KR20230080999A - Device For Reading Frame And Its Method - Google Patents

Abstract

The present invention obtains an image including at least one frame and divides the image into a region of interest, thereby providing an effect of inspecting one or a plurality of frames and providing effects corresponding to frames of various shapes.

Description

Frame reading device and its method {Device For Reading Frame And Its Method}

The present invention relates to a frame inspection apparatus and method, and more particularly, to a frame inspection apparatus and method for determining the size of a material coupled to a frame from an image including a plurality of frames.

Although the term artificial intelligence first appeared in the 1970s, it is a field that is being studied very actively today as the amount of data increases, advanced algorithms, and computing power and storage have improved. Such artificial intelligence is currently being used in various fields such as defense, medical care, healthcare, education, games, and security.

On the other hand, attempts to measure the length, size, etc. of an object on an image by using an image are appearing in various fields. Conventionally, there is a method of measuring an area of an outdoor signboard using an image.

However, when an outdoor signboard is actually photographed from the outside, there is a problem in that manual work is required after photographing to specify the target signboard because adjacent signboards are often photographed together rather than when the target outdoor signage is photographed alone. In addition, there is a problem in that measurement is difficult when the outdoor sign has a different shape.

Accordingly, a method of extracting features of an object using artificial intelligence is being continuously studied.

Publication No. 10-2021-0037047 Patent Publication

In order to solve the above problems, an object of the present invention is to provide a frame inspection device and method for measuring the size of a virtual figure generated by dividing an image into at least one region of interest and connecting feature points extracted for each region of interest. to be

In order to achieve the above object, a frame inspection method according to an embodiment of the present invention, stored in a memory or performed by a processor, includes the steps of acquiring an image including at least one frame; dividing the image into at least one region of interest to be analyzed for each frame; extracting a plurality of feature points from each region of interest for each region of interest; and obtaining a virtual figure created using a virtual line connecting the plurality of feature points.

The dividing into the ROI may include extracting reference color information corresponding to the at least one frame from the image; extracting a plurality of edges of each frame using the reference color; and dividing the at least one ROI including one closed region formed by the plurality of corners.

In addition, in the extracting of the plurality of feature points, vertices of each frame within each region of interest may be extracted as the plurality of feature points.

The dividing into the ROI may include extracting reference color information corresponding to the at least one frame from the image; and extracting a plurality of support points connected to each frame using the reference color. and dividing the plurality of regions of interest including a plurality of supporting points connected only to the respective frames.

In addition, in the step of extracting the plurality of feature points, the plurality of support points within each region of interest may be extracted as the plurality of feature points.

Also, the virtual line may be a set of lines connecting the first feature point to the second feature point and the first feature point to the third feature point among the plurality of feature points.

In addition, the step of classifying into the region of interest may be characterized in that a first artificial intelligence model is used.

The frame inspection method may further include determining a size of a material to be combined with each frame based on the virtual figure.

In addition, the image includes a reference object whose absolute size is known, and in the obtaining of the virtual figure, the size of the virtual figure may be measured in consideration of a relative ratio between the virtual figure and the reference object. there is.

Meanwhile, in the frame inspection apparatus including a memory and a processor, an algorithm stored in the memory or executed by the processor may generate an image including at least one frame. acquisition, divides the image into a plurality of regions of interest to be analyzed for each frame, extracts a plurality of feature points from each region of interest for each region of interest, and generates them using a virtual line connecting the plurality of feature points It may include obtaining a virtual figure.

The frame inspection device may further include a photographing means, and the photographing means may acquire the image by photographing the at least one frame.

In addition, the frame inspection apparatus may further include an output unit, and the output unit may provide information including the acquired virtual figure.

The frame inspection apparatus may further include a communication unit, and the communication unit may acquire the image from an external device and provide information including the virtual figure to the external device.

The frame inspection apparatus further includes an input unit, wherein the input unit determines a size difference between the virtual figure and the material to be coupled to each frame to determine the size of the material to be coupled to each frame based on the virtual figure. You can enter the correction value corresponding to .

An apparatus and method for inspecting a frame according to the present invention acquires an image including at least one frame and divides the image into a region of interest, thereby providing an effect of inspecting one or a plurality of frames. A corresponding effect can be provided.

