WO2015008873A1 - Apparatus and method for recognizing 3d image using heterogeneous cameras - Google Patents

Apparatus and method for recognizing 3d image using heterogeneous cameras Download PDF

Info

Publication number
WO2015008873A1
WO2015008873A1 PCT/KR2013/006301 KR2013006301W WO2015008873A1 WO 2015008873 A1 WO2015008873 A1 WO 2015008873A1 KR 2013006301 W KR2013006301 W KR 2013006301W WO 2015008873 A1 WO2015008873 A1 WO 2015008873A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
camera
infrared
rgb
light source
Prior art date
Application number
PCT/KR2013/006301
Other languages
French (fr)
Korean (ko)
Inventor
김연수
Original Assignee
엘지전자 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 엘지전자 주식회사 filed Critical 엘지전자 주식회사
Priority to PCT/KR2013/006301 priority Critical patent/WO2015008873A1/en
Publication of WO2015008873A1 publication Critical patent/WO2015008873A1/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/50Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00Image enhancement or restoration
    • G06T5/001Image restoration
    • G06T5/002Denoising; Smoothing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/30Transforming light or analogous information into electric information
    • H04N5/33Transforming infra-red radiation
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10024Color image
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00Indexing scheme for image analysis or image enhancement
    • G06T2207/10Image acquisition modality
    • G06T2207/10048Infrared image

Abstract

Provided is an apparatus for recognizing a 3D image using heterogeneous cameras comprising: a light source for irradiating a laser line onto an object; an infrared camera for obtaining an infrared-photographed image of the object onto which the laser line has been projected by the irradiation of the light source; an RGB camera for obtaining an RGB-photographed image of the object onto which the laser line has been projected by the irradiation of the light source; and a control unit for obtaining a laser line from which noise is removed by an external light by comparing the infrared-photographed image obtained by the infrared camera and the RGB-photographed image obtained by the RGB camera, and recognizing the 3D image.

Description

Three-dimensional image recognition apparatus and method using a heterogeneous camera

The present invention, specifically, the external light such as sun light with respect to the image signal obtaining the object of the laser light source is irradiated to recording using an infrared camera and an RGB camera of the three-dimensional image recognition apparatus and method using a heterogeneous camera by performing the signal processing for removing the noise signal according to the present invention relates to three-dimensional image recognition apparatus and method using a heterogeneous camera to accurately detect the laser line.

Typically the robot has been developed as a way to automate the process in a factory or industry. The robot has been recently put into practical use as well as industrial robots used in industry, etc. home or office as housework or office assistant. This is a typical example to be mentioned that the cleaning robot, robot instructions, security robots and the like.

A mobile robot such as cleaning robots, in order from the district to the path and the robot moves the robot to perform a given functional activities must first robot to recognize objects and obstacles. Recognition of the target object, such as object or obstacle can be applied to various methods such as using a visual sensor or an ultrasonic sensor, or touch sensor.

A method using a 3D sensor in a method of recognizing an object or obstacle.

1 is a view for explaining the principle of a three-dimensional image recognition method using a conventional camera.

1, a method of using a 3D sensor, the laser line by using an active light source (active vision) (10) using an infrared camera (20) reflecting the light to the target object 30 in a predetermined shape such as a laser It has been detected.

Conventionally, in order to detect the laser line projected onto the target object in the 3D sensor in the static environment by using the difference image of the image signal and the background image was detected by increasing the signal-to-noise ratio.

However, in the dynamic environment, a method using the image difference between the image signal and the background image was difficult to separate the signal and the noise by an effect of enlarging the noise generating zone conditions.

In addition, an infrared captured image 50 obtained by the infrared camera 20 includes the noise component 52 due to external light 40 with the laser line 51. Therefore, there was a precise detection of the constraints on the laser line 31 is projected onto the target object 30 due to the noise component 52 due to external light 40, when detecting the laser line.

Object of the present invention, two kinds by performing signal processing for removing the noise signal due to external light such as sun light with respect to a video signal obtained by photographing an object, a laser light source is irradiated to accurately detect the laser line to provide a three-dimensional image recognition apparatus and method using a camera.

