification portion 32 from the first image, and determine the type of the processed figure 300 from the recognized numerical portion 31 and the classification portion 32. , drawing information about size and shape can be obtained.

The drawing generating method described above may be performed by the processor 1300, and the drawing generating device may include the processor 1300. Additionally, the drawing generating device may be the electronic device 1000.

Next, let's take a closer look at how to create drawings.

Figure 9 is a flowchart showing a drawing generation method using a second image according to an embodiment of the present invention. Figure 10 is a diagram showing the analysis area divided in the second image of the roadmap.

Referring to FIG. 9, the drawing generation method according to an embodiment of the present invention includes a second image acquisition step (S2100) of acquiring a second image of a road surface including a road shape; A road shape information acquisition step (S2200) of acquiring road shape information about the road shape on the second image; and a first final shape acquisition step (S2300) of acquiring the first final shape 150 using the temporary shape information.

First, a second image acquisition step (S2100) of acquiring a second image of a road including a road shape may be performed.

The street shape 11 may be made of a polygon such as a triangle or square. However, it is not limited to this. The street shape 11 may be formed in a two-dimensional form. Additionally, the gable shape 11 may be a thin material such as glass, plywood, or steel plate. However, it is not limited to this. Furthermore, the street shape 11 does not directly reflect the material, but can be shown in a thin two-dimensional form.

The drawing creation method can process temporary drawings. Here, processing of the rough surface means formalization of the rough surface. Provisional drawings are drawings drawn manually or in an arbitrary manner, so they may require formalization. Forming means processing a triangular shape (11) that is not completely straight into a triangular shape (11) that is completely straight.

In other words, regularization may mean processing it into a straight line if the line forming the slanted shape 11 is curved, and if the line forming the slanted shape 11 is discontinuous, it may mean processing it into a continuous state. can do.

The second image may be obtained by photographing the sketch, and may be obtained from the sketch drawn by the electronic device 1000. That is, the second image may include the road shape 11. The second image may be stored in the memory 1100 as a file. At this time, the second image file may be created by photographing the sketch using the photographing means of the electronic device 1000, and may be generated from a diagram directly drawn by the electronic device 1000, and may be sent to the communication unit 1500. It can be transmitted from an external device. Here, the external device may be a computing device such as a mobile device, laptop, desktop, or kiosk, but is not limited to those listed.

The electronic device 1000 and/or the processor 1300 according to an embodiment of the present invention may acquire a second image of a conspicuous surface including the conspicuous shape 11 and standardize the conduit shape 11 or the conduit surface. there is.

Subsequently, a road shape information acquisition step (S2200) of acquiring road shape information regarding the road shape 11 on the second image may be performed by the electronic device 1000 or the processor 1300.

Referring to FIG. 10 , the electronic device 1000 can specify an analysis area 70 and obtain road shape information for each analysis area 70 . The hypothetical form information may include information about the hypothetical form (11), the false number (12), and the primary symbol. The street shape 11 may include a plurality of lines forming the outline of the street shape 11. The mantissa number 12 may be length information corresponding to the plurality of lines. The primary symbol may be information on the angle formed between the plurality of lines. Additionally, the main symbol of each display may include angle information.

The unit of all mantissa numbers 12 appearing in the embodiments of the present invention may be at least one of mm, cm, and m, but is not limited thereto.

That is, the electronic device 1000 can obtain street shape information from the second image. Here, the street shape information may include the shape of the street shape 11, a plurality of lines forming the outline of the street shape 11, and mantissa numbers 12, etc. At this time, the road shape information may be stored in the memory 1100.

Subsequently, the first final shape 150 acquisition step (S2300) of obtaining the first final shape 150 by applying the symbol information to the geometric shape information may be performed by the electronic device 1000 or the processor 1300. there is.

Figure 11 is a diagram showing the process of forming the first final shape based on the shape information.

Referring to Figures 10 and 11, the road shape information may include a road shape (11). Additionally, the constellation information may include a constellation shape (11) and a constellation number (12). In other words, the processor 1300 can obtain the first final shape 150 using the triangular shape 11. In addition, the processor 1300 may obtain the first final shape 150 using the diagonal shape 11 and the addendum 12.

When the second image includes only the horizontal shape 11, the first final shape 150 may only specify the shape of the horizontal shape 11, and the size of the horizontal shape 11 may not be specified. Conversely, when the second image includes the consonant shape 11 and the mantissa 12, the first final figure 150 can specify both the shape of the conspicuous shape 11 and the size of the conspicuous figure 11. .

Next, we look at the road shape information acquisition step (S2200).

In the step of obtaining the street shape information (S2200) according to an embodiment of the present invention, a plurality of lines forming the outline of the street shape 11 can be recognized based on the color of the pixel of the second image. The method of recognizing a plurality of lines in the second image is the same as the method of recognizing a plurality of lines in the third image, and specific details regarding this will be described later.

Next, we look at the correction step.

The drawing generation method according to an embodiment of the present invention further includes a correction step of removing the distortion element when the gauze shape 11 has a distortion element according to the gavage shape information obtained in the gavage shape information acquisition step. It can be included. Specific details regarding this will be described later.