1 is a flowchart illustrating a frame inspection method.
2 is a diagram illustrating an image including a plurality of frames according to an embodiment of the present invention.
3 is a diagram illustrating an image including a plurality of frames according to an embodiment of the present invention.
4 is a flowchart illustrating a step of classifying a region of interest according to an embodiment of the present invention.
5 is a diagram illustrating a plurality of regions of interest according to an embodiment of the present invention.
6 is a flowchart illustrating a step of classifying a region of interest according to an embodiment of the present invention.
7 is a diagram illustrating a plurality of regions of interest according to an embodiment of the present invention.
8 is a diagram showing a plurality of feature points according to an embodiment of the present invention.
9 is a diagram showing the size of a virtual figure according to an embodiment of the present invention.
10 is a view showing a frame inspection device and materials installed using the method.
11 is a block diagram illustrating a frame inspection device.

Hereinafter, a frame inspection apparatus and method according to an embodiment of the present invention will be described.

A person skilled in the art can develop various devices that embody the principles of the invention and fall within the concept and scope of the invention, even though not explicitly described or shown herein. In addition, all conditional terms and examples listed in this specification are, in principle, expressly intended only for the purpose of making the concept of the invention understood, and should be understood as not being limited to such specifically listed examples and conditions.

The above objects, features and advantages will become more apparent through the following detailed description of the invention in conjunction with the accompanying drawings, and accordingly, those skilled in the art to which the invention belongs will be able to easily implement the technical idea of the invention.

Embodiments described in this specification will be described with reference to sectional views and/or perspective views, which are ideal exemplary views of the present invention. Films and thicknesses of regions shown in these drawings are exaggerated for effective description of technical content. The shape of the illustrative drawings may be modified due to manufacturing techniques and/or tolerances. Therefore, embodiments of the present invention are not limited to the specific shapes shown, but also include changes in shapes generated according to manufacturing processes.

While describing various embodiments, the same names and the same reference numbers will be given to components performing the same functions even if the embodiments are different. Furthermore, configurations and operations already described in other embodiments will be omitted for convenience.

10 is a block diagram illustrating a frame inspection device.

First, the configuration of a frame inspection device (hereinafter referred to as “electronic device”) used in the frame inspection method will be described.

As an embodiment of the present invention, the electronic device 1000 may be referred to as a terminal, a device, an electronic device, and the like. The electronic device 1000 includes a smart phone, a tablet PC, a PC, a smart TV, a mobile phone, a personal digital assistant (PDA), a laptop, a media player, a micro server, a global positioning system (GPS) device, an e-book reader, and a digital broadcasting terminal. , navigation, kiosks, MP3 players, digital cameras, home appliances and other computing devices, but are not limited thereto. Also, the electronic device 1000 may be a wearable device having a display function and a data processing function, such as a watch, glasses, a hair band, and a ring. However, it is not limited thereto, and the electronic device 1000 may include all types of devices capable of processing data and providing the processed data.

For example, as shown in FIG. 10 , an electronic device 1000 according to an 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, the electronic device 1000 may be implemented with more components, or the electronic device 1000 may be implemented with fewer components. may be

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

The memory 1100 may include a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (eg SD or XD memory, etc.), 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 , an optical disk, and at least one type of storage medium.

The output unit 1200 may output an audio signal, a video signal, or a vibration signal, and 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 photographed 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 . Also, the sound output unit outputs sound signals related to functions performed by the electronic device 1000 (eg, call signal reception sound, message reception sound, and notification sound).

The processor 1300 typically controls overall operations of the electronic device 1000 . For example, the processor 1300 may generally control the output unit 1200, the communication unit 1500, the photographing means 1600, and the like 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). Another 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 an infrared data association (IrDA) communication unit, a Wi-Fi Direct (WFD) communication unit, an ultra wideband (UWB) communication unit, an Ant+ communication unit, etc., but is not limited thereto.

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

The photographing unit 1600 may obtain an image frame such as a still image or a moving image through an image sensor in a video call mode or a photographing mode. An image captured by 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 capture devices, and is not limited to the types.

Next, a frame inspection method according to an embodiment of the present invention will be described.