Is not limited to this aspect are the technical problems referred to above of the invention, still another aspect are not mentioned will be clearly understood by those skilled in the art from the following description.

According to an aspect of the invention, a light source for irradiating a laser line on the object; An infrared camera for obtaining an infrared image taken for the laser line is projected onto a target object by the irradiation of the light source; RGB camera for acquiring a taken image RGB on the projected object laser line by irradiation of the light source; And a control unit for obtaining a noise due to external light is removed the laser line by comparing the RGB-up image obtained by the infrared-up image and the RGB camera acquired by the infrared camera, and performs three-dimensional image recognition the three-dimensional image recognition apparatus using a camera which is provided with two kinds.

Wherein the controller detected the laser line that contains the noise component by removing the background image from the infrared-up image, and removes the background image from the RGB-up image, and the background image is removed infrared-up image and the background image is removed RGB it is possible to detect a common noise component included in the photographed image, and obtaining the said the detected noise component is removed from the laser line is detected in the infrared image taken a noise component caused by the outside light is removed the laser line.

The light source may be irradiated with a near infrared line laser (line laser) beam.

The light source may be irradiated with a near infrared line laser beam having a wavelength more than 890nm.

The light source is centered with the infrared camera and the RGB camera can be arranged to the left and right. At this time, the light source and irradiated with a laser line with a cross-shaped pattern, the cross-shaped pattern has a vertical direction may have a major axis than to the horizontal direction.

Wherein the infrared camera and the RGB camera with respect to the light source may be disposed at the upper and lower sides. At this time, the light source and irradiated with a laser line with a cross-shaped pattern, the cross-shaped pattern has a horizontal direction may have a major axis compared to the vertical direction.

According to another aspect of the invention, the step of irradiating the object with a laser line from the light source; Obtaining an infrared image taken for the laser line is projected onto a target object by the irradiation of the light source; Obtaining an RGB image photographed on the projected object laser line by irradiation of the light source; And by comparing the RGB-up image and the infrared-up image obtained by the laser line noise it is removed by the external light, and three-dimensional three-dimensional image recognition method using the two kinds of cameras including a step of performing image recognition is provided .

Performing the three-dimensional image is recognized, the method comprising detecting the laser line that contains the noise component by removing the background image from the infrared image taken; Removing the background image from the RGB-up image, and detecting the common noise components contained in the background image is removed infrared captured image and the background image is removed, an RGB image photographed; And this way 3-D image recognition using a heterologous camera comprising the step of obtaining the above the detected noise component is removed from the laser line is detected in the infrared-up image is a noise component due to the outside light is removed the laser line is provided .

The light source may be irradiated with a near infrared line laser (line laser) beam.

The light source may be irradiated with a near infrared line laser beam having a wavelength more than 890nm.

The infrared camera image and the RGB camera image can be obtained that the infrared camera and the RGB camera with respect to the light source are disposed on the left and right sides. At this time, the light source and irradiated with a laser line with a cross-shaped pattern, the cross-shaped pattern has a vertical direction may have a major axis than to the horizontal direction.

Detecting the noise component, the background image is removed, the infrared-up image and the background image is said to detect the noise component included in removing the RGB common to the captured image infrared camera image and the left and right with respect to the RGB camera image it can perform a scan.

The infrared camera image and the RGB camera image can be obtained that the infrared camera and the RGB camera with respect to the light source is disposed at the upper and lower sides. At this time, the light source and irradiated with a laser line with a cross-shaped pattern, the cross-shaped pattern has a horizontal direction may have a major axis compared to the vertical direction.

Detecting the noise component, the background image is removed, the infrared-up image and the background image is said to detect the noise component included in removing the RGB common to the captured image infrared camera image and the up and down with respect to the RGB camera image it can perform a scan.

According to the present invention, the accuracy of the laser line is detected by photographing the object, a laser light source is irradiated using an infrared camera and an RGB camera and by performing signal processing for the video signal to remove the noise component due to external light such as sunlight It can be improved, thereby improving the accuracy of the three-dimensional image recognition.

1 is a view for explaining the principle of a three-dimensional image recognition method using a conventional camera.

2 is a block diagram of a configuration of a three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

3 is a view for explaining the principle of a three-dimensional image recognition method using the two kinds of cameras according to one embodiment of the present invention.

4 and 5 are views for explaining a pattern of the laser line that can be used accordingly, a light source, an infrared camera, the arrangement of RGB camera and the three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

6 is a view for explaining the characteristics of an image obtained by the infrared camera and an RGB camera in a three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

7 is a view for explaining detecting a laser line over the line matching component in the three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

8 is a flow chart for explaining a three-dimensional image recognition method using the two kinds of cameras according to one embodiment of the present invention.

Specific details of the embodiments are included in the detailed description and drawings.

Methods of accomplishing the advantages and features of the present invention and reference to the embodiments that are described later in detail in conjunction with the accompanying drawings will be apparent. However, the invention is not limited to the embodiments set forth herein may be embodied in many different forms, but the present embodiments, and the disclosure of the present invention to complete, and ordinary skill in the art is furnished the chair in order to fully convey the concept of the invention to have, the present invention will only be defined by the appended claims.

Hereinafter, with reference to the drawings for block diagrams or processing flow chart for explaining the three-dimensional image recognition apparatus and method using a heterogeneous camera with embodiments of the present invention will be described for the present invention. At this time, and combinations of each block and flow chart diagrams of a flow chart diagram will be understood that can be implemented by computer program instructions. These computer program instructions general purpose computer, such that the instructions, which special applications it may be mounted on the computer or other programmable data processing processor of the equipment, done through the computer or other programmable data processing apparatus of the processors described in the flowchart block (s) , it creates means for implementing the functions. These computer program instructions may also be stored in a computer or other programmable machine can be used in a data processing apparatus that can direct a computer-readable memory to function in a particular manner, that a computer is available or computer readable memory the instructions stored in their it is also possible to produce an article of manufacture including instruction means that implement the function specified in the flowchart block or blocks. Computer program instructions It is also possible that tapje on the computer or other programmable data processing apparatus, computer, or other programmable series of operational steps on the data processing apparatus are performed, perform the data processing apparatus is a computer or other programmable to generate a process executed by a computer instructions are also possible to provide steps for implementing the functions specified in the flowchart block (s).

In addition, each block of the flowchart illustrations may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function (s). In addition, in some alternative implementations, it should also be noted that the functions noted in the blocks may occur out of the order. For example, it is also possible that is also possible that two blocks shown in succession may in fact performs substantially the same time or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.

2 is a block diagram of a configuration of a three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

2, the three-dimensional image recognition apparatus using the two kinds of the camera in accordance with one embodiment of the present invention may be configured to include a light source 110, an infrared camera (120), RGB camera 130, a controller 140 can.

The light source 110 irradiates the laser line on the object 200 to a three-dimensional image recognition. Source 110 may, for example, the active light source (active vision), such as a laser (laser) may be used. The light source 110, for example, can be used a near infrared line laser (line laser) beam. The near infrared ray line laser beam it is possible to obtain the video information in a complete absence of light.

Infrared camera 120 may obtain an infrared captured image containing the image of the laser line projected onto the target object by irradiation with a light source 110.

RGB camera 130 may obtain an RGB image of the laser line is projected onto a target object by irradiation with a light source 110. In this case, RGB camera 120 may obtain the image information is located close to the light source 110 while maintaining the desired distance from the light source 110. For example, RGB camera 130 may be positioned or located on the right or on the left-position, or the position and direction on the direction of the light source 110.

The control unit 140 compares the RGB-up image obtained by the infrared-up image and the RGB camera 130 is generated by the infrared camera 130 is obtained a laser line noise is removed by the external light, and three-dimensional image It can perform recognition.

3 is a view for explaining the principle of a three-dimensional image recognition method using the two kinds of cameras according to one embodiment of the present invention.

3, the infrared laser line 201 from the light source 110 is irradiated to a target object (200).

In the infrared camera image taken by the infrared camera 120, 121 may include a noise component 123 is caused by the external light 300, such as sun light with the laser line component (122).

RGB camera 130, the RGB camera image 130 taken by the contains the noise component 133, according to the external light 300, such as sun light, laser line component is not included.

The control unit 140 performs the matching with respect to the RGB line-up image 131 generated by an infrared captured image 121 and the RGB camera 130 is generated by the infrared camera 120. Here, the line matching means for detecting the component included in the same infrared captured image 121 and the RGB-up image 131. The

Accordingly, the controller 140 recall detecting a noise component caused by the external light 300 such as an infrared captured image 121 and the RGB-up image 131 is common to solar light in. The control unit 140 removes the noise component in the infrared-up image and outputs the remaining components to the final laser line laser line.

4 and 5 are views for explaining a pattern of the laser line that can be used accordingly, a light source, an infrared camera, the arrangement of RGB camera and the three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

4, the light source 110 consisting of laser diodes are disposed in the center, the infrared camera 120 is disposed on the right side of the light source 110, an RGB camera 130 to the left of the light source (110) It is arranged. In another embodiment of the present invention there is infrared camera 120 around the light source 110 is disposed on the left side, it can be a RGB camera 130 is disposed on the right side.

In this embodiment the infrared camera 120 and the RGB camera 130 is placed in the center of the light source 110 in consideration of the angle of view is that plenty is greater than the angle of view of the light source 110 using a laser diode.

Laser line pattern can be used in various forms. For example, a laser line pattern of the four-way type may be used. At this time, as shown in (a), the light source 110, an infrared camera (120), RGB camera 130 is in the case to be arranged to the left and right, top and bottom, rather than the left and right as shown in (b) constituting the major axis it is better to use a laser line pattern of the cross-shaped pattern. This is to facilitate this, yigyeokdo (disparity) calculated when using the long axis of the four-way line to the 3D restoration.

A In a modified embodiment, if also arranged in the infrared camera (120), RGB camera 130 around the light source 110, as shown in 5 (a) up and down has, as shown in (b) described as it is better to use a laser line pattern of the cross-shaped pattern constituting the left and right, rather than up and down the long axis.

On the other hand, in the modified embodiment it is possible in several different ways, the placement of the light source 110, an infrared camera (120), RGB camera 130. For example, the center light source (110), RGB camera 130 is arranged to the left or right or the infrared camera 120, it may be arranged in the vertical direction. There are also other example, arranged in a RGB camera 130 is centered by the light source 110 and an infrared camera 120, the left and right, or may be arranged in the vertical direction.

6 is a view for explaining the characteristics of an image obtained by the infrared camera and an RGB camera in a three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

6, the light source 110 can be irradiated with a laser line pattern 201 having a wavelength band of 890nm or more to the target object (200).

Infrared camera 120 and the RGB camera 130 photographing the laser lines 201, the irradiation target object (200). Here, it is possible to take the wavelength band of 890nm or more infrared cameras 120. On the other hand, RGB camera 130 does not absorb one infrared wavelength band. Meanwhile,

An infrared captured image 121 obtained by the infrared camera 120 includes a component 123 of the cross pattern 122 and the horizontal line. On the other hand, in the photographed image 131 obtained by the RGB camera 130 is detected, only the component (133) of horizontal lines. That is, in the infrared photographing image 121 of the infrared camera 120, a laser line pattern 122 is detected from the right side region. On the other hand, not the RGB camera 130 is a laser line pattern detection.

On the other hand, the horizontal line in the infrared captured image 121 and the RGB-up image 131 (123, 133) are commonly detected. For example, sunlight is a horizontal line (123, 133) that are commonly detected in the image of the infrared camera 120 and the RGB camera 130, since it contains the wavelength band of the entire region of the external light such as sunlight (300 ) means that the noise component due to the.

The control unit 130 detects the pattern of the laser line 122 in the infrared camera image 121 in accordance with known that the pattern of the laser lines projected from the light source 110. For example, the control unit 130 can be detected through the image processing of the four-way line irradiated from the light source 110.

7 is a view for explaining detecting a laser line over the line matching component in the three-dimensional image recognition apparatus using the two kinds of cameras according to one embodiment of the present invention.

7, the small cross-line pattern in infrared photographic image 121 and the RGB-up image 131 (123, 133) is detected. Here, a state in which the light source 110, an infrared camera (120), RGB camera 130 is disposed to the left and right. Accordingly, the controller 140 is a left-right scan for infrared captured image 121 and the RGB-up image 131 can be performed to find corresponding points between two images (L -> L -> right, right) .

On the other hand, in case where the light source 110, an infrared camera (120), RGB camera 130 is disposed vertically as shown in Figure 5, the vertical scanning by performing (a -> a> ha ha) You can find the corresponding points.

The control unit 130 may match the corresponding points between the infrared captured image 121 and the RGB-up image 131 through the left and right scan. That is the laser line pattern is irradiated from the light source 110 is not detected in the photographed RGB image 121 obtained in the RGB camera 120 is different depending on the wavelength laser line being in the infrared region. Therefore, the RGB camera 120 and the infrared common pattern (123, 133) which is detected by the camera 130 at 7 is the external light 300, such as sun light as the laser line pattern illuminated by the light source 110, noise component by being removed in accordance with the pattern.

8 is a flow chart for explaining a three-dimensional image recognition method using the two kinds of cameras according to one embodiment of the present invention.

8, and a laser beam line 201 to an object 200 by the light source (110) (S1). Here, the light source 110 consisting of laser diodes are disposed in the center, the infrared camera 120 is disposed on the right side of the light source 110, can be a RGB camera 130 is disposed on the left side of the light source (110) is. At this time, the light source 110 can be irradiated with a laser line 201 having a wavelength band of 890nm or more to the target object (200). Laser line 201 has a vertical direction may be the major axis of cross-shaped pattern used in comparison to the horizontal direction.

And obtains an infrared captured image 121 on the irradiated object, the laser line by an infrared camera (120) (S2).

Here, the infrared camera 120 may take a wavelength band of 890nm or more. An infrared captured image 121 obtained by the infrared camera 120 includes a component 123 of the cross pattern 122 and the horizontal line.

And by the control unit 140 to remove the background image from the infrared image taken 121 detects the laser line contains a noise component (S3).

Obtains the RGB image photographed on the irradiated object, the laser line by the RGB camera (130) (S4).

RGB camera 130 does not absorb one infrared wavelength band. The photographed image 131 obtained by the RGB camera 130, is detected, only the noise component by the external light. That is, when the detected infrared captured image 121 laser line pattern is not detected in the RGB camera 130.

To remove the background image from the RGB image recording (S5) by the controller 140.

It detects a common noise component included in the infrared image and the background image is taken to remove the RGB image photographed background image is removed by the control unit (140) (S6).

PV is the signal component that is commonly detected in the image of the infrared camera 120 and the RGB camera 130, since it contains the wavelength band of the entire region means a noise component caused by the external light 300, such as sunlight.

It removes the noise component from the laser line is detected in the infrared captured image 121 by the control unit (140) (S7). The control unit 130 may detect the pattern of the laser line 122 in the infrared camera image 121 in accordance with known that the pattern of the laser lines projected from the light source 110.

To obtain a laser line noise components removed by the control unit (140) (S8).

Using the noise component is removed obtaining the laser line, and performs three-dimensional image recognized by the control unit (140) (S9).

One of ordinary skill in the art will appreciate that the present invention without changing departing from the scope and spirit be embodied in other specific forms. Thus the embodiments described above are only to be understood as illustrative and non-restrictive in every respect. The scope of the present invention to fall within the scope of the is represented by the claims below rather than the foregoing description, and all such modifications as derived from the meaning and range and equivalents concept of the claims of this invention It should be interpreted.

Claims (18)

  1. A light source for irradiating a laser line on the object;
    An infrared camera for obtaining an infrared image taken for the laser line is projected onto a target object by the irradiation of the light source;
    RGB camera for acquiring a taken image RGB on the projected object laser line by irradiation of the light source; And
    By comparing the RGB-up image obtained by the infrared-up image and the RGB camera acquired by the infrared camera obtain a laser line noise is removed by the external light, and to a control unit for performing 3-D Image Recognition three-dimensional image recognition apparatus using the two kinds of the camera.
  2. According to claim 1,
    Wherein the controller detected the laser line that contains the noise component by removing the background image from the infrared-up image, and removes the background image from the RGB-up image, and the background image is removed infrared-up image and the background image is removed RGB detecting a common noise component included in the photographed image, wherein the detected noise component is removed from the laser line is detected in the infrared-up image using a heterologous camera for obtaining a laser line noise components caused by the outside light is removed three-dimensional image recognition apparatus.
  3. The method of claim 1 wherein said light source is a three-dimensional image recognition apparatus using the two kinds of cameras for irradiating a near infrared line laser (line laser) beam.
  4. The method of claim 3 wherein said light source is a three-dimensional image recognition apparatus using the two kinds of cameras for irradiating a near infrared line laser beam having a wavelength more than 890nm.
  5. According to claim 1,
    Three-dimensional image recognition apparatus using the two kinds of cameras are disposed to the infrared camera and an RGB camera left and right with respect to the light source.
  6. 6. The method of claim 5,
    The light source and irradiated with a laser line with a cross-shaped pattern,
    The cross-shaped pattern is a three-dimensional image recognition apparatus using the two kinds of the camera in a vertical direction with a long axis in comparison to the horizontal direction.
  7. According to claim 1,
    Three-dimensional image recognition apparatus using the two kinds of the infrared camera RGB camera and the camera is arranged on the upper and lower side with respect to the light source.
  8. The method of claim 7,
    The light source and irradiated with a laser line with a cross-shaped pattern,
    The cross-shaped pattern is a three-dimensional image recognition apparatus using the two kinds of camera having a major axis in the horizontal direction than in the vertical direction.
  9. Irradiating the target object with a laser line from the light source;
    Obtaining an infrared image taken for the laser line is projected onto a target object by the irradiation of the light source;
    Obtaining an RGB image photographed on the projected object laser line by irradiation of the light source; And
    Three-dimensional image recognition method using the two kinds of cameras including the step of comparing the RGB image photographed with the infrared-up image obtained by the laser line noise is removed by the outside light, and performs three-dimensional image recognition.
  10. The method of claim 9, wherein the step of performing the three-dimensional image is recognized,
    Detecting the laser line that contains the noise component by removing the background image from the infrared image taken;
    Removing the background image from the RGB-up image, and detecting the common noise components contained in the background image is removed infrared captured image and the background image is removed, an RGB image photographed;
    Three-dimensional image recognition method using the two kinds of cameras including the step of removing the noise component detected from the line of the laser detected by the infrared image taken obtain a laser line noise component is removed by the external light.
  11. 10. The method of claim 9, wherein the light source is a three-dimensional image recognition method using the two kinds of cameras for irradiating a near infrared line laser (line laser) beam.
  12. 12. The method of claim 11, wherein the light source is a three-dimensional image recognition method using the two kinds of cameras for irradiating a near infrared line laser beam having a wavelength more than 890nm.
  13. 10. The method of claim 9,
    The infrared camera image and the RGB camera image is a 3D image recognition method using a camera heterologous obtain the RGB camera and an infrared camera is arranged on the right and left with respect to the light source.
  14. 14. The method of claim 13,
    The light source and irradiated with a laser line with a cross-shaped pattern,
    The cross-shaped pattern is three-dimensional image recognition methods have a vertical direction with a heterologous camera having a long axis in comparison to the horizontal direction.
  15. 15. The method of claim 14,
    Detecting the noise component,
    3 using two kinds of cameras for performing left and right scanning with respect to the infrared camera image and the RGB camera image in order to detect the common noise components contained in the background image is removed, the infrared-up image and the background RGB shot image in which the image is removed D image recognition method.
  16. 10. The method of claim 9,
    The infrared camera image and the RGB camera image is a 3D image recognition method using a camera heterologous obtain the RGB camera and an infrared camera is disposed at the upper and lower sides with respect to the light source.
  17. 17. The method of claim 16,
    The light source and irradiated with a laser line with a cross-shaped pattern,
    The cross-shaped pattern is a three-dimensional image recognition method using the two kinds of camera having a major axis in the horizontal direction than in the vertical direction.
  18. 18. The method of claim 17,
    Detecting the noise component,
    3 using two kinds of cameras to perform the vertical scan relative to the infrared camera image and the RGB camera image in order to detect the common noise components contained in the background image is removed, the infrared-up image and the background RGB shot image in which the image is removed D image recognition method.
PCT/KR2013/006301 2013-07-15 2013-07-15 Apparatus and method for recognizing 3d image using heterogeneous cameras WO2015008873A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/KR2013/006301 WO2015008873A1 (en) 2013-07-15 2013-07-15 Apparatus and method for recognizing 3d image using heterogeneous cameras

