Three-dimensional object detection device, and three-dimensional object detection method

• US 9,591,274 B2
• Filed: 07/26/2013
• Issued: 03/07/2017
• Est. Priority Date: 07/27/2012
• Status: Active Grant

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Clarion Co., Ltd. (Faurecia SA), Nissan Motor Co., Ltd.
• Inventors: Hayakawa, Yasuhisa, Fukata, Osamu, Takemura, Masayuki, Utagawa, Akira, Muramatsu, Shoji, Irie, Kota
• Primary Examiner(s): Kim, Hee-Yong
• Application Number:
  US14/410,760
• Publication Number:
  US 20150195496A1
• Time in Patent Office: 1,320 Days
• Field of Search: 348/118, 382/103
• US Class Current: 1/1
• CPC Class Codes: B60R 2300/607 from a bird's eye viewpoint
• CPC Class Codes: B60R 2300/8026 in addition to a rear-view mirror system
• CPC Class Codes: G03B 2217/005 Blur detection
• CPC Class Codes: G06T 2207/10028 Range image; Depth image; 3D point clouds
• CPC Class Codes: G06T 2207/30261 Obstacle
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/136 involving thresholding
• CPC Class Codes: G06V 10/141 Control of illumination
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: H04N 2101/00 Still video cameras
• CPC Class Codes: H04N 23/61 based on recognised objects
• CPC Class Codes: H04N 23/81 for suppressing or minimising disturbance in the image signal generation
• CPC Class Codes: H04N 7/188 Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position signal generation from motion picture films H04N5/253

Abstract

A three-dimensional object detection device has an image capturing unit, an object detection unit, first and second edge intensity calculation units, a day/night assessment unit and a controller. The day/night assessment unit assess whether it is currently daytime or nighttime when detecting a three-dimensional object based on the captured images. Upon assesses it is daytime, edges of a subject are extracted from a first edge extraction area, including a horizon reference area, and a threshold value for detecting the three-dimensional object is set based on the intensity of the edges in the first edge extraction area. Upon assesses it is nighttime, the edges of a subject are extracted from a second edge extraction area, including a road edge reference area, and a threshold value for detecting the three-dimensional object is set based on the intensity of the edges that are extracted from the second edge extraction area.

Claims

1. A three-dimensional object detection device comprising:

an image capturing device mounted on a host vehicle, and comprising a lens for capturing images behind the host vehicle;

a three-dimensional object detection unit programmed to detect a presence of a three-dimensional object rearward of the host vehicle based on captured images captured by the image capturing device;

a first edge intensity calculation unit programmed to extract edges of a subject in a first edge extraction area that includes at least a horizon reference area that references a horizon near a vanishing point of a road and to calculate an intensity of the edges in the first edge extraction area as a first edge intensity based on a distribution of the edges extracted from the first edge extraction area, the horizontal reference area being area including the horizon as well as an area from which extraction of a predetermined quantity of edges is expected in an image area corresponding to the horizon;

a second edge intensity calculation unit programmed to extract edges of a subject in a second edge extraction area that includes at least a road edge reference area that references a road edge, and to calculate an intensity of the edges in the second edge extraction area as a second edge intensity based on a distribution of the edges that are extracted from the second edge extraction area;

a day/night assessment unit programmed to assess whether currently daytime or nighttime exists; and

a controller programmed to control detection of the three-dimensional object by the three-dimensional object detection unit based on the first edge intensity in the first edge extraction area when an assessment has been made that daytime currently exists by the day/night assessment unit, and programmed to control the detection of the three-dimensional object by the three-dimensional object detection unit based on the second edge intensity in the second edge extraction area when an assessment has been made that nighttime currently exists.