1 is a flowchart illustrating a frame inspection method. 2 is a diagram illustrating an image 10 including a plurality of frames 100 according to an embodiment of the present invention. 3 is a diagram illustrating an image 10 including a plurality of frames 100 according to an embodiment of the present invention.

Referring to FIG. 1 , a frame inspection method according to an embodiment of the present invention may be performed by an electronic device 1000 and stored in a memory 1100 or performed by a processor 1300. In the method, obtaining an image 10 including at least one frame 100 (S100); Dividing the image 10 into at least one region of interest 200 to be analyzed for each frame 100 (S200); extracting a plurality of feature points 300 from each region of interest 200 for each region of interest 200 (S300); and acquiring a virtual figure 400 created using a virtual line connecting the plurality of feature points 300 (S400).

First, obtaining an image 10 including at least one frame 100 (S100) may be performed.

Referring to Figures 2 and 3, the frame 100 means a frame on which the material 500 can be seated, coupled, connected or applied. The frame 100 is not necessarily made of a rectangle, and may vary depending on the shape of the material 500. That is, the frame 100 may be formed in various shapes such as a polygonal shape and a circular shape.

Materials 500 applicable to the frame 100 may include glass, doors, partitions, or blinds. In this specification, glass will be mainly described as an example, but it is not necessarily limited to glass.

Image 10 may include at least one frame 100 . The image 10 may include one frame 100 of the image 10 or may include a plurality of frames 100 . In addition, the image 10 may be captured by the photographing unit 1600 of the electronic device 1000, may be received by the communication unit 1500 from an external device in the form of a file, and may be provided online in a download format. and may be provided by a storage medium such as USB. The acquired image 10 may be stored in the memory 1100 of the electronic device 1000 .

Subsequently, a step (S200) of dividing the image 10 into at least one ROI 200 to be analyzed for each frame 100 may be performed.

The electronic device 1000 may individually analyze each frame 100 . In order to individually analyze each frame 100 , the electronic device 1000 may divide the image 10 into a plurality of regions of interest 200 . By introducing the ROI 200 , the electronic device 1000 may analyze each frame 100 even when the image 10 includes a plurality of frame images 10 . That is, the present invention has the advantage of individually analyzing even when a plurality of frame images 10 are included.

For example, the image 10 according to FIG. 2 comprises seven glass frames 100 and one door frame 100 . In this case, the electronic device 1000 may recognize eight frames 100 by recognizing each frame 100 . In this case, if the frame 100 does not form a closed region, it may be excluded from the region of interest 200 .

The image 10 according to FIG. 3 comprises five glass frame images 10 . At this time, the electronic device 1000 may recognize five frames 100 by recognizing each frame 100 . At this time, since the leftmost frame 100 shown on the image 10 is cut off, it can be excluded from the ROI 200 in the case where a closed region is not formed. Through this, only the part requiring analysis can be classified as the region of interest 200 to be analyzed.

The division of the region of interest 200 may be implemented by a general algorithm, but may be implemented through an artificial intelligence model to be described later.

Next, a step of extracting a plurality of feature points 300 from each region of interest 200 for each region of interest 200 (S300) may be performed.

The feature point 300 means a feature point constituting the frame 100 . That is, in the image 10 of FIG. 2 , each ROI 200 may include the image 10 of each frame 100, and the feature point 300 may include an inner vertex of each frame 100. Also, in the image 10 of FIG. 3 , the feature point 300 may include a support point connected to each frame 100 .

That is, the electronic device 1000 may obtain a plurality of feature points 300 required for each of the divided regions of interest 200 . A virtual figure 400 to be described later may be acquired using a plurality of feature points 300, and the virtual figure 400 may be subject to measurement in the present invention.

Subsequently, a step of obtaining a virtual figure 400 created by using a virtual line connecting the plurality of feature points 300 (S400) may be included.

The electronic device 1000 may connect a plurality of feature points 300 using virtual lines and obtain a virtual figure 400 . The virtual figure 400 may include size information useful for determining the size of the material 500 to be measured and applied to the actual frame 100 . For example, the size of the virtual figure 400 means the horizontal and vertical lengths when the virtual figure 400 is a rectangle, and the lengths of the base and height when the virtual figure 400 is a triangle, and the mathematical represents the number of elements.