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/KR2013/006301 WO2015008873A1 (en) 2013-07-15 2013-07-15 Apparatus and method for recognizing 3d image using heterogeneous cameras

Publications (1)

Publication Number Publication Date
WO2015008873A1 true WO2015008873A1 (en) 2015-01-22

Family

ID=52346310

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2013/006301 WO2015008873A1 (en) 2013-07-15 2013-07-15 Apparatus and method for recognizing 3d image using heterogeneous cameras

Country Status (1)

Country Link
WO (1) WO2015008873A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110069189A1 (en) * 2008-05-20 2011-03-24 Pelican Imaging Corporation Capturing and processing of images using monolithic camera array with heterogeneous imagers
JP2012060602A (en) * 2010-09-13 2012-03-22 Konica Minolta Opto Inc Imaging device
KR20120038431A (en) * 2009-06-03 2012-04-23 플리어 시스템즈, 인크. Infrared camera systems and methods for dual sensor applications
KR20120071551A (en) * 2010-12-23 2012-07-03 한국전자통신연구원 Method and apparatus for user interraction by using structured lights
US20130002823A1 (en) * 2011-06-28 2013-01-03 Samsung Electronics Co., Ltd. Image generating apparatus and method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110069189A1 (en) * 2008-05-20 2011-03-24 Pelican Imaging Corporation Capturing and processing of images using monolithic camera array with heterogeneous imagers
KR20120038431A (en) * 2009-06-03 2012-04-23 플리어 시스템즈, 인크. Infrared camera systems and methods for dual sensor applications
JP2012060602A (en) * 2010-09-13 2012-03-22 Konica Minolta Opto Inc Imaging device
KR20120071551A (en) * 2010-12-23 2012-07-03 한국전자통신연구원 Method and apparatus for user interraction by using structured lights
US20130002823A1 (en) * 2011-06-28 2013-01-03 Samsung Electronics Co., Ltd. Image generating apparatus and method

Similar Documents

Publication Publication Date Title
US4611292A (en) Robot vision system
JP4983905B2 (en) Imaging device, 3d modeling data generating method, and a program
US6031941A (en) Three-dimensional model data forming apparatus
CN101068344B (en) Object detection apparatus
KR100948161B1 (en) Camera corrector
EP2275990A1 (en) 3D sensor
US9715734B2 (en) Image processing apparatus, imaging apparatus, and image processing method
US20150042759A1 (en) Device for optically scanning and measuring an environment
WO2009141998A1 (en) Calibration method, calibration device, and calibration system having the device
JP5266953B2 (en) Way projection-type display device and a display
JP2012123781A (en) Information processing device, information processing system and information processing method
US9300954B2 (en) Surrounding information-obtaining device for working vehicle
JP4095491B2 (en) Distance measuring device, a distance measuring method and a distance measuring program
KR20050052486A (en) Stereo door sensor
US9217637B2 (en) Device for optically scanning and measuring an environment
EP1526477A1 (en) Image processing apparatus
US20170193673A1 (en) Device for optically scanning and measuring an environment
JPH1098646A (en) Object extraction system
JP2004050390A (en) Work taking out device
JP5699788B2 (en) Screen area detection method and system
JP2011123071A (en) Image capturing device, method for searching occlusion area, and program
CN101274432A (en) Apparatus for picking up objects
JP2006071471A (en) Moving body height discrimination device
Vidas et al. A mask-based approach for the geometric calibration of thermal-infrared cameras
CN103413313B (en) Binocular vision navigation system and method for robot-based electricity

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13889572

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase in:

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13889572

Country of ref document: EP

Kind code of ref document: A1