2. The three-dimensional object detection device according to claim 1, further comprising

an image conversion unit programmed to perform viewpoint conversion of the captured images obtained by the image capturing device into bird'"'"'s-eye view images;

the three-dimensional object detection unit being programmed to detect a presence of a three-dimensional object based on differential waveform information when a peak value of the differential waveform information is equal to or greater than a predetermined second threshold value, and in which the differential waveform information is generated by aligning in bird'"'"'s-eye view positions of the bird'"'"'s-eye view images obtained at different times by the image conversion unit in bird'"'"'s-eye view, and counting a number of pixels that indicate a predetermined difference of equal to or greater than a first threshold value to form a frequency distribution; and

the controller is further programmed to control the detection of the three-dimensional object by the three-dimensional object detection unit by setting one of the first threshold value and the second threshold value based on the first edge intensity in the first edge extraction area when an assessment has been made that currently daytime exists by the day/night assessment unit, and programmed to control the detection of the three-dimensional object by the three-dimensional object detection unit by setting one of the first threshold value and the second threshold value based on the second edge intensity in the second edge extraction area when an assessment has been made that currently nighttime exists.

3. The three-dimensional object detection device according to claim 1, further comprising

an image conversion unit programmed to perform viewpoint conversion of the captured images obtained by the image capturing device into bird'"'"'s-eye view images;

the three-dimensional object detection unit programmed to detect a presence of a three-dimensional object based on edge information by detecting components having a luminance difference between adjacent pixel areas of equal to or greater than a predetermined first threshold value from the bird'"'"'s-eye view images obtained from the image conversion unit, and determine the three-dimensional object exists when a quantity of the edge information based on the edge components is equal to or greater than a predetermined second threshold value; and

the controller is further programmed to control the detection of the three-dimensional object by the three-dimensional object detection unit by setting one of the first threshold value and the second threshold value based on the first edge intensity in the first edge extraction area when an assessment has been made that currently daytime exists by the day/night assessment unit and controls the detection of the three-dimensional object by the three-dimensional object detection unit by setting one of the first threshold value and the second threshold value based on the second edge intensity in the second edge extraction area when an assessment has been made that currently nighttime exists.

4. The three-dimensional object detection device according to claim 2, wherein

the controller is further programmed to change one of the first threshold value and the second threshold value to a higher value as one of the first edge intensity and the second edge intensity increases and programmed to change one of the first threshold value and the second threshold value to a lower value one of as the first edge intensity and the second edge intensity decreases.

5. The three-dimensional object detection device according to claim 2, wherein

the three-dimensional object detection unit is further programmed to detect the three-dimensional object in a left-side detection area and a right-side detection area that are set to the left and right rear of the host vehicle;

when changing the first threshold value based on one of the first edge intensity and the second edge intensity, the controller is further programmed to change the first threshold value so that a difference between the first threshold values in the left-side detection area and the right-side detection area when an assessment has been made that the currently daytime exists is equal to or less than a difference between the first threshold values in the left-side detection area and the right-side detection area when an assessment has been made that currently nighttime exists; and

when changing the second threshold value based on one of the first edge intensity and the second edge intensity, the controller is further programmed to change the second threshold value so that a difference between the second threshold values in the left-side detection area and the right-side detection area when an assessment has been made that the currently daytime exists is equal to or less than a difference between the second threshold values in the left-side detection area and the right-side detection area when an assessment has been made that currently nighttime exists.

6. The three-dimensional object detection device according to claim 2, wherein

the three-dimensional object detection unit programmed to detect the three-dimensional object in a left-side detection area and a right-side detection area that are set at the left and right rear of the host vehicle;

the second edge intensity calculation unit is further programmed to set a left-side road edge reference area and a right-side road edge reference area that reference the left and right road edges as the second edge extraction areas, programmed to calculate the second edge intensity in the left-side road edge reference area based on the intensity of the edges in the left-side road edge reference area, and programmed to calculate the second edge intensity in the right-side road edge reference area based on the intensity of the edges in the right-side road edge reference area; and

the controller is further programmed to set the value of one of the first threshold value and the second threshold value in the left-side detection area based on the second edge intensity in the left-side road edge reference area, and programmed to set the value of one of the first threshold value and the second threshold value in the right-side detection area based on the second edge intensity in the right-side road edge reference area.

7. The three-dimensional object detection device according to claim 2, wherein

the controller is further programmed to prohibit a change in the value of the first threshold value or the second threshold value when the three-dimensional object is being detected by the three-dimensional object detection unit.

8. The three-dimensional object detection device according to claim 2, wherein

the controller being further programmed to increase the value of one of the first threshold value and the second threshold value when the luminance of the captured image is less than a predetermined value, as compared to when the luminance of the captured image is equal to or greater than the predetermined value.