The size of the virtual figure 400 may be the same as the size of the material 500 to be applied to the actual frame 100, but may not be the same. For example, in the case of FIG. 2, the size of the virtual figure 400 and the frame 100 are almost the same, but in the case of FIG. 3, the size of the virtual figure 400 and the frame 100 cannot be regarded as the same.

At this time, the electronic device 1000 may receive a correction value through an input means. The correction value means a numerical value representing a difference between the size of the virtual figure 400 and the size of the material 500 to be applied to the actual frame 100 . The electronic device 1000 may determine the size of the material 500 to which the frame 100 is actually applied through the correction value. That is, a step of determining the size of the actual material 500 using the acquired virtual figure 400 may proceed.

Next, the step of dividing the ROI 200 in the image 10 as shown in FIGS. 2 and 3 will be described.

4 is a flowchart illustrating a step of classifying into a region of interest 200 according to an embodiment of the present invention. 5 is a diagram showing a plurality of regions of interest 200 according to an embodiment of the present invention. FIG. 6 is a flowchart illustrating a step of classifying into a region of interest 200 according to an embodiment of the present invention. 7 is a diagram illustrating a plurality of regions of interest 200 according to an embodiment of the present invention. 8 is a diagram showing a plurality of feature points 300 according to an embodiment of the present invention.

Referring to FIG. 4 , in step S200 of classifying into regions of interest 200 according to an embodiment of the present invention, information of a reference color corresponding to the at least one frame 100 is extracted from the image 10 Step (S210); Extracting a plurality of edges of each frame 100 using the reference color (S220); and dividing the at least one region of interest 200 including one closed region formed by the plurality of corners (S230).

First, a step S210 of extracting reference color information corresponding to the at least one frame 100 from the image 10 may be performed.

The electronic device 1000 may extract reference color information using a color corresponding to the frame 100 . Extraction of the reference color is for recognizing a color of the same or similar series as the reference color as the area of the frame 100 . The electronic device 1000 may identify the region of interest 200 by recognizing the same or similar color as the reference color as the frame 100 .

Information of the reference color may be extracted by an algorithm or artificial intelligence model. Reference color information may be extracted for each ROI 200 including each frame 100 . For example, although the plurality of frames 100 may be of the same color, since the plurality of frames 100 may be of different colors, the reference color information may be the same as the number of divided regions of interest 200. there is.

Subsequently, a step of extracting a plurality of edges of each frame 100 using the reference color (S220) may proceed.

The electronic device 1000 may recognize the image 10 having the same or similar color to the extracted reference color as the frame 100 . The recognized frame 100 is a combination of one material 500 and may be classified into each region of interest 200 .

Subsequently, a step of dividing the at least one region of interest 200 including one closed region formed by the plurality of corners (S230) may be performed.

Each of the plurality of edges extracted on the frame 100 may form one closed area. The electronic device 1000 may classify the ROI 200 based on the region where one closed region is formed.

Referring to FIG. 5 , edges recognized along colors of the same or similar series as the reference color constitute eight frames 100, and the electronic device 1000 includes a region of interest (ROI) each including the eight frames 100 ( 200) can be distinguished. The electronic device 1000 may extract feature points 300 within the eight regions of interest 200 .

Next, a step of extracting a plurality of feature points 300 from the image 10 as shown in FIG. 2 will be described.

8(a) and 8(b) , in the step of extracting the plurality of feature points 300 according to an embodiment of the present invention, each frame 100 within each region of interest 200 The vertices of can be extracted as the plurality of feature points 300 .

In the above-described method, the electronic device 1000 classifies each region of interest 200, and extracts the inner vertices of each frame 100 as a plurality of feature points 300 for each divided region of interest 200. there is. The plurality of feature points 300 may be vertices of the virtual figure 400 .

In summary, in the case of an image 10 including a frame 100 as shown in FIG. 2 , the electronic device 1000 first extracts color information of a pixel corresponding to the frame 100 to find a corner and find a closed area, The region of interest 200 including the closed region is divided to find the subject and number of analysis, and the inner vertex of the frame constituting the closed region for each region of interest 200 can be found and used as the feature point 300 . Finally, the electronic device 1000 may form the virtual figure 400 using the plurality of feature points 300 and measure the size of the virtual figure 400 . This can be done by algorithms or artificial intelligence models.