Next, we will look at the drawing file creation step (S2300).

The drawing generation method according to the embodiment of the present invention may further include generating a drawing file using the first final shape 150 (S2300). The method of generating a drawing file using the first final shape 150 is the same as the method of generating a drawing file using the second final shape 200, and specific details regarding this will be described later.

Next, the drawing generating device according to the embodiment of the present invention includes a processor 1300, where the processor 1300 acquires a second image of the drawing including the drawing 11, and the second It is possible to obtain street shape information about the street shape 11 on the image.

The drawing generating method described above may be performed by the processor 1300, and the drawing generating device may include the processor 1300. Additionally, the drawing generating device may be the electronic device 1000.

Next, we will look further at the drawing generation method according to an embodiment of the present invention.

Figure 12 is a flowchart showing a drawing generation method using a third image according to an embodiment of the present invention. Figure 13 is a diagram showing a third image of a road plan including a plurality of road shapes. Figure 14 is a diagram showing the analysis area divided in the third image of the roadmap. Figure 15 is a diagram showing symbols recognized within a region of interest.

Referring to FIG. 12, the drawing generation method according to an embodiment of the present invention is performed by the processor 1300, and the third drawing including the drawing shape 11 and the symbol 20 A third image acquisition step of acquiring an image (S3100); A road shape information acquisition step (S3200) of acquiring road shape information about the road shape 11 on the third image; A symbol information acquisition step (S3300) of acquiring symbol information about the symbol 20 on the third image; and a second final shape 200 acquisition step (S3400) of obtaining the second final shape 200 by applying the symbol information to the random shape information.

First, a third image acquisition step (S3100) of acquiring a third image of a tracheal surface including the tractive pattern 11 and the symbol 20 may be performed by the electronic device 1000.

Referring to FIG. 13, the street shape 11 may be made of a polygon such as a triangle or square. However, it is not limited to this. The street shape 11 may be formed in a two-dimensional form. Additionally, the gable shape 11 may be a thin material such as glass, plywood, or steel plate. However, it is not limited to this. Furthermore, the street shape 11 does not directly reflect the material, but can be shown in a thin two-dimensional form.

The symbol 20 may be depicted on the horizontal shape 11 or may be shown so that at least a portion of the symbol 20 overlaps the horizontal shape 11. The symbol 20 may indicate a state in which a material is scheduled to be processed. The symbol 20 is not limited in type, but may be mainly depicted to comply with symbol conventions. However, the symbol 20 does not necessarily have to be drawn in accordance with the symbol convention, but in this case, it may be requested by the electronic device 1000 to be drawn again. Related details will be described later.

The traverse surface may include a trajectories 11 and a symbol 20. A temporary drawing refers to a drawing drawn temporarily. The roadway can be drawn manually. Additionally, the drawing may be drawn by a device such as a computer. Specifically, the provisional drawing may be a drawing temporarily drawn before being processed by the drawing generating device according to an embodiment of the present invention.

The drawing creation method can process temporary drawings. Here, processing of the rough surface means formalization of the rough surface. Provisional drawings are drawings drawn manually or in an arbitrary manner, so they may require formalization. Forming means processing a triangular shape (11) that is not completely straight into a triangular shape (11) that is completely straight. In addition, formalization means applying the processing schedule indicated by the symbol 20 to the road shape 11. The standardized second final shape 200 may be similar to a preview of the processing state of the rough shape.

In other words, regularization may mean processing it into a straight line if the line forming the slanted shape 11 is curved, and if the line forming the slanted shape 11 is discontinuous, it may mean processing it into a continuous state. can do.

The third image may be obtained by photographing the sketch, and may be obtained from the sketch drawn by the electronic device 1000. That is, the third image may include a road shape (11) and a symbol (20). The third image may be stored in the memory 1100 as a file. At this time, the third image file may be created by photographing the sketch using the photographing means of the electronic device 1000, and may be generated from a diagram directly drawn by the electronic device 1000, and may be generated from the diagram directly shown by the electronic device 1000. It can be transmitted from an external device. Here, the external device may be a computing device such as a mobile device, laptop, desktop, or kiosk, but is not limited to those listed.

The electronic device 1000 according to an embodiment of the present invention may acquire a third image of a conspicuous surface including a conduit 11 and a symbol 20, and standardize the conduit 11 or the conduit 20. .

Subsequently, a road shape information acquisition step (S3200) of acquiring road shape information regarding the road shape 11 on the third image may be performed by the electronic device 1000.

Referring to FIG. 14 , the electronic device 1000 can specify the analysis area 70 and obtain road shape information for each analysis area 70 . The hypothetical form information may include information about the hypothetical form (11), the false number (12), and the primary symbol. The street shape 11 may include a plurality of lines forming the outline of the street shape 11. The mantissa number 12 may be length information corresponding to the plurality of lines. The primary symbol may be information on the angle formed between the plurality of lines. Additionally, each display symbol may include angle information.

The unit of all mantissa numbers 12 appearing in the embodiments of the present invention may be at least one of mm, cm, and m, but is not limited thereto.

That is, the electronic device 1000 can obtain street shape information from the third image. Here, the street shape information may include the shape of the street shape 11, a plurality of lines forming the outline of the street shape 11, and mantissa numbers 12, etc. At this time, the road shape information may be stored in the memory 1100.

Subsequently, a symbol information acquisition step (S3300) of acquiring symbol information about the symbol 20 on the third image may be performed by the electronic device 1000.

Referring to FIG. 15, symbol information may include information about the symbol 20. The symbol 20 may mean a state in which processing is scheduled, and may include a graphic portion 21 and a symbol numerical portion 22.

One symbol 20 may be shown in one conspicuous shape 11, but a plurality of symbols 20 may be shown in one conspicuous shape 11. The fact that a plurality of symbols 20 are shown means that a plurality of areas to be processed exist in one material.

The symbol information acquisition step (S3300) may be carried out after the road shape information acquisition step (S3200), may be carried out simultaneously with the road shape information acquisition step (S3200), and may be carried out before the road shape information acquisition step (S3200). Finally, symbol information and road shape information can be used in the process of processing the drawing shown on the road map.

Subsequently, a second final shape 200 acquisition step (S3400) in which the second final shape 200 is obtained by applying the symbol information to the geometric shape information may be performed by the electronic device 1000.

Figure 16 is a diagram showing the process of forming a second final shape from an intermediate shape. Figure 17 is a diagram showing the second final shape.

Referring to FIG. 16, symbol information can be applied to street shape information. The symbol 20 is shown on or at least partially overlapped with the convex shape 11, and indicates a state in which processing is scheduled.

Specifically, the electronic device 1000 can form the intermediate figure 100 using the triangle shape information. The intermediate figure 100 means that the symbol 20 is shown as is on the horizontal figure 11. The intermediate figure 100 may have distortion elements that will be described later removed.

Referring to FIG. 17, the electronic device 1000 can form the second final figure 200 by removing the symbol 20 from the intermediate figure 100 and applying the removed symbol 20 to the removed position. there is. In other words, the second final shape 200 means the final shape 11 in a standardized state by applying the processing schedule information indicated by the symbol 20 to the graphical shape information.

The electronic device 1000 may obtain a third image of the sketch, obtain geometry information and symbol information, and apply the symbol information to the geometry information to form the second final shape 200. That is, the electronic device 1000 can formalize the rough drawing into the second final figure 200.

Next, we will look at the road shape information acquisition process (S3200).

The step of obtaining the street shape information (S3200) according to an embodiment of the present invention is a step of recognizing a plurality of lines forming the outline of the street shape 11 based on the color of the pixel of the third image (S3210). ; Recognizing a plurality of addends 12 corresponding to the plurality of lines (S3220); And it may include a step (S3230) of acquiring the hypothetical shape information based on the recognized plurality of lines and the recognized plurality of pseudodigits 12.

First, a step (S3210) of recognizing a plurality of lines forming the outline of the diagonal shape 11 based on the color of the pixel of the third image may be performed by the electronic device 1000.

In order to obtain street shape information, the electronic device 1000 may recognize the outline of the street shape 11 according to the color of the pixels of the third image. The electronic device 1000 can recognize lines and non-line parts based on the color of the pixel. The step of recognizing the plurality of lines (S3210) may be performed through a line recognition algorithm.

For example, pixels corresponding to black may be recognized as a line, and pixels corresponding to white may be recognized as a background. However, the colors of the line and background are not limited to black and white, respectively, and can be any color that is distinct from each other.

Additionally, before the step of recognizing a plurality of lines (S3210), a step of gray scaling the third image or a step of binarizing the third image may be performed. Grayscaling or binarization makes it easy to distinguish between lines that are the target of recognition and the background that is not the target of recognition. In particular, when binarizing the third image, the line can be converted to white and the background can be converted to black.

That is, in order to facilitate line recognition, the third image may be expressed in grayscale or expressed in only two colors, black and white.

Subsequently, a step (S3220) of recognizing a plurality of mantissa numbers 12 corresponding to the plurality of lines may be performed by the electronic device 1000.

The plurality of lines refers to the outer lines of the road shape (11). Accordingly, since a plurality of mantissa numbers 12 corresponding to a plurality of lines may be shown together in the drawing, the mantissa numbers 12 need to be recognized along with the plurality of lines.

The electronic device 1000 can recognize each addend number 12 corresponding to each line. Recognition of the mantissa number (12) can be performed by a mantissa number recognition model. The mantissa recognition model is a second artificial intelligence model and can be learned in advance using the mantissa (12) data set. The addend number recognition model can recognize the addendum numbers 12 on the third image based on the learned data.

Subsequently, a step (S3230) of acquiring the hypothetical shape information based on the recognized plurality of lines and the recognized plurality of pseudodigits 12 may be performed.

The electronic device 1000 can recognize a plurality of lines and can recognize a plurality of mantissa numbers 12 corresponding to the plurality of lines. The electronic device 1000 may obtain the horizontal shape information using information about a plurality of lines and information about a plurality of mantissa numbers 12. At this time, the lateral shape information may include a plurality of lines, angles formed by the plurality of lines, and a plurality of mantissa numbers 12 corresponding to the plurality of lines. Furthermore, the horizontal shape information may include an angle formed by a plurality of lines and a false number 12 indicating the angle.

When the size of an angle is displayed at a corner of the diagonal shape 11, the electronic device 1000 may recognize the display as the angle of the corner. Conversely, when the size of the angle is not displayed at the corner of the horizontal shape 11, the electronic device 1000 may recognize the angle of the vertex as 90 degrees as horizontal shape information. Accordingly, the drawing generation method according to an embodiment of the present invention can reduce material manufacturing and processing time by standardizing the irregularly drawn street shape 11.

The electronic device 1000 can obtain the geometry information and use it as basic data for forming the second final shape 200.

Next, we will look at the step of correcting the road shape 11 on the road surface.

Figure 18 is a diagram showing an example of a distortion element.

Referring to FIG. 18, in the drawing generation method according to an embodiment of the present invention, when the horizontal shape 11 has a distortion element according to the horizontal shape information obtained in the horizontal shape information acquisition step (S3200), the distortion A correction step for removing elements may be further included.

The electronic device 1000 may obtain road shape information including the road shape 11 from the third image of the road surface. In this case, the road shape 11 may include a distortion element. Distortion elements can be divided into types 1 to 4. However, it is not limited to this.

The first type refers to a case where the wavy shape 11 includes a curved portion 13, as shown in FIG. 18(a). At this time, the electronic device 1000 may perform correction to change the curved portion 13 into a straight line.

The second type refers to a case where the convex shape 11 includes a protrusion 14 as shown in FIG. 18(b). At this time, the electronic device 1000 may perform correction to remove the protrusion 14.

The third type refers to a case where the angle shape 11 includes a corner 15 that is not 90 degrees, as shown in FIG. 18(c). The case of including a corner 15 other than 90 degrees means a case where there is no indication of an angle at the corner of the rectangular shape 11. At this time, the electronic device 1000 can correct the corner 15 that is not 90 degrees by changing it to 90 degrees.

The fourth type refers to a case where the diagonal shape 11 includes a cut portion 16, as shown in FIG. 18(d). At this time, the electronic device 1000 can make corrections by extending and connecting the line corresponding to the cut portion 16 of the diagonal shape 11.

Furthermore, one constellation 11 may include a plurality of distortion elements. At this time, the electronic device 1000 can correct all of the plurality of distortion elements.

When drawing the triangular shape 11 by hand, the triangular shape 11 may include at least one of a curved portion 13, a protruding portion 14, a corner 15 other than 90 degrees, and a cut portion 16, as described above. Therefore, the electronic device 1000 can obtain a standardized traversal shape 11 through a correction step to remove distortion elements. This can omit the process of drawing a new 2D drawing, thereby reducing time and cost in the process of manufacturing and processing materials.

Next, we look at the symbol information acquisition step (S3300).

The symbol information acquisition step (S3300) according to an embodiment of the present invention includes specifying a region of interest (60) including the horizontal shape (11) and the symbol (20) on the third image (S3310); Classifying the symbol 20 within the region of interest 60 using a second artificial intelligence model (S3320); and obtaining the symbol information for the classified symbol 20 (S3330).

First, the electronic device 1000 may perform a step (S3310) of specifying the region of interest 60 including the wavy shape 11 and the symbol 20 on the third image.

The electronic device 1000 may search for the symbol 20 within the region of interest 60. When the third image includes a plurality of road shapes 11, a plurality of analysis areas 70 each including a plurality of road shapes 11 may be specified, so the region of interest 60 is a plurality of analysis areas. It may mean one of the areas of (70).

The region of interest 60 may include a road shape 11 and a symbol 20. The electronic device 1000 may find a symbol 20 that is located on the conduit 11 or that at least partially overlaps the conduit 11 within the region of interest 60 . In other words, this step specifies one analysis area 70 as the area of interest 60 in order to find the symbol 20 corresponding to the corresponding road shape 11 among the plurality of analysis areas 70.

Figure 19 is a diagram showing a standardized state of a rough drawing. Figure 20 is a diagram showing a standardized state of a rough drawing.

Subsequently, a step (S3320) of classifying the symbol 20 within the region of interest 60 using a second artificial intelligence model may be performed by the electronic device 1000.

The second artificial intelligence model refers to the learned model. The second artificial intelligence model can be learned using a symbol image representing the symbol 20. The symbol image representing the symbol 20 is labeling data and can train a second artificial intelligence model. Additionally, the second artificial intelligence model can be learned using pseudo-number images and/or character images. Number images and/or character images can be used to train a second artificial intelligence model as labeling data.

The learned second artificial intelligence model can classify the symbol 20 and classify pseudo numbers related to the symbol 20. An example of the symbol 20 can be seen to the left of the arrow in FIGS. 19(a) to 19(d) and 20(e) to 20(h). The symbol 20 is not limited to the examples described below.

The symbol 20 in FIG. 19(a) means drilling. '30' means the distance from the left line of the gable shape (11), '70' means the distance from the lower line of the gauze shape (11), and '*?*50' is the circumference of the circle to be drilled. means.

The symbol 20 in Figure 19(b) indicates that the corners will be rounded off. 'R10' means curvature with a radius of 10. In other words, the symbol 20 means that the corner of the gable shape 11 is cut to a curvature of R10.

The symbol 20 in FIG. 19(c) indicates that the corner will be cut in a straight line. '100*100' is written in a horizontal*vertical form and means the length of 100 horizontal and 100 vertical. In other words, the symbol 20 means that the corners of the gable shape 11 are cut to a size of 100 in width and 100 in height.

The symbol 20 in Figure 19(d) indicates that the corner will be cut in a straight line. '100' refers to the length of the newly formed line after cutting the corner. That is, the symbol 20 means that the corner of the gable shape 11 is cut into a straight line of 100 length.

The symbol 20 in FIG. 20(e) indicates that the corner will be cut. '100' means the distance from the upper line of the gable shape 11 to the line extending outward from the gauze shape 11 in FIG. 20(e), and '50' means the length of the corresponding line. . That is, the symbol 20 means that the corner is to be cut as shown on the right side of the arrow in FIG. 20(e).

The symbol 20 in FIG. 20(f) indicates that the corner will be cut. '100' means the distance from the upper line of the gable shape 11 to the line extending outward from the gauze shape 11 in FIG. 20(f), and '50' means the length of the corresponding line. . That is, the symbol 20 means that the corner is to be cut as shown on the right side of the arrow in FIG. 20(f).

The symbol 20 in Figure 20(g) indicates that the corners will be cut into a square shape. '100*100' is written in a horizontal*vertical form and means the length of 100 horizontal and 100 vertical. That is, the symbol 20 means that the corner of the gable shape 11 is cut to a length of 100 width and 100 length.

The symbol 20 in FIG. 20(h) indicates that the corners will be drilled in a square shape. '100' means the distance from the lower line to the corresponding line. '50' refers to the horizontal and vertical lengths of the square shape, respectively. That is, the symbol 20 means to perforate as shown on the right side of the arrow in FIG. 20(h).

As described above, the electronic device 1000 can find the symbol 20 within the region of interest 60 and classify the symbol 20 using a second artificial intelligence model. Since there may be a plurality of symbols 20 in the region of interest 60, the electronic device 1000 can classify the plurality of symbols 20 using the second artificial intelligence model. Additionally, the electronic device 1000 may classify letters and/or pseudo-numbers corresponding to the symbol 20 using the second artificial intelligence model.

Subsequently, a step (S3330) of acquiring the symbol information for the classified symbol 20 may be performed by the electronic device 1000.

The electronic device 1000 can obtain symbol information about the classified symbol 20. Additionally, it is possible to obtain symbol information about letters and/or pseudo-digits corresponding to the classified symbols 20. The symbol information may include data about the processing schedule using the symbol 20 and letters and/or pseudo-numbers. Such symbol information can be used in the second final shape 200 acquisition step (S3400), which will be described later.

Next, we will look at the second final shape 200 acquisition step (S3400).

Referring again to FIG. 16, the step (S3400) of acquiring the second final shape 200 according to an embodiment of the present invention includes forming an intermediate shape 100 using the obtained temporary shape information (S3410); removing a portion corresponding to the symbol 20 from the formed intermediate shape 100 (S3420); And it may include forming the second final shape 200 using the symbol information at the position where the symbol 20 is removed from the intermediate shape 100 (S3430).

First, a step (S3410) of forming an intermediate shape 100 using the obtained shape information may be performed. The electronic device 1000 may obtain road shape information and symbol information from the third image of the road surface. The electronic device 1000 may form the intermediate figure 100 using the acquired shape information and/or symbol information. The intermediate figure 100 refers to a triangular shape 11 composed of a triangular shape 11 and a symbol 20. The intermediate figure 100 may be composed of the geometric figure 11 and the symbol 20 on the initially acquired third image, and may be a corrected intermediate figure 100 that has undergone correction to remove distortion elements.

Subsequently, a step (S3420) of removing a portion corresponding to the symbol 20 from the formed intermediate figure 100 may be performed by the electronic device 1000.

Since the intermediate shape 100 includes both the symbol 20 and the horizontal shape 11, it is necessary to delete the symbol 20. Referring to the central drawing of FIG. 16, the electronic device 1000 can remove the symbol 20 of the intermediate figure 100. That is, the electronic device 1000 can remove the part that overlaps the symbol 20.

Subsequently, a step (S3430) of forming the second final shape 200 using the symbol information may be performed by the electronic device 1000.

Since the part overlapping with the symbol 20 has been removed from the second final shape 200, the electronic device 1000 applies the processing scheduled state indicated by the removed symbol 20 to the part from which the symbol 20 has been removed. 2The final figure 200 can be formed. The second final figure 200 refers to a drawing showing the outline of the actual material and the state to be processed. That is, the electronic device 1000 can form the second final figure 200 using the intermediate figure 100 and standardize the rough drawing.

In this way, through the second final drawing 200 acquisition step (S3400), the electronic device 1000 can automatically obtain drawing data for generating drawing files such as CAD. As a result, it is possible to save time performing drawing work through a schematic drawing drawn manually by the drawing generation method according to an embodiment of the present invention.

Next, we will look at the step (S3430) of forming the second final shape 200 using symbol information in more detail.

According to an embodiment of the present invention, the step of forming the second final shape 200 using the symbol information includes forming the shape portion 21 of the symbol information using the symbol numerical value portion 22 of the symbol information. This may be a step of substituting standardized data corresponding to the intermediate figure 100.

Referring again to FIG. 16, the symbol information may include a symbol numerical value portion 22 and a graphic portion 21. In addition, the symbol numerical value unit 22 refers to a number and/or a family character corresponding to the symbol 20, and may refer to a machining position, a machining size, etc. that indicate a scheduled machining state. The graphic portion 21 refers to the shape of the symbol 20 itself and may signify a state in which processing is scheduled.

The electronic device 1000 can obtain standardized data corresponding to the figure unit 21 using the symbol numerical unit 22, and substitute the obtained standardized data into the intermediate graphical shape 11. As shown on the far right side of FIG. 16, the standardized data is the machining of a temporary surface that involves cutting the lower right part of the cutting shape (11) in a straight line with the processing information of a cutting surface that cuts the upper left part of the cutting shape (11) in a curved line. You can have information.

That is, the electronic device 1000 can form the shape of the second final figure 200 through standardized data. Accordingly, the electronic device 1000 can secure standardization data from the third image of the symbol 20, thereby standardizing the rough drawing and shortening the time for performing drawing work.

Next, the drawing generation method according to an embodiment of the present invention may further include generating a drawing file using the second final shape 200 (S3500).

The electronic device 1000 can form the second final shape 200. The electronic device 1000 may generate a drawing file through the second final shape 200. Drawing files may include 2D or 3D images of materials. A drawing file may consist of at least one of STEP, STP, STL, and DWG. However, the drawing file is not limited to the listed examples and can be used as a computer program and may have various extensions capable of expressing the second final shape 200.

Next, we will look at a case in which the symbol 20 cannot be recognized.

According to an embodiment of the present invention, the drawing generation method provides, in the symbol information acquisition step, if the symbol 20 is not recognized, a notification signal to re-draw the symbol 20 on the rough drawing according to the symbol convention. It may further include a notification providing step (S3600) of providing to the output unit 1200.

The second artificial intelligence model can obtain symbol information about the symbol 20. However, there may be cases where the symbol 20 is not recognized by the second artificial intelligence model. If the symbol 20 is not recognized, it means that the graphic part 21 of the symbol 20 cannot be recognized or that an essential element of the symbol numerical part 22 of the symbol 20 is defective.

The case where the graphic portion 21 of the symbol 20 cannot be recognized may mean a case where the match rate of the symbol 20 is less than 0.5 when the second artificial intelligence model classifies the symbol 20. In addition, a case where an essential element of the symbol numeric part 22 of the symbol 20 is defective means a case where there is no addend corresponding to the symbol 20 or an incorrect family character is written.

If the symbol 20 is not recognized, the second final shape 200 may be formed with incorrect symbol information. To increase accuracy, the electronic device 1000 may provide a notification signal through the output unit 1200 to re-show the symbol 20 on the third image.

At this time, the electronic device 1000 may provide a notification signal to re-draw the symbol 20 according to the symbol convention. The symbol 20 that the electronic device 1000 can recognize may be limited to the type of symbol 20 stored in the memory 1100 and may be limited to the type of symbol 20 learned by the second artificial intelligence model. there is. In order to use symbol information as information that can be utilized by the electronic device 1000, the symbol 20 must follow the already promised symbol convention. An example drawn according to the symbol convention can be seen to the left of the arrow in FIGS. 19 and 20.

Figure 21 is a diagram showing a case where a symbol violates the symbol convention.

Referring to FIG. 21, a symbol 20 that is not shown according to the symbol convention is a meaningless symbol 20 and cannot be utilized by the electronic device 1000. Accordingly, the electronic device 1000 may provide a notification signal through the output unit 1200 to re-show the symbol 20 on the third image.

Unlike in FIG. 19(b), the symbol 20 in FIG. 21(a) does not contain 'R', which violates the symbol convention, and this is a symbol ( 20) may not be recognized.

The notation of the symbol 20 in Figure 21(b) violates the symbol convention because the graphic part 21 is a 'triangle' and the symbol numerical part 22 expresses a curvature of 'R10', which violates the symbol 20. This may apply to cases where it is not recognized. Additionally, in this case, the match rate may be less than 0.5 during the process of the second artificial intelligence model classifying the symbol (20).

Unlike in FIG. 19(b), the notation of the symbol 20 in FIG. 21(c) violates the symbol convention because '10' is not written, and this may correspond to a case where the symbol 20 is not recognized.

As described above, when the symbol 20 violates the symbol convention or when the symbol 20 is not recognized, the electronic device 1000 outputs a notification signal so that the symbol 20 can be displayed again according to the symbol convention. It can be provided to unit 1200.

In addition, the second artificial intelligence model can provide the output unit 1200 with an input guide for the graphic part 21 and the symbol numerical part 22 of the symbol 20 that conforms to the symbol convention with the highest matching rate. That is, when the symbol 20 is not recognized as shown in FIG. 21(a), the electronic device 1000 may provide an input guide as shown in FIG. 19(b). At this time, the user can refer to the input guide and redraw the drawing to comply with the symbol convention.

Ultimately, the electronic device 1000 can improve the recognition rate and accuracy of the drawing generation method through the step of providing a notification signal (S3600). Additionally, the electronic device 1000 can reduce the drawing task execution time by accurately forming the second final shape 200 and generating a drawing file.

Next, we will look at the case where the third image includes a plurality of diagonal shapes 11.

In the drawing generation method according to an embodiment of the present invention, the third image includes a plurality of diagonal shapes 11, and after the step of acquiring the third image, a plurality of diagonal shapes 11 are generated from the third image. A step of specifying a plurality of analysis areas 70 each including may be performed for each of the plurality of analysis areas 70.

Referring again to FIG. 4, the third image may include only one conspicuous shape 11, but may include a plurality of conspicuous shapes 11. At this time, the electronic device 1000 may specify a plurality of analysis areas 70 including each conspicuous shape 11 in order to recognize the plurality of conspicuous shapes 11 . A plurality of analysis areas 70 may be expressed as an individual unit, a region of interest 60.

To specify multiple objects, an object detection model can be used. The object detection model can recognize each road type 11 and specify a plurality of analysis areas 70 including each road shape 11. The object detection model may be a model made of artificial intelligence, and may be a representative model of the CNN series. The object detection model can be learned using a third image composed of various road shapes 11, and the learned object detection model can recognize various types of road shapes 11 and specify the analysis area 70. .

As a result, the time and cost of performing drawing work can be saved by standardizing a plurality of road shapes 11 at the same time using one third image of the road surface including a plurality of road shapes 11.

Below, we will look at a drawing generating device according to an embodiment of the present invention.

Figure 22 is a block diagram showing a drawing generating device according to an embodiment of the present invention.

Referring to FIG. 22, the drawing generating device according to an embodiment of the present invention includes a processor 1300, wherein the processor 1300 generates a drawing of a drawing including a drawing 11 and a symbol 20. Acquire a third image, acquire road shape information about the road shape 11 on the third image, obtain symbol information about the symbol 20 on the third image, and store the symbol information in the above. The second final shape 200 can be obtained by applying it to the geometry information.

A drawing generating device according to an embodiment of the present invention may include a processor 1300. The processor 1300 may perform the drawing generation method described above. Additionally, the drawing generating device according to an embodiment of the present invention may further include a memory 1100, and the memory 1100 may store a program that performs a drawing generating method. Additionally, the memory 1100 may store at least one of information about the first to third images, a drawing file, the generated shape 300, the first final shape 150, and the second final shape 200.

Additionally, the drawing generating device according to an embodiment of the present invention may further include an output unit 1200. The output unit 1200 may display the machined shape 300, the first final shape 150, and the second final shape 200. Additionally, the output unit 1200 can transmit a notification signal as an audio signal to the user through the sound output unit and transmit it to the user through the display unit. The output unit 1200 may deliver the input guide to the user through an audio output unit or a display unit.

Additionally, the drawing generating device according to an embodiment of the present invention may further include a communication unit 1500. The communication unit 1500 transmits at least one of the processed drawing 300, the first final drawing 150, and the second final drawing 200 information, a drawing file, and the first to third images to or from an external device. You can receive it.

Additionally, the drawing generating device according to an embodiment of the present invention may further include a photographing means 1600. The photographing means 1600 may acquire first to third images by photographing the numerical part, the classification part, and the rough plane.

Meanwhile, a computer-readable recording medium can record a program that performs the drawing generation method according to an embodiment of the present invention. The recording medium can operate a drawing generating device. The recording medium may be memory 1100.

The formation of the first final shape 150 is a process using the triangular shape 11 to which the symbol 20 is not applied, and the formation of the second final shape 200 is a process using the triangular shape 11 to which the symbol 20 is applied. This is the process used. The process of forming the first final shape 150 and the second final shape 200 can be interpreted equally except for the symbol 20. All parts of the second image and the third image can be interpreted the same except for the symbol 20.

In addition, the drawing generating method for forming the processed drawing 300, the first final drawing 150, and the second final drawing 200 may be performed by the processor 1300, and may specifically be performed by the drawing generating device. You can. Furthermore, the drawing generating device may be comprised of an electronic device 1000.

As described above, although the present invention has been described with reference to preferred embodiments, those skilled in the art may modify the present invention in various ways without departing from the spirit and scope of the present invention as set forth in the following patent claims. Or, it can be carried out in modification.

10: conspicuous form and mantissa 11: conspicuous form
12: Addendum 13: Curved part
14: Protrusion 15: Corner other than 90 degrees
16: cutting part 20: symbol
21: graphic part of symbol 22: symbol numerical part of symbol
31: Numerical unit 32: Discrimination unit
40: Quantity part 41: Quantity symbol
42: Quantity 50: Processing part
51: First processing matter 52: Second processing matter
53: Third processing matter 60: Area of interest
70: Analysis area 100: Intermediate figure
150: 1st final shape 200: 2nd final shape
300: Machining diagram 1000: Electronic device
1100: Memory 1200: Output unit
1300: Processor 1500: Communication unit
1600: Photography means

Claims (20)

In a drawing generation method performed by a processor,
A first image acquisition step of acquiring a first image having a numerical part including numbers and a classification part including symbols;
An image recognition step of recognizing the numerical part and the classification part from the first image; And
A data generating step of acquiring drawing information about the type, size, and shape of the processed figure from the recognized numerical part and classification part. A drawing generating method comprising a. According to claim 1,
The image recognition step is,
A drawing generation method that classifies the numbers in the numeric part and the symbols in the classification part based on data learned by a first artificial intelligence model. According to claim 1,
The first image further includes a quantity portion,
A management step of uploading the obtained drawing information and the quantity described in the quantity unit to an inventory management system. A drawing generating method further comprising a. According to claim 1,
The first image further includes a processing part,
A drawing generation method further comprising a management step of uploading the obtained drawing information and processing details described in the processing unit to an inventory management system. According to claim 1,
A drawing generation method further comprising a storage step of storing the obtained drawing information in a memory in the form of a drawing file. According to claim 1,
The data creation step is,
A type creation step of determining the type of the processed figure according to the symbol of the classification unit, and
A drawing generating method comprising a shape generating step of determining the size and shape of the machined figure according to the determined type of the machined figure and the numerical value. According to claim 6,
The type creation step is,
A drawing generation method that determines the type of the processed figure by finding a match in the symbol list for the symbol of the classification unit. According to claim 1,
If at least one of the numerical part and the classification unit cannot be recognized in the image recognition step, or the type of the processed figure cannot be obtained from the classification unit in the data generation step,
A drawing generation method further comprising a notification step of providing a notification to correct the first image. In the drawing generating device including a processor,
The processor acquires a first image having a numeric portion including numbers and a classification portion including symbols, recognizes the numerical portion and the classification portion from the first image, and processes the recognized numerical portion and the classification portion. A drawing generating device that obtains drawing information about the type, size, and shape of a figure. In a drawing generation method performed by a processor,
A second image acquisition step of acquiring a second image of a road surface including a road shape;
A road shape information acquisition step of acquiring road shape information about the road shape on the second image; and
A drawing generating method including; a first final shape acquisition step of acquiring a first final shape using the temporary shape information. According to claim 10,
The road type information acquisition step is,
A drawing generation method that recognizes a plurality of lines forming the outline of the street shape based on the color of the pixel of the second image. According to claim 10,
A drawing generation method further comprising: generating a drawing file using the first final shape. In the drawing generating device including a processor,
The processor acquires a second image of a trajectories including a trajectories, acquires trajectories information about the trajectories on the second image, and obtains a first final shape using the trajectories information. , drawing generation device. In a drawing generation method performed by a processor,
A third image acquisition step of acquiring a third image of the road surface including the road shape and symbols;
A road shape information acquisition step of acquiring road shape information about the road shape on the third image;
A symbol information acquisition step of acquiring symbol information about the symbol on the third image; and
A second final shape acquisition step of applying the symbol information to the random shape information to obtain a second final shape. According to claim 14,
The road type information acquisition step is,
Recognizing a plurality of lines forming an outline of the shape based on the color of pixels of the third image;
Recognizing a plurality of mantissa numbers corresponding to the plurality of lines; and
A drawing generating method including; acquiring the random shape information based on the recognized plurality of lines and the recognized plurality of addends. The method of claim 10 or 14,
If the horizontal shape has a distortion element according to the horizontal shape information acquired in the horizontal shape information acquisition step,
A drawing generation method further comprising a correction step of removing the distortion element. According to claim 14,
The symbol information acquisition step is,
specifying a region of interest including the road shape and the symbol on the third image;
Classifying the symbol within the region of interest using a second artificial intelligence model; and
A drawing generation method including; acquiring the symbol information for the classified symbols. According to claim 14,
A drawing generating method further comprising: generating a drawing file using the second final shape. In the drawing generating device including a processor,
The processor acquires a third image of the road surface including a road shape and a symbol, obtains road shape information about the road shape on the third image, and obtains symbol information about the symbol on the third image. A drawing processing device for obtaining a second final shape by applying the symbol information to the gauze shape information. A computer-readable recording medium on which a program for performing the method according to any one of claims 1 to 8, 10 to 12, and 14 to 18 is recorded.

Images