9. The three-dimensional object detection device according to claim 1, wherein

the controller being further programmed to control the detection of the three-dimensional object by the three-dimensional object detection unit by conducting exposure control of the image capturing device based on the first edge intensity in the first edge extraction area when an assessment has been made that currently daytime exists by the day/night assessment unit and controls the detection of the three-dimensional object by the three-dimensional object detection unit by conducting exposure control of the image capturing device based on the second edge intensity in the second edge extraction area when an assessment has been made that currently nighttime exists.

10. The three-dimensional object detection device according to claim 9, wherein

an image conversion unit programmed to perform viewpoint conversion of the captured images obtained by the image capturing device into bird'"'"'s-eye view images;

the three-dimensional object detection unit being programmed to detect a presence of a three-dimensional object based on differential waveform information when a peak value of the differential waveform information is equal to or greater than a predetermined second threshold value, and in which the differential waveform information is generated by aligning in bird'"'"'s-eve view positions of the bird'"'"'s-eye view images obtained at different times by the image conversion unit in bird'"'"'s-eye view, and counting a number of pixels that indicate a predetermined difference in a differential image of the aligned bird'"'"'s-eye view images to form a frequency distribution.

11. The three-dimensional object detection device according to claim 9, wherein

an image conversion unit programmed to perform viewpoint conversion of the captured images obtained by the image capturing device into bird'"'"'s-eye view images;

the three-dimensional object detection unit programmed to detect a presence of a three-dimensional object based on edge information from the bird'"'"'s-eye view images obtained by the image conversion unit.

12. The three-dimensional object detection device according to claim 9, wherein

the three-dimensional object detection unit is further programmed to detect the three-dimensional object in a left-side detection area and a right-side detection area that are set to the left and right rear of the host vehicle;

and

the controller is further programmed to conduct exposure control of the image capturing device so that a difference in exposures between the left-side detection area and the right-side detection area when an assessment has been made that currently daytime exists becomes equal to or less than a difference in the exposures between the left-side detection area and the right-side detection area when an assessment has been made that currently nighttime exists.

13. The three-dimensional object detection device according to claim 9, wherein

the three-dimensional object detection unit is further programmed to detect the three-dimensional object in a left-side detection area and a right-side detection area that are set to the left and right rear of the host vehicle;

the second edge intensity calculation unit is further programmed to set a left-side road edge reference area and a right-side road edge reference area that reference the left and right road edges as the second edge extraction areas, programmed to calculate the second edge intensity in the left-side road edge reference area based on the intensity of the edges in the left-side road edge reference area, and programmed to calculate the second edge intensity in the right-side road edge reference area based on the intensity of the edges in the right-side road edge reference area; and

the controller is further programmed to set an exposure in the left-side detection area based on the second edge intensity in the left-side road edge reference area, and programmed to set an exposure in the right-side detection area based on the second edge intensity in the right-side road edge reference area.

14. The three-dimensional object detection device according to claim 9, wherein

the controller is further programmed to maintain an exposure of the image capturing device regardless of the first edge intensity or the second edge intensity when the three-dimensional object is being detected by the three-dimensional object detection unit.

15. The three-dimensional object detection device according to claim 9, wherein

the control controller being further programmed to decrease an exposure amount of the image capturing device when the luminance of the captured image is less than a predetermined value, as compared to when the luminance of the captured image is equal to or greater than the predetermined value.

16. The three-dimensional object detection device according to claim 9, wherein

the controller is further programmed to conduct exposure control of the image capturing device by changing one of an aperture value of the image capturing device to a value on an open side, a capturing sensitivity of the image capturing device to a higher value, and an exposure time of the image capturing device to a longer time as the value of one of the first edge intensity and the second edge intensity decreases.

17. The three-dimensional object detection device according to claim 1, wherein

the first edge intensity calculation unit is further programmed to set an area that references the horizon near the vanishing point as the horizon reference area.

18. The three-dimensional object detection device according to claim 1, wherein

the first edge intensity calculation unit is further programmed to set, in addition to the horizon reference area, a road edge reference area that references the road edge as the first edge extraction area, and further programmed to calculate the intensity of the edges in the horizon reference area as the first edge intensity when the intensity of the edges in the horizon reference area is equal to or greater than a predetermined value, and further programmed to calculate the intensity of the edges in the area with the highest edge intensity, from among the horizon reference area and the road edge reference area, as the first edge intensity when the intensity of the edges in the horizon reference area is less than the predetermined value.

19. A three-dimensional object detection method comprising:

converting viewpoints of captured images into bird'"'"'s-eye view images;

aligning positions of the bird'"'"'s-eye view images that are obtained at different times in a bird'"'"'s-eye view;

generating differential waveform information by counting a number of pixels that indicate a predetermined difference that is equal to or greater than a predetermined first threshold value in a differential image of aligned bird'"'"'s-eye view images to form a frequency distribution, and

detecting a presence of a three-dimensional object based on the differential waveform information when a peak value of the differential waveform information is equal to or greater than a predetermined second threshold value;

assessing whether currently present daytime or nighttime exists;

extracting edges of a subject in a first edge extraction area, including at least a horizon reference area that references the horizon near a vanishing point of a road, when the assessment has been made that currently daytime exists, the horizontal reference area being area including the horizon as well as an area from which extraction of a predetermined quantity of edges is expected in an image area corresponding to the horizon;

calculating an intensity of the edges in the first edge extraction area as a first edge intensity based on a distribution of the edges extracted from the first edge extraction area, when the assessment has been made that currently daytime exists;

setting one of the first threshold value and the second threshold value based on the first edge intensity, when the assessment has been made that currently daytime exists;

extracting edges of a subject in a second edge extraction area, including a road edge reference area that references the road edge, when an assessment has been made that currently nighttime exists;

calculating an intensity of the edges in the second edge extraction area as a second edge intensity based on a distribution of the edges extracted from the second edge extraction area, when the assessment has been made that currently nighttime exists; and

setting one of the first threshold value and the second threshold value based on the second edge intensity, when the assessment has been made that currently nighttime exists.

20. A three-dimensional object detection method comprising:

converting viewpoints of captured images into bird'"'"'s-eye view images;

detecting a edge components having a luminance difference between adjacent pixel areas of equal to or greater than a predetermined first threshold value are detected from the bird'"'"'s-eye view images;

detecting a presence of a three-dimensional object based on the edge information when a quantity of the edge information based on edge components is equal to or greater than a predetermined second threshold value;

assessing whether currently daytime or nighttime exists;

extracting edges of a subject in a first edge extraction area, including at least a horizon reference area that references the horizon near a vanishing point of a road, when an assessment has been made that currently daytime exists, the horizontal reference area being area including the horizon as well as an area from which extraction of a predetermined quantity of edges is expected in an image area corresponding to the horizon;

calculating an intensity of the edges in the first edge extraction area as a first edge intensity based on a distribution of the edges extracted from the first edge extraction area, when the assessment has been made that currently daytime exists; and

setting one of the first threshold value and the second threshold value based on the first edge intensity, when the assessment has been made that currently daytime exists;

extracting edges of a subject in a second edge extraction area, including a road edge reference area that references the road edge, when an assessment has been made that currently nighttime exists;

calculating an intensity of the edges in the second edge extraction area as a second edge intensity based on a distribution of the edges extracted from the second edge extraction area, when the assessment has been made that currently nighttime exists; and

setting one of the first threshold value and the second threshold value based on the second edge intensity, when the assessment has been made that currently nighttime exists.

21. A three-dimensional object detection method comprising:

converting viewpoints of captured images from an image capturing device into bird'"'"'s-eye view images;

generating differential waveform information based on a difference between the bird'"'"'s-eye view images obtained at different times, and

detecting a presence of a three-dimensional object based on the differential waveform information;

assessing whether currently daytime or nighttime exists;

extracting edges of a subject in a first edge extraction area, including at least a horizon reference area that references the horizon near a vanishing point of a road, when the assessment has been made that currently daytime exists, the horizontal reference area being area including the horizon as well as an area from which extraction of a predetermined quantity of edges is expected in an image area corresponding to the horