KR20230130890A - Device and method of controlling vehicle body inspection - Google Patents

Abstract

The present invention relates to a vehicle body inspection control device and control method, and to a vehicle body inspection control device and control method that can automate the setting of regions of interest required for external inspection of a vehicle body. According to some embodiments of the present invention, a vehicle body inspection control method includes obtaining a three-dimensional modeling image and an actual image of an inspection object, respectively; Setting a plurality of regions of interest in the 3D modeling image based on preset conditions; and overlaying a plurality of regions of interest set on the 3D modeling image onto the actual image.

Description

Vehicle body inspection control device and control method {DEVICE AND METHOD OF CONTROLLING VEHICLE BODY INSPECTION}

The present invention relates to a vehicle body inspection control device and control method, and to a vehicle body inspection control device and control method that can automate the setting of regions of interest required for external inspection of a vehicle body.

Various panels are attached to the body of a vehicle to form the exterior of the vehicle body. An exterior inspection of the produced vehicle body is performed to prevent defects such as scratches and irregularities that may exist on the exterior of the vehicle body.

The exterior inspection of the car body was mainly performed based on the operator's visual and tactile sensations. In this case, defects tend to be determined based on the operator's subjective judgment standards, which reduces reliability and makes uniform quality control difficult.

Accordingly, Korean Patent Publication No. 10-1610148 (“Patent Document 1”) discloses a system that enables quantitative and visual analysis of surface defects in a car body by applying a vision inspection method using pattern lighting.

Registered Patent Publication No. 10-1610148 (Registration Date: 2016.04.01)

The present invention was devised to solve the above-mentioned problems,

The aim is to provide a car body inspection control device and method that can significantly reduce the time required for car body exterior inspection.

The purpose of the present invention is not limited to the purposes mentioned above, and other purposes not mentioned will be clearly understood by those skilled in the art (hereinafter referred to as 'ordinary skilled in the art') in the technical field to which the present invention pertains from the description below. It could be.

In order to achieve the purpose of the present invention as described above and perform the characteristic functions of the present invention described later, the features of the present invention are as follows.

According to some embodiments of the present invention, a vehicle body inspection control method includes obtaining a three-dimensional modeling image and an actual image of an inspection object, respectively; Setting a plurality of regions of interest in the 3D modeling image based on preset conditions; and overlaying a plurality of regions of interest set on the 3D modeling image onto the actual image.

According to some embodiments of the present invention, a vehicle body inspection control device includes: a modeling image providing unit configured to provide a three-dimensional modeling image of an inspection object; an image providing unit configured to provide an actual vision image of the inspection target; a controller configured to acquire the 3D modeling image from the modeling image providing unit and an actual vision image from the image providing unit, and set a plurality of regions of interest in the 3D modeling image based on preset conditions; and an output unit configured to output a 3D modeling image in which the region of interest is set.

According to the present invention, a car body inspection control device and method are provided that can significantly shorten the time required for a car body exterior inspection by automatically setting a region of interest based on a 3D modeling image without manual work.

The effects of the present invention are not limited to those described above, and other effects not mentioned will be clearly recognized by those skilled in the art from the description below.

Figure 1a shows an example of manually setting the region of interest (ROI) within the measurement area (FOV),
1B to 1E display each set region of interest (ROI),
Figure 2 shows a configuration diagram of a vehicle body inspection control device according to the present invention;
Figure 3 shows a flowchart of automatic area-of-interest setting control by the car body inspection control device according to the present invention;
Figure 4 shows a control flowchart of a region growth technique applied when setting a region of interest according to an embodiment of the present invention;
5A-5F illustrate exemplary processes in which region growth techniques are applied according to embodiments of the present invention;
Figure 6 illustrates a method of determining a common area when applying a region growth technique according to an embodiment of the present invention;
Figure 7 shows a flowchart of automatic region-of-interest setting control according to an embodiment of the present invention;
Figure 8 shows a judgment flowchart when adjacent common areas meet when automatically setting a region of interest according to an embodiment of the present invention;
Figure 9 shows judgment conditions when adjacent common areas meet when automatically setting a region of interest according to an embodiment of the present invention.

The specific structural or functional descriptions presented in the embodiments of the present invention are merely illustrative for the purpose of explaining the embodiments according to the concept of the present invention, and the embodiments according to the concept of the present invention may be implemented in various forms. In addition, it should not be construed as being limited to the embodiments described in this specification, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention.

Meanwhile, in the present invention, terms such as first and/or second may be used to describe various components, but the components are not limited to the above terms. The above terms are used only for the purpose of distinguishing one component from other components, for example, within the scope of the rights according to the concept of the present invention, the first component may be named the second component, Similarly, the second component may also be referred to as the first component.

When a component is said to be “connected” or “connected” to another component, it should be understood that it may be directly connected or connected to the other component, but that other components may exist in between. something to do. On the other hand, when it is mentioned that a component is “directly connected” or “in direct contact” with another component, it should be understood that there are no other components in between. Other expressions to describe the relationship between components, such as “between” and “immediately between” or “adjacent to” and “directly adjacent to”, should be interpreted similarly.

Like reference numerals refer to like elements throughout the specification. Meanwhile, the terms used in this specification are for describing embodiments and are not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated in the context. As used in the specification, “comprises” and/or “comprising” means that a referenced component, step, operation and/or element is present in one or more other components, steps, operations and/or elements. or does not rule out addition.

When determining the presence or absence of defective parts of a car body through a vision inspection such as in Patent Document 1, a recipe setting operation is required after obtaining a measurement image (Field of View, FOV) using a vision camera. Recipe settings refer to setting the region of interest (ROI) and defect detection standards for the measurement image (FOV). Here, the region of interest (ROI) corresponds to the analysis area within the measurement area (FOV), and is managed by dividing the measurement area (FOV) into multiple parts.

However, the number of measurement images (FOV) of the exterior surface area of the vehicle body that is subject to inspection reaches hundreds of images per vehicle body, so it takes a considerable amount of time to manually set the region of interest (ROI).

Accordingly, the present invention does not require manually setting a region of interest (ROI) for the measurement image (FOV) of a vision camera, but automatically creates a region of interest (ROI) for the 3D modeling file of the vehicle body, thereby creating a measurement image (FOV). The aim is to provide a car body inspection control device and control method that can shorten the recipe work time.

Hereinafter, the present invention will be described in detail with reference to the attached drawings.

Since there is a shape in the car body panel, when pattern lighting is irradiated to the panel, the shape of the pattern (width, direction, etc.) changes depending on the surface shape. For example, the pattern shape appears like a contour line. Different pattern shapes mean that the slope and shape of the surface are different, and when inspecting the surface quality of the car body (using oil stones), the direction of scraping the oil stones varies depending on the surface shape of the panel.

In a conventional method such as Patent Document 1, moiré patterns due to phase difference are visually confirmed, surfaces with the same slope are manually distinguished, and each work area is drawn as a polygon. For example, as shown in Figure 1a, four regions of interest (ROI, purple) within one measurement area (FOV) are manually distinguished (indicated by P, Q, R, and S, respectively, in Figures 1b to 1e), etc. Therefore, the work takes a very long time.

The present invention applies the region growing technique to the 3D modeling file of the car body to create the same or similar slope without the operator having to directly check the pattern shape of the measurement image (FOV) to distinguish the region of interest (ROI). The area may be automatically set as a region of interest (ROI). By using a region of interest (ROI) that is automatically set in advance during vision inspection of car body panels in the actual field, the time required to manually set the region of interest (ROI) in the field can be greatly reduced. The region of interest (ROI) that is automatically set in advance is finally inspected on site, and optimization work may be performed.

As shown in FIG. 2, the vehicle body inspection control device according to the present invention includes an input unit 100, a controller 300, and an output unit 500.

The input unit 100 provides various input information to the controller 300. A user's input may be input through the input unit 100, and various information for vehicle body inspection may be input. According to an implementation example of the present invention, the input unit 100 includes a modeling image providing unit 120 and an image providing unit 140.

The modeling image provider 120 is configured to provide the controller 300 with a 3D modeling file of the vehicle body that is an inspection target. In one implementation, a 3D modeling file of a car body may be created based on a known computer aided design (CAD) program. A variety of formats can be used as the format of the 3D modeling file, and the controller 300 is configured to read the input 3D modeling file.

The image provider 140 provides an actual image of the vehicle body. The image provider 140 is configured to collect actual images or measured images (FOV) captured from various angles of the vehicle body being inspected by a vision camera and transmit them to the controller 300.

The controller 300 includes a storage unit 320 and a processing unit 340. The storage unit 320 includes instructions executable by the processing unit 340. In some implementations, instructions stored in memory 320 may include a series of steps for controlling vehicle body inspection. In some implementations, instructions in memory 320 may include a series of steps to perform a vehicle body inspection based on conditions entered by a user. The memory unit 320 may include non-volatile memory.

The processing unit 340 is configured to execute commands stored in the memory unit 320 based on input from the input unit 100 and perform various calculations necessary for vehicle body inspection.

The output unit 500 includes a display device. Input through the input unit 100 and calculation results calculated through the controller 300 may be displayed on the output unit 500. The output unit 500, together with the input unit 100, serves as a user interface.

With reference to FIG. 3, a control method of a vehicle body inspection control device according to some embodiments of the present invention will be introduced. In step S10, control for automatically setting a region of interest (ROI) begins.

The controller 300 obtains a 3D modeling file of the vehicle body to be inspected from the modeling image provider 120 (S12). The acquired 3D modeling file may be stored temporarily or permanently in the memory unit 320. Additionally, the controller 300 can display the obtained 3D modeling file on the output unit 500.

The controller 300 causes the processing unit 340 to execute a command stored in the memory unit 320 to create a region of interest (ROI) in the acquired 3D modeling file based on preset conditions (S14).

In step S16, the controller 300 receives the actual vision image of the captured vehicle body from the image provider 140. In some implementations, acquiring the 3D modeling image in step S12 and acquiring the image of the vehicle body in step S16 may be performed simultaneously or separately. Also, the order of their acquisition can be changed.

The controller 300 is configured to overlay the region of interest (ROI) generated in the acquired 3D modeling file of the vehicle body onto the image of the vehicle body (S18). The present invention automatically sets a region of interest (ROI) in the 3D modeling image of the car body and overlays it on the actual image, greatly reducing the enormous amount of time previously required to set the region of interest (ROI). there is.

The controller 300 may perform a final inspection of the overlaid region of interest (ROI) and perform optimization as necessary (S20). Additionally, a defect detection standard may be set for each generated region of interest (ROI) (S22). Through this process, control for automatically setting the region of interest (ROI) is terminated (S24).

According to the present invention, a preset condition for generating a region of interest (ROI) may be a region growth technique. Referring to Figure 4, in some implementations, the region growing technique may be performed through steps S41 to S45.

As shown in Figure 5a, a predetermined measurement area (FOV, indicated by a box) is opened in the 3D modeling file of the vehicle body. As shown in FIG. 5B, in step S41, the controller 300 divides the portion of the 3D modeling image enlarged in FIG. 5A into a plurality of micro regions (I). And, as shown in FIG. 5C, the controller 300 sets specific micro-regions (I) among the plurality of divided micro-regions (I) as seed points (S) (S43). The seed point (S) may be placed at regular intervals in the micro region (I). The controller 300 uses a region growing technique to group common regions (eg, A1, A2, A3, A4) around each seed point (S) (see S45 and FIG. 5D).

According to an embodiment of the present invention, the surface normal of each micro-region (I) is compared with the surface normal of the adjacent micro-region (I) to determine whether it belongs to the same region or a common region. Specifically, referring to FIG. 6, the controller 300 obtains the surface normal (n) for each micro-region (I). Then, the controller 300 compares the slope of the surface normal line (n) and merges areas with the same or similar slope of the surface normal line (n) and sets them as a common area (A). The range of the same or similar slope can be set in advance. For example, small regions (I) disposed between slope inflection lines may be determined to be one common region (A).

According to some embodiments of the present invention, the region of interest (ROI) may be set automatically as shown in FIG. 7. As described above, the controller 300 divides the 3D modeling image into a plurality of small regions (I) in steps S41 and S43, and sets specific small regions among the plurality of small regions as seed points (S). A plurality of seed points (S) are set, and one of them is referred to as the first seed point (S1).

In step S51, the controller 300 obtains the first seed point S1 and the surface normal line n of the micro region I surrounding the first seed point S1. The controller 300 determines whether the surface normal (n) of the micro region (I) surrounding the first seed point (S1) has a slope within a preset range (S53). If the surface normal (n) of a specific surrounding micro-area (I) has a slope within a preset range, the micro-area (I) is divided into the first common area (A1), which is the area of the first seed point (S1). Incorporate (S55, see again Figure 5d). In the opposite case, that is, if the surface normal (n) of the surrounding specific micro-region (I) does not have a slope within a preset range, the controller 300 determines that the micro-region (I) is the first seed point (S1). 1 It is not determined to belong to the common area (A1) and is not included in the first common area (A1) (S57). That is, the first common area A1 can be expanded by including micro-areas I having a slope within a common range around the first seed point S1.

Referring to Figures 8 and 9, the determination method when the first common area (A1) of the first seed point (S1) and the second common area (A2) of the second seed point (S2), which is another seed point, meet, explained.

The controller 300 determines whether the first common area A1 of the first seed point S1 and the second common area A2 of the second seed point S2 meet (S61). When the first common area (A1) and the second common area (A2) meet, the controller 300 connects the first common area (A1) and the second common area (A1) of the first seed point (S1) with respect to the cross-sectional direction of the inspection target or vehicle body. 2 Obtain the change in surface slope of each common area of the seed point (S2) (S63). That is, the controller 300 can calculate the amount of change in surface inclination by differentiating the cross-sectional profile of the vehicle body with respect to the cross-sectional direction of the vehicle body. Specifically, the controller 300 acquires the amount of change in the surface slope of the micro-region belonging to the first common area A1 and the micro-region belonging to the second common area A2.

Then, the controller 300 compares the obtained change in surface slope with a preset tolerance range (R) (S65). If the change in surface slope is within the allowable range (R), the first common area (A1) and the second common area (A2) are merged and expanded into one area (S67, see FIG. 5e). That is, in the case of a small area where the change in surface slope is within the allowable range (R), the first common area (A1) and the second common area (A2) are considered to be the same area in common. On the other hand, when the change in surface slope is not within the allowable range (R), that is, in the case of a micro region where the change in surface slope is not within the allowable range (R), the first common area (A1) and the second common area (A2) Stop merging and manage both parts separately (S69). Additionally, as shown in FIG. 5F, a region of interest (ROI) can be set excluding edge portions and portions where the surface normal changes rapidly.

In this way, according to the present invention, the recipe work time for vision measurement images can be greatly shortened by automatically generating a region of interest (ROI) based on the 3D modeling file of the vehicle body that is the inspection target.

The present invention described above is not limited to the above-described embodiments and the accompanying drawings, and various substitutions, modifications, and changes are possible without departing from the technical spirit of the present invention as is known in the technical field to which the present invention pertains. It will be clear to those who have the knowledge of.

100: input unit 120: modeling image providing unit
140: video provider 300: controller
320: memory unit 340: processing unit
500: Output FOV: Measurement area
ROI: Region of interest I: Micro region
S: seed point S1: first seed point
A, A1, A2, A3, A4: common area n: surface normal
R: tolerance range

Claims (13)

Obtaining a 3D modeling image and an actual image of the inspection target, respectively;
Setting a plurality of regions of interest in the 3D modeling image based on preset conditions; and
Overlaying a plurality of regions of interest set on the 3D modeling image onto the actual image;
A control method for vehicle body inspection comprising: The method according to claim 1, wherein setting the plurality of regions of interest comprises:
Dividing the 3D modeling image into a plurality of micro regions;
setting a plurality of seed points including a first seed point in some of the plurality of micro-regions;
Comparing the surface slope of the first seed point and the surface slope of a micro region around the first seed point; and
If the surface slope of a small area around the first seed point is within a certain range with respect to the surface slope of the first seed point, grouping the small area with the first seed point;
A control method comprising: The method of claim 2, wherein the comparing step includes:
calculating a surface normal of the first seed point and a small region around the first seed point;
When the slope of the surface normal of the first micro region among the micro regions is within a preset range with respect to the slope of the surface normal of the first seed point, a first common center with the first micro region centered on the first seed point Incorporating into an area; and
If the slope of the surface normal of the first micro-region is not within the preset range, excluding the first micro-region from the first common area;
A control method comprising: The method of claim 3, wherein the seed point further includes a second seed point neighboring the first seed point among the seed points,
Comparing the surface slope of the second seed point with the surface slope of a micro region around the second seed point; and
If the surface slope of a micro-region around the second seed point is within a certain range with respect to the surface slope of the second seed point, grouping the micro-region with the second seed point;
A control method further comprising: The method of claim 4, wherein comparing the surface slope of the second seed point with the surface slope of a micro region around the second seed point comprises:
calculating a surface normal of the second seed point and a micro region around the second seed point;
If the slope of the surface normal of the second micro-region among the micro-regions around the second seed point is within a preset range with respect to the surface normal of the second seed point, the second micro-region is centered around the second seed point. incorporating into a second common area; and
If the slope of the surface normal of the second micro-region is not within the preset range, excluding the second micro-region from the second common area;
A control method comprising: In claim 5,
determining whether the first common area and the second common area meet;
calculating a change in surface slope of each of the first common area and the second common area with respect to the cross-sectional direction of the inspection object; and
Merging or separating a first common area and a second common area based on the change in the surface slope;
A control method further comprising: The method of claim 6, wherein the merging or separating step includes:
If the change in surface slope of a specific micro-area in the first or second common area is within a preset allowable range, the micro-area is considered to belong to an area common to the first common area and the second common area. A control method that is configured to make a decision. The method of claim 6, wherein the merging or separating step includes:
If the amount of change in surface slope of a specific micro-area within the first or second common area is outside the preset allowable range, the micro-area does not belong to an area common to the first common area and the second common area. A control method that is configured to determine that it is not. The control method according to claim 1, wherein the inspection object is the exterior of the vehicle body. The control method according to claim 2, wherein the seed points are arranged at a predetermined distance apart from the plurality of micro-regions. a modeling image providing unit configured to provide a three-dimensional modeling image of the inspection target;
an image providing unit configured to provide an actual vision image of the inspection target;
a controller configured to acquire the 3D modeling image from the modeling image providing unit and an actual vision image from the image providing unit, and set a plurality of regions of interest in the 3D modeling image based on preset conditions; and
an output unit configured to output a 3D modeling image with the region of interest set;
A car body inspection control device comprising: The method of claim 11, wherein the controller:
A car body inspection control device configured to overlay the set region of interest on the actual vision image. The vehicle body inspection control device according to claim 11, wherein each region of interest among the plurality of regions of interest is composed of portions having a surface slope within a preset certain range.

Images