KR20210030230A - Data collection device for augmented reality - Google Patents

Abstract

Disclosed is a data collection device for augmenting various information on an RGB image. The disclosed data collection device for augmented reality comprises: a vision sensor generating an RGB image; a thermal imaging camera which generates a thermal image and is disposed to be spaced apart from an RGB camera; an optical system matching optical axes of the vision sensor and the thermal imaging camera; and an image processing unit augmenting the thermal image on the RGB image.

Description

Data collection device for augmented reality {DATA COLLECTION DEVICE FOR AUGMENTED REALITY}

The present invention relates to a data collection device for augmented reality, and more particularly, to a data collection device for augmenting various information on an object recognized in an RGB image.

In general augmented reality (AR), virtual content such as status information and images are superimposed on a relative position from a specific target object detected and identified from a real-life image, and the detection of the target and acquisition of status information are limited to those that can be visually distinguished. .

Recently, with the development of artificial intelligence technology, the recognition rate of image-based object recognition technology has been improved and the target type has been expanded. However, in augmented reality applications, information obtained from visually distinguishable objects is limited to augmentation. Accordingly, in particular, in the field of industrial or management augmented reality that augments and shows state information of a specific object, there is a limit to content application.

As an example, in order to augment the temperature information of the object surface obtained from the thermal imaging camera in the real-life image, that is, the RGB image, it is necessary to match the RGB image and the thermal image. When the optical axes of the RGB camera and the thermal imaging camera do not coincide, There is a problem that it is difficult to match the RGB image and the thermal image. Through a separate operation for registration, the RGB image and the thermal image can be matched, but a high amount of computation and accurate calibration are required. Although the amount of computation can be reduced by lowering the frame rate of an image, it is difficult to achieve satisfactory performance in this case.

Related prior documents include Korean Patent Registration Nos. 10-1997770 and 10-1962755.

The present invention is to provide a data collection device for collecting image data necessary for augmented reality.

In addition, the present invention is to provide a data collection device for augmented reality that can easily match images used in augmented reality.

In addition, the present invention is to provide a data collection device for augmented reality that can easily correct the missing depth information.

According to an embodiment of the present invention for achieving the above object, a vision sensor for generating an RGB image; A thermal imaging camera that generates a thermal image and is spaced apart from the RGB camera; An optical system that matches the optical axis of the vision sensor and the thermal imaging camera; And an image processing unit for augmenting the thermal image on the RGB image.

In addition, according to another embodiment of the present invention for achieving the above object, an RGB camera for generating an RGB image; A thermal imaging camera that generates a thermal image and is spaced apart from the RGB camera; A depth camera that generates a depth image and is spaced apart from the RGB camera and the thermal imaging camera; An optical system that matches optical axes of the RGB camera, the thermal imaging camera, and the depth camera; And an image processing unit for augmenting the thermal image or the depth image on the RGB image.

According to an embodiment of the present invention, by collecting an RGB image, a thermal image, and a depth image while the optical axes of the RGB camera, the thermal image camera, and the depth camera are matched, matching between the RGB image, the thermal image, and the depth image This can be done easily.

In addition, according to an embodiment of the present invention, missing depth information may be obtained by using the matched image.

In addition, according to an embodiment of the present invention, the accuracy of object recognition may be improved by using the 3D modeling result of the target object for recognition of the target object.

1 is a view for explaining a data collection device for augmented reality according to an embodiment of the present invention.
2 is a diagram illustrating an image processing unit according to an embodiment of the present invention.
3 is a view for explaining a data collection device according to another embodiment of the present invention.
4 is a diagram illustrating an image processing unit according to another embodiment of the present invention.
5 is a diagram illustrating an edge of a target object in which a part of the RGB image is covered.
6 is a diagram illustrating a target object in an RGB image and a depth image.
7 is a diagram illustrating a hole area in a depth image.

In the present invention, various modifications may be made and various embodiments may be provided, and specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to a specific embodiment, it should be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the present invention. In describing each drawing, similar reference numerals have been used for similar elements.

The present invention relates to an apparatus for collecting data for augmentation on an RGB image.

According to an embodiment of the present invention, data to be augmented in an RGB image may be 3D geometric information or visual feature information for object recognition, such as temperature information or depth information on the surface of an object included in the RGB image. , Depending on the camera or sensor used, other types of data may be collected and augmented.

As an embodiment, a thermal imaging camera may be used to obtain temperature information on a surface of an object. In order to augment such temperature information in an RGB image, it is necessary to match the RGB image with the thermal image. According to an embodiment of the present invention, in order to obtain a matched image without separate data processing for such registration, the optical axes of the RGB cameras and the thermal imaging cameras disposed at different positions are matched with each other.

Since the same object is located at different points in the RGB image and the thermal image acquired while the optical axes of the RGB camera and the thermal imaging camera are not aligned, a separate correction process is required for image matching, but the RGB camera and the thermal imaging camera Since the same object is located at the same point in different images when the optical axes of are matched, a matched image can be obtained without a separate correction process for registration.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a view for explaining a data collection device for augmented reality according to an embodiment of the present invention.

Referring to FIG. 1, a data collection device according to an embodiment of the present invention includes a vision sensor 110, a thermal imager 120, an optical system 130, and an image processing unit 140. As an embodiment, the vision sensor may be an RGB camera or an RGB-D camera that simultaneously generates an RGB image and a depth image.

The vision sensor 110 for generating an RGB image and the thermal imager 120 for generating a thermal image are disposed to be spaced apart from each other, and the optical system 130 includes the vision sensor 110 and the thermal camera 120 Match the optical axis of ).

As an embodiment, the optical system 130 proceeds to the first path 161 corresponding to the optical axis of the vision sensor 110 and transmits the beam input to the optical system 130 to the second path 162. And a beam splitter 131 that separates the beam of the wavelength band and the beam of the remaining wavelength band proceeding to the first path 161. The beam of the far-infrared wavelength band is input to the thermal imaging camera 120 along the second path 162 corresponding to the optical axis of the thermal imaging camera 120, and the beam of the remaining wavelength band is visualized along the first path 161. It is input to the sensor 110. The beam in the remaining wavelength band includes light in the visible wavelength band.

As shown in FIG. 1, light moving along a path corresponding to the same optical axis called the optical axis of the vision sensor 110 is divided into two beams by the beam splitter 131, respectively, and the vision sensor 110 and the column Since the image camera 120 is input, the beam input to the thermal image camera 120 can be regarded as a beam traveling along the optical axis of the vision sensor 110. Accordingly, it can be seen that the optical axes of the vision sensor 110 and the thermal imaging camera 120 coincide with the optical axes of the vision sensor 110.

At this time, in the first path 161, the distance L1 from the beam splitter 131 to the vision sensor 110 is the distance L2 of the second path, that is, the beam splitter in the second path 162 ( It may be the same as the distance from 131) to the thermal imaging camera 120.

Accordingly, according to an embodiment of the present invention, not only an RGB image and a thermal image are matched, but an RGB image and a thermal image including objects of the same size may be obtained without a difference in perspective.

As another embodiment, the positions of the vision sensor and the thermal imaging camera may be opposite to each other. In this case, the optical system proceeds to the first path corresponding to the optical axis of the thermal imaging camera, and the beam input to the optical system proceeds to the far-infrared wavelength band beam traveling to the first path and the second path corresponding to the optical axis of the vision sensor. It may include a beam splitter for separating the beams of the remaining wavelength band. In this case, in the first path corresponding to the optical axis of the thermal imaging camera, the length from the beam splitter to the thermal imaging camera may be the same as the length of the second path corresponding to the optical axis of the vision sensor.

The image processing unit 140 receives an RGB image and a thermal image from the vision sensor 110 and the thermal imaging camera 120 and enhances the thermal image on the RGB image. Depending on the embodiment, the entire thermal image may be augmented on the RGB image, or a part of the thermal image of the target object may be augmented on the target object included in the RGB image.

If the RGB-D camera is used as a vision sensor, the image processing unit 140 may augment the depth image to the RGB image.

2 is a diagram illustrating an image processing unit according to an embodiment of the present invention.

Referring to FIG. 2, the image processing unit 140 according to an embodiment of the present invention may include an image recognition unit 142 and an augmentation processing unit 143, and further includes an image matching unit 141 according to the embodiment. Can include.

The image matching unit 141 matches the RGB image and the thermal image. As described above, a matched RGB image and a thermal image may be obtained through an optical system. For post-processing for more accurate registration, the image matching unit 141 may be used.

The image matching unit 141 may perform registration using the feature points of the RGB image and the thermal image. When the angles of view of the RGB camera 110 and the thermal imaging camera 120 are different, after performing calibration for the angle of view, RGB images and thermal images can be matched through image warping.

The image recognition unit 142 recognizes and divides a target object from an RGB image and a thermal image. Since the RGB camera 110 and the thermal imaging camera 120 receive a beam traveling in the same optical axis, the RGB image and the thermal image include the same target object.

The image recognition unit 142 can recognize an object included in an RGB image through various object recognition algorithms using feature point detection, edge detection, etc., and object recognition and object recognition using a machine learning algorithm such as semantic segmentation. Splitting can also be performed at the same time.

As an embodiment, the image recognition unit 142 performs object recognition and segmentation in an RGB image, and then applies the object recognition and segmentation result to a thermal image matched to the RGB image, thereby recognizing and segmenting an object for a thermal image. You can do it.

The augmentation processing unit 143 augments a thermal image of the target object on the target object included in the RGB image. For example, when the first and second objects are included in the RGB image and the first object is a target object, the augmentation processing unit 143 may overlap the thermal image with the first object of the RGB image, and the second object is the RGB image. It is expressed as the RGB pixel value at.

Meanwhile, according to an embodiment of the present invention, the vision sensor 110, the thermal imaging camera 120, the optical system 130, and the image processing unit 140 may receive external light as shown in FIG. 1. A portion of the hardware frame 150 may be disposed inside the open hardware frame 150 so that the hardware frame 150 may be used in various places with various types of mounting devices attached thereto.

As an embodiment, the mounting device may be a mount, a handle, a glasses frame, etc., and the data collection device according to an embodiment of the present invention is used in combination with a harness, a backpack, etc., or worn on the user's body. I can.

Returning to FIG. 2 again, the data collection device according to an embodiment of the present invention may further include a display, a temperature sensor, a humidity sensor, an inertial sensor, and the like. An RGB image generated by the augmentation processing unit 143 with augmented thermal image of the target object may be displayed on a display. The image processing unit 140 may additionally display sensing data of a temperature sensor and a humidity sensor on the RGB image in which the thermal image is enhanced. In addition, the position of the object may be accurately calculated through the inertial sensor, and the accuracy of 3D modeling, which will be described later, may be improved.

3 is a view for explaining a data collection device according to another embodiment of the present invention.

Referring to FIG. 3, a data collection device according to an embodiment of the present invention includes an RGB camera 110, a thermal imaging camera 120, a depth camera 170, an optical system 330, and an image processing unit 340. . In FIG. 3, an embodiment in which a depth camera is additionally used is described, but a camera capable of acquiring other information other than the depth camera may be used according to the embodiment.

Compared with FIG. 1, the embodiment illustrated in FIG. 3 further includes a vision sensor 110, which is an RGB camera, and a depth camera 170, which is disposed to be spaced apart from the thermal imaging camera 120. The depth camera 170 generates a depth image by irradiating near-infrared light and receiving near-infrared light reflected from an object.

And as the depth camera 170 is added, the optical system 330 matches the optical axes of the RGB camera 110, the thermal imaging camera 120, and the depth camera 170, and for this purpose, the second beam splitter 132 additionally It may further include.

As described above, the first beam splitter 131 separates the beam input to the optical system 330 into a beam of a far-infrared wavelength band and a beam of the remaining wavelength band. In addition, the second beam splitter 132 splits the beam of the remaining wavelength band separated by the first beam splitter 131 into a beam of a visible light band and a beam of a near-infrared wavelength band.

As an embodiment, the first beam splitter 131 proceeds to the first path 161 corresponding to the optical axis of the RGB camera 110 and transmits a beam input to the optical system. It is divided into a beam of a far-infrared wavelength band traveling to the second path 162 and a beam of the remaining wavelength band traveling to the first path 161. In addition, the second beam splitter 132 transmits the beam of the remaining wavelength bands traveling to the first path 161 corresponding to the optical axis of the RGB camera 110 and the third path 163 corresponding to the optical axis of the depth camera 170. A beam of a near-infrared wavelength band traveling to) and a beam of a visible light wavelength band traveling to the first path 161 are separated.

The beam of the far-infrared wavelength band is input to the thermal imaging camera 120 along the second path 162, and the beam of the near-infrared wavelength band is input to the depth camera 170 along the third path 163. In addition, the beam of the visible light wavelength band is input to the RGB camera 110 along the first path 161.

As shown in FIG. 3, light moving along a path corresponding to the same optical axis called the optical axis of the RGB camera 110 is divided into three beams by the beam splitters 131 and 132, and each of the RGB cameras 110 , Since it is input to the thermal imaging camera 120 and the depth camera 170, both the beams input to the thermal imaging camera 120 and the depth camera 170 can be regarded as beams traveling along the optical axis of the RGB camera 110. have. Accordingly, it can be seen that the optical axes of the RGB camera 110, the thermal imaging camera 120, and the depth camera 170 are all aligned with the optical axes of the RGB camera 110.

At this time, the distance (L0 + L1) of the moving path of the beam moving from the first beam splitter 131 to the RGB camera 110 is a beam moving from the first beam splitter 131 to the thermal imaging camera 120 It may be the same as the distance L2 of the moving path of and may be the same as the distance L0 + L3 of the moving path of the beam moving from the first beam splitter 131 to the depth camera 170. In other words, in the first path 161, the distance (L0 + L1) from the first beam splitter 131 to the RGB camera 110 is the same as the distance L2 of the second path, and the first path It is equal to the sum of the distance L0 from the first beam splitter 131 to the second beam splitter 132 at 161 and the distance L3 of the third path.

Accordingly, according to an embodiment of the present invention, not only the RGB image, the thermal image, and the depth image are matched, but an RGB image, a thermal image, and a depth image including objects of the same size may be obtained without a difference in perspective.

As another embodiment, the positions of the RGB camera, the thermal imaging camera, and the depth camera may be changed to each other. In this case, the optical system proceeds in a path corresponding to the optical axis of the depth camera or the thermal imaging camera, and the beam input to the optical system may be separated so that the beam is input to the remaining cameras.

Meanwhile, as described above, in the optical system 330, depending on the arrangement positions of the RGB camera 110, the thermal imaging camera 120, and the depth camera 170, the separated beams are respectively RGB camera 110 and thermal imaging camera. Another optical element, such as a lens or a mirror, may be further included so as to be input to the 120 and the depth camera 170.

In addition, the image processing unit 340 augments a thermal image to an RGB image or a depth image to an RGB image. The image processing unit 340 may augment the entire depth image to the RGB image, or may augment a part of the depth image of the target object to the target object included in the RGB image, according to embodiments. The image processing unit 340 may additionally display sensing data of a temperature sensor and a humidity sensor on an RGB image in which a thermal image or a depth image is augmented.

FIG. 4 is a diagram for explaining an image processing unit according to another embodiment of the present invention, and FIG. 5 is a diagram illustrating an edge of a target object in which a part of an RGB image is covered. 6 is a diagram illustrating a target object in an RGB image and a depth image, and FIG. 7 is a diagram illustrating a hole area in the depth image.

Referring to FIG. 4, the image processing unit 340 according to an embodiment of the present invention may include an image recognition unit 342 and an augmentation processing unit 343, and further includes an image matching unit 341 according to the embodiment. Can include.

As described above, for post-processing for more accurate image registration, the image matching unit 341 may be used, and the image matching unit 341 matches the RGB image, the thermal image, and the depth image using feature points. Or, you can calibrate the angle of view.

The image recognition unit 342 recognizes and divides a target object from at least one of an RGB image, a depth image, and a thermal image.

As an embodiment, the image recognition unit 342 performs object recognition and segmentation in an RGB image, and then applies the object recognition and segmentation result to a thermal image or a depth image matched to the RGB image, thereby providing a thermal image or a depth image. Object recognition and segmentation can be performed.

In addition, the image recognition unit 342 may generate a 3D model for the recognized and segmented target object by using a depth value for the target object included in the depth image, that is, 3D geometric information. As an example, the image recognition unit 342 may generate a 3D model for a target object using a point cloud or a modeling technique using a voxel.

The augmentation processing unit 343 augments a thermal image or a depth image of the target object on the target object included in the RGB image. For example, when the first and second objects are included in the RGB image and the first object is the target object, the augmentation processing unit 143 may overlap the depth image or the thermal image with the first object of the RGB image, and the second object May be expressed as an RGB pixel value in an RGB image.

If a 3D model for the target object is generated, the augmentation processing unit 343 may augment a thermal image or a depth image for the target object on the 3D model.

Meanwhile, according to another embodiment of the present invention, the image recognition unit 342 uses an RGB image, a depth image, and a thermal image to determine a depth value for a hole region in which the depth value is missing from the target object region of the depth image. Can be corrected. As described above, since the depth image is generated using near-infrared light reflected from the object from the object, the depth value may not be obtained for the part where total reflection occurs, or the depth value may not be obtained because it is obscured by another object. . In FIG. 7, the left image is an RGB image, and the right image is a depth image for an RGB image, and a hole area 710 in which a depth value is missing due to total reflection and a hole area 720 in which a depth value is missing due to occlusion are included. Is shown. The image recognition unit 342 may detect such a hole area in the depth image and correct a depth value for the hole area by using the RGB image and the thermal image.

As an embodiment, the image recognition unit 342 may include an edge determination unit 3421, a hole search unit 3422, and a depth value generator 3423.

The edge determiner 3421 determines the edge of the target object using the RGB image and the thermal image. If there is a hidden area of the target object, the edge of the target object cannot not be accurately extracted from the RGB image, but since the hidden area of the target object may be displayed in the thermal image, the edge determination unit 3421 may determine the RGB image and the thermal image. By using complementarily, the edge for the target object can be determined.

As an embodiment, the edge determiner 3421 may recognize and segment a target object in an RGB image, and correct an edge of the target object by using a target object region of a thermal image corresponding to the segmented target object. The edge determination unit 3421 determines a target object area having the same shape as the edge of the target object extracted from the RGB image from the thermal image, and using the target object area of the thermal image and pixel values around the target object area, The edge of the target object can be corrected.

For example, when the lower right portion of the target object 510 having a rectangular shape as shown in FIG. 5(a) is covered by another object, the edge 530 of the target object extracted from the RGB image 520 as shown in FIG. 5(b) The shape of) may not be a square shape such as the shape of the target object 510. In this case, if the shape of the area in which the same or similar pixel values are distributed among the target object area and the pixel values around the target object area of the thermal image is a square, the edge determination unit 3421 determines the edge of the target object in FIG. 5 ( It can be corrected with the same square as a).

Alternatively, the edge determiner 3421 may recognize the target object using the thermal image and detect an edge of the recognized target object in the RGB image.

The hole search unit 3422 determines a hole area in the target object area of the depth image corresponding to the edge by using the edge of the target object determined by the edge determiner 3421. When the edge 620 for the triangular target object is determined in the RGB image 610 as shown in FIG. 6(a), the hole search unit 3422 uses the edge 620, as shown in FIG. 6(b). The target object region 640 corresponding to the edge 620 is extracted from the depth image 630. The hatched portion of the target object area 640 is an area in which the depth value exists, and the hole search unit 3422 may determine the hole area 650 in the target object area 640 where the depth value does not exist.

As an embodiment, the hole search unit 3422 may determine the hole area by comparing a threshold value determined from the extracted depth value in the target object area with the depth value in the target object area. For example, the threshold value may be an average value or a maximum value of the threshold value included in the target object area. In general, since the pixel value of the hole area where the depth value does not exist is 0 or close to 0, the hole search unit 3422 An area including a pixel value smaller than the threshold value in the target object area may be determined as the hole area.

Alternatively, the hole search unit 3422 may determine the hole area in an area located within a preset distance from the boundary line of the target object area. This focuses on the fact that a hole area generally occurs near the edge. The hole search unit 3422 may determine an area including a pixel value smaller than the threshold value as the hole area in an area located within a preset distance from the boundary line of the target object area.

The depth value generator 3423 generates a depth value for the hole area by using the depth value around the hole area determined by the hole search unit 3422.

In one embodiment, the depth value generator 3423 determines a depth value of an area adjacent to the hole area as a depth value for the hole area, or an average value of the depth values of the area adjacent to the hole area is a depth value for the hole area. Can be determined by

As described above, in the present invention, specific matters such as specific components, etc., and limited embodiments and drawings have been described, but this is provided only to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. , If a person of ordinary skill in the field to which the present invention belongs, various modifications and variations are possible from these descriptions. Therefore, the spirit of the present invention is limited to the described embodiments and should not be defined, and all things that are equivalent or equivalent to the claims as well as the claims to be described later fall within the scope of the spirit of the present invention. .

Claims (15)

A vision sensor that generates an RGB image;
A thermal imaging camera that generates a thermal image and is spaced apart from the RGB camera;
An optical system that matches the optical axis of the vision sensor and the thermal imaging camera; And
An image processing unit to augment the thermal image on the RGB image
Data collection device for augmented reality comprising a.
The method of claim 1,
The optical system
A beam of a far-infrared wavelength band proceeding to a first path corresponding to the optical axis of the vision sensor and input to the optical system to a second path corresponding to the optical axis of the thermal imaging camera and the first path Beam splitter that separates the beams of the remaining wavelength bands
Data collection device for augmented reality comprising a.
The method of claim 2,
In the first path, the distance from the beam splitter to the vision sensor is,
In the second path, the same as the distance from the beam splitter to the thermal imaging camera
Data collection device for augmented reality.
The method of claim 1,
The optical system
Proceeding to a first path corresponding to the optical axis of the thermal imaging camera and proceeding to a beam input to the optical system in a far-infrared wavelength band traveling to the first path and a second path corresponding to the optical axis of the vision sensor. Beam splitter that separates the beams of the remaining wavelength bands
Data collection device for augmented reality comprising a.
The method of claim 4,
In the first path, the length from the beam splitter to the thermal imaging camera is,
Equal to the length of the second path
Data collection device for augmented reality.
The method of claim 1,
The image processing unit
An image recognition unit for recognizing and segmenting a target object in the RGB image and the thermal image; And
An augmentation processing unit for augmenting a thermal image of the target object on the target object included in the RGB image
Data collection device for augmented reality comprising a.
The method of claim 1,
Further comprising a temperature sensor and a humidity sensor,
The image processing unit
Displaying the sensing data of the temperature sensor and the humidity sensor on the RGB image in which the thermal image is augmented
Data collection device for augmented reality.
An RGB camera that generates an RGB image;
A thermal imaging camera that generates a thermal image and is spaced apart from the RGB camera;
A depth camera that generates a depth image and is spaced apart from the RGB camera and the thermal imaging camera;
An optical system that matches optical axes of the RGB camera, the thermal imaging camera, and the depth camera; And
Image processing unit to augment the thermal image or the depth image on the RGB image
Data collection device for augmented reality comprising a.
The method of claim 8,
The image processing unit
An image recognition unit for recognizing and segmenting a target object from at least one of the RGB image, the depth image, and the thermal image; And
An augmentation processing unit for augmenting a thermal image or a depth image of the target object on the target object included in the RGB image
Data collection device for augmented reality comprising a.
The method of claim 9,
The image recognition unit
Using the RGB image, the depth image, and the thermal image, correcting a depth value for a hole area in which the depth value is missing from the target object area of the depth image
Data collection device for augmented reality.
The method of claim 10,
The image recognition unit
An edge determiner for determining an edge of the target object using the RGB image and the thermal image;
A hole search unit for determining the hole area in the target object area of the depth image by using the edge; And
A depth value generator that generates a depth value for the hole area by using a depth value around the hole area
Data collection device for augmented reality comprising a.
The method of claim 11,
The hole search unit
Using the edge, extracting the target object region from the depth image,
Comparing the threshold value determined from the extracted depth value in the target object area with the depth value in the target object area, determining the hole area
Data collection device for augmented reality.
The method of claim 11,
The hole search unit
Using the edge, extracting the target object region from the depth image,
Determining the hole area in an area located within a preset distance from the boundary line of the target object area
Data collection device for augmented reality.
The method of claim 9,
The optical system
A first beam splitter for separating the beam input into the optical system into a beam of a far infrared wavelength band and a beam of a remaining wavelength band; And
And a second beam splitter for separating the beam of the remaining wavelength band into a beam of a visible wavelength band and a beam of the near-infrared wavelength band,
The distance of the moving path of the beam moving from the first beam splitter to the RGB camera is
The same as the distance of the moving path of the beam moving from the first beam splitter to the thermal imaging camera and the moving path of the beam moving from the first beam splitter to the depth camera
Data collection device for augmented reality.
The method of claim 8,
Further comprising a temperature sensor and a humidity sensor,
The image processing unit
Displaying sensing data of the temperature sensor and the humidity sensor on the RGB image in which the thermal image or the depth image is augmented
Data collection device for augmented reality.

Images