Accordingly, even when a plurality of frames 100 appear on the image 10, the size of the virtual figure 400 can be collectively measured.

Referring to FIG. 6 , in step S200 of classifying the region of interest 200 according to an embodiment of the present invention, information of a reference color corresponding to the at least one frame 100 in the image 10 is Extracting (S240); and extracting a plurality of support points connected to each frame 100 using the reference color (S250); and dividing the at least one region of interest 200 including a plurality of support points connected only to each frame 100 (S260).

First, a step of extracting reference color information corresponding to the at least one frame 100 from the image 10 (S240) may be performed. Since this step proceeds in the same way as the above method, it will be omitted.

Subsequently, a step of extracting a plurality of support points connected to each frame 100 using the reference color (S250) may proceed.

The electronic device 1000 may recognize the image 10 having the same or similar color to the extracted reference color as the frame 100 . The recognized frame 100 is a combination of one material 500 and may be classified into each region of interest 200 . In the image 10 shown in FIG. 3 , the electronic device 1000 may extract a plurality of support points connected to each frame 100 instead of extracting corners.

The support 110 refers to a point where the frame 100 couples, supports, or grips the material 500 . The support 110 may include a support point. Preferably, the support point may mean the center of the support 110 . The electronic device 1000 may divide the ROI 200 by using support points.

In the step of extracting a plurality of supporting points (S250), the supporting points on the image 10 may be detected by an algorithm or an artificial intelligence model, but may also be performed manually.

There may be a method of first obtaining the image 10 and applying the reflective mark 130 to the supporting point on the image 10, and applying the reflective mark 130 to the supporting point and taking a picture of the frame 100. 10) may be obtained.

As a result, when the electronic device 1000 lowers the contrast of the image 10 including the reflective mark 130, the point to which the reflective mark 130 is applied appears brightest, and the point that looks bright may be used as a plurality of fulcrum points.

Accordingly, the electronic device 1000 may automatically recognize a plurality of support points using the color information of the frame 100 or may recognize a plurality of support points using the reflective mark 130 .

Subsequently, a step of dividing the at least one region of interest 200 including a plurality of support points connected only to each frame 100 (S260) may be performed.

The electronic device 1000 may divide the ROI 200 based on a plurality of support points. The plurality of support points are parts in contact with the glass connected to the frame 100, and the number and shape of the material 500 to be analyzed may be determined.

Referring to FIG. 7 , the expression 'only for each frame' means that one ROI 200 includes a plurality of supporting points corresponding to one frame 100 when dividing the ROI 200 . That is, since the leftmost region of interest 200 includes four support points and the four support points are connected to one frame 100, the leftmost region of interest 200 is the leftmost frame 100. It can include four supporting points corresponding only to .

Referring back to FIG. 7 , a plurality of support points recognized along the same or similar series colors as the reference color constitute five frames 100, and the electronic device 1000 includes a region of interest (ROI) each including a plurality of support points ( 200) can be distinguished. The electronic device 1000 may extract feature points 300 within the five regions of interest 200 .

Next, a step of extracting a plurality of feature points 300 from the image 10 as shown in FIG. 3 will be described.

Referring to FIGS. 8(c) and 8(d) , in the step 300 of extracting the plurality of feature points 300 according to an embodiment of the present invention, the plurality of feature points within each region of interest 200 Support points may be extracted as the plurality of feature points 300 .

In the above method, the electronic device 1000 classifies each region of interest 200, and for each divided region of interest 200, a plurality of support points connected only to each frame 100 are extracted as feature points 300. there is. The plurality of feature points 300 may be vertices of the virtual figure 400 .

In summary, in the case of the image 10 including the frame 100 as shown in FIG. 3 , the electronic device 1000 first extracts color information of pixels corresponding to the frame 100 or support points to include a plurality of support points. It is possible to search for a set to do, to classify the region of interest 200 including each set, to find an object of analysis and to find a number, and to find a support point for each region of interest 200 and set it as the feature point 300 . Finally, the electronic device 1000 may form the virtual figure 400 using the plurality of feature points 300 and measure the size of the virtual figure 400 . This can be done by algorithms or artificial intelligence models.

Next, a virtual line constituting the virtual figure 400 will be described.

Referring to FIG. 8 , the virtual line according to an embodiment of the present invention may be a set of lines connecting the first feature point to the second feature point and the first feature point to the third feature point in the plurality of feature points 300. there is.

The virtual figure 400 may be a subject of measurement through the image 10 . A virtual line means an arbitrary line constituting the virtual figure 400 . The first feature point represents one arbitrary feature point 300 . The second feature point means the feature point 300 other than the first feature point, and the third feature point means the feature point 300 excluding the first feature point and the second feature point.

Specifically, the virtual line may include a first line connecting the first feature point to the second feature point and a second line connecting the first feature point to the third feature point. At this time, the second feature point and the third feature point selected may be selected so that the angle formed by the first line and the second line is the largest.

For example, looking at the feature points 300 located at the upper left corner, upper right corner, lower left corner, and lower right corner in the frame 100 shown in FIG. When the upper right corner and the third feature point are located at the lower left corner, the angle formed by the first line and the second line may be 90 degrees.

Conversely, when the first feature point is located at the upper left corner, the second feature point is located at the upper right corner, and the third feature point is located at the lower right corner, the angle formed by the first line and the second line may be smaller than 90 degrees. When comparing the two cases, the former case in which the first line and the second line maintain 90 degrees can be selected.

As such, the electronic device 1000 may obtain the virtual figure 400 by repeatedly connecting virtual lines using each feature point 300 as a first feature point. The virtual figure 400 is an object of measurement on the image 10, and the electronic device 1000 may acquire the size of the virtual figure 400 obtained on the image 10. That is, the user can easily perform the frame inspection method through the captured image 10 or the provided image 10 .

In addition, the step (S200) of dividing into the region of interest 200 according to an embodiment of the present invention may be characterized by using an artificial intelligence model.

An artificial intelligence model according to an embodiment of the present invention may be stored in the memory 1100. The artificial intelligence model may output the size of the virtual figure 400 when the image 10 is input.

Meanwhile, the frame inspection method according to an embodiment of the present invention may further include a learning step for learning an artificial intelligence model. Specifically, the artificial intelligence model can be learned using a large amount of images (10). In particular, the artificial intelligence model may learn the image 10 including at least one frame 100 .

The artificial intelligence model may be composed of a machine learning model, a deep learning model, and the like. In particular, the artificial intelligence model may be composed of R-CNN series, and specifically, may be composed of R-CNN, Fast R-CNN, Faster R-CNN or Mask R-CNN.

The learned artificial intelligence model can distinguish a region of interest 200 including a specific object on the image 10 . A specific object may be a closed area composed of corners and may be a set of a plurality of support points.

Accordingly, the electronic device 1000 may correspond to each of the frames 100 of different types as shown in FIGS. 2 and 3 . In addition, the electronic device 1000 has an advantage of being able to correspond to each of the frames 100 of the type of FIG. 2 and the type of FIG. 3 even when they are in one image 10 .

Next, look at the step of determining the size of the material (500).

The frame inspection method according to an embodiment of the present invention may further include determining the size of the material 500 to be coupled to each frame 100 based on the virtual figure 400 . The electronic device 1000 may measure the size of the virtual figure 400 and requires a correction value to obtain a size to be applied to the actual frame 100 .

9 is a diagram showing the size of a virtual figure 400 according to an embodiment of the present invention. 10 is a diagram showing a frame inspection device and a material 500 installed using the method.

According to FIG. 9(a), a virtual figure 400 formed along a plurality of feature points 300 has a rectangular shape. The electronic device 1000 may measure the horizontal length h1 and the vertical length h2 through the virtual figure 400 . However, the size of the actual material 500 must be larger than the virtual figure 400 so that it can be coupled to the frame 100. Accordingly, the electronic device 1000 may receive length information corresponding to h3, h41, and h42 as correction values. The electronic device 1000 may receive the values of h3, h41, and h42 in advance before acquiring the image 10 or after acquiring the image 10 .

Also, according to FIG. 9(b), the virtual figure 400 formed along the inner vertex of the frame 100 has a rectangular shape. The electronic device 1000 may measure the horizontal length h5 and the vertical length h6 through the virtual figure 400 . However, since the size of the actual material 500 may be different from that of the virtual figure 400, the electronic device 1000 may receive necessary length information as a correction value.

The electronic device 1000 may determine the size of the material 500 to be coupled to each frame 100 by substituting a correction value based on the virtual figure 400 . In addition, the electronic device 1000 may output the size of the virtual figure 400 and/or the size of the material 500 to be combined with each frame 100 .

That is, the frame inspection apparatus and frame inspection method according to an embodiment of the present invention introduces the concept of the virtual figure 400 to provide a uniform measurement method, and additionally provides a correction value to correspond to the actual frame 100. It is possible to determine the size of the material 500 to which the actual frame 100 is applied. There is an advantage that can be applied to various frames (100).

Referring to FIG. 10 , it can be seen that a material 500 manufactured using size information of the material 500 is coupled to each frame 100 .

Next, look at the reference object 600 that allows the absolute size of the virtual figure 400 to be measured.

Referring to FIG. 8 , in the frame inspection method according to an embodiment of the present invention, the image 10 includes a reference object 600 whose size is known, and obtaining the virtual figure 400 (S400), the size of the virtual figure 400 may be measured in consideration of the relative ratio between the virtual figure 400 and the reference object 600.

Image 10 may include reference object 600 . The reference object 600 may be located adjacent to the frame 100 and photographed by the photographing unit 1600 . Also, the reference object 600 may be displayed on the image 10 after acquiring the image 10 .

The reference object 600 must know the absolute value of its size. For example, since only the relative length of h1 on the image 10 can be known, but the absolute value cannot be known, the absolute length of h1 can be known using the reference object 600 .

On the image 10, the horizontal length (r1) of the reference object 600, the horizontal length (h1) of the virtual figure 400, and the absolute length of the reference object 600 are used to determine the size of the virtual figure 400 through a proportional formula. Absolute length can be calculated.

Meanwhile, the frame inspection apparatus and method according to the embodiment of the present invention can measure the size of the virtual figure 400 even when the reference object 600 does not exist. The artificial intelligence model may measure the size of the virtual figure 400.

The artificial intelligence model may learn the image 10 including at least one frame 100 . Specifically, the artificial intelligence model may learn data that labels the image 10. After learning is completed, the artificial intelligence model can measure the length of at least one frame 100 using the image 10 .

Hereinafter, a frame inspection apparatus according to an embodiment of the present invention will be described.

11 is a block diagram illustrating a frame inspection device.

Referring to FIG. 11 , in the frame inspection apparatus including a memory 1100 and a processor 1300, the frame inspection apparatus according to an embodiment of the present invention is stored by the memory 1100 or the processor 1300. The algorithm performed by ) acquires an image 10 including at least one frame 100, and divides the image 10 into a plurality of regions of interest 200 for analysis for each frame 100. A virtual figure 400 generated by extracting a plurality of feature points 300 from each region of interest 200 for each region of interest 200 and using a virtual line connecting the plurality of feature points 300 It may include obtaining.

The electronic device 1000 may include a memory 1100 and a processor 1300 . An algorithm performed by the frame 100 inspection method according to the present invention may be performed by the processor 1300 of the electronic device 1000 and may be stored in the memory 1100 of the electronic device 1000.

Next, the frame inspection device according to an embodiment of the present invention further includes a photographing means 1600, and the photographing means 1600 captures the at least one frame 100 to obtain the image 10. can be obtained

The image 10 may be acquired through the photographing unit 1600, the memory 1100 may store the photographed image 10, and the processor 1300 may perform a frame inspection method through the stored image 10. can do.

Next, the frame inspection apparatus according to an embodiment of the present invention may further include an output unit 1200, and the output unit 1200 may provide information including the obtained virtual figure 400. .

The output unit 1200 may provide information including the virtual figure 400 . The output unit 1200 may display the shape of the virtual figure 400 . The output unit 1200 may display the size of the virtual figure 400 . Also, the output unit 1200 may display the size of the virtual figure 400 and the size of the actual material 500 using the correction value. The output unit 1200 may provide the above size in the form of visual or auditory information according to the type.

Next, the frame inspection apparatus according to an embodiment of the present invention further includes a communication unit 1500, and the communication unit 1500 acquires the image 10 from an external device and converts the virtual figure 400 to The included information can be provided to an external device.

The image 10 may be obtained from an external device. At this time, the communication unit 1500 may receive image 10 data from an external device. Meanwhile, the communication unit 1500 may transmit data on the shape of the virtual figure 400, the size of the virtual figure 400, and/or the size of the actual material 500 to an external device.

Next, the frame inspection device according to an embodiment of the present invention further includes an input unit, wherein the input unit determines the size of the material 500 to be coupled to each frame 100 based on the virtual figure 400. To determine, a correction value corresponding to a difference in size between the virtual figure 400 and the material 500 to be combined with each frame 100 may be input.

Since the size of the material 500 to be combined with each frame 100 may be determined by the correction value, the input unit may provide the correction value to the memory 1100 and/or the processor 1300. Through this, the processor 1300 may determine the size of the material 500 .

As described above, although it has been described with reference to preferred embodiments of the present invention, those skilled in the art can variously modify the present invention within the scope not departing from the spirit and scope of the present invention described in the claims below. or by transforming

can be carried out.

10: Image
100: frame
110: support
130: reflective mark
200: area of interest
300: feature point
400: virtual figure
500: material
600: reference object
1000: electronic device
1100: memory
1200: output unit
1300: processor
1500: Ministry of Communications
1600: shooting means

Claims (14)

In the frame inspection method stored in a memory or performed by a processor,
obtaining an image including at least one frame;
dividing the image into at least one region of interest to be analyzed for each frame;
extracting a plurality of feature points from each region of interest for each region of interest; and
and obtaining a virtual figure generated by using a virtual line connecting the plurality of feature points. According to claim 1,
The step of classifying into the region of interest,
extracting reference color information corresponding to the at least one frame from the image;
extracting a plurality of edges of each frame using the reference color; and
and dividing the at least one region of interest including one closed region formed by the plurality of corners. According to claim 1,
The step of extracting the plurality of feature points,
A frame inspection method of extracting vertices of each frame in each region of interest as the plurality of feature points. According to claim 1,
The step of classifying into the region of interest,
extracting reference color information corresponding to the at least one frame from the image; and
extracting a plurality of support points connected to each frame using the reference color; and
and dividing the at least one region of interest including a plurality of support points connected only to each frame. According to claim 1,
The step of extracting the plurality of feature points,
The frame inspection method of extracting the plurality of support points in each region of interest as the plurality of feature points. According to claim 1,
The virtual line is a set of lines connecting the first feature point to the second feature point and the first feature point to the third feature point among the plurality of feature points. According to claim 1,
The step of classifying into the region of interest,
Characterized in that using an artificial intelligence model, a frame meter reading method. According to claim 1,
The frame inspection method further comprising determining a size of a material to be combined with each frame based on the virtual figure. According to claim 1,
The image includes a reference object whose absolute size is known;
The step of obtaining the virtual figure,
A frame inspection method of measuring a size of the virtual figure in consideration of a relative ratio between the virtual figure and the reference object. In the frame inspection device including a memory and a processor,
The algorithm stored by the memory or executed by the processor,
An image including at least one frame is acquired, the image is divided into a plurality of regions of interest for analysis for each frame, a plurality of feature points are extracted from each region of interest for each region of interest, and the plurality of feature points are A frame inspection device comprising obtaining a virtual figure generated by using a connected virtual line. According to claim 10,
Including more shooting means,
wherein the photographing means obtains the image by photographing the at least one frame. According to claim 10,
further comprising an output unit;
wherein the output unit provides information including the obtained virtual figure. According to claim 10,
Further comprising a communication unit,
wherein the communication unit obtains the image from an external device and provides information including the virtual figure to the external device. According to claim 10,
Including more input,
The input unit inputs a correction value corresponding to a size difference between the virtual figure and the material to be combined with each frame in order to determine the size of the material to be combined with each frame based on the virtual figure. .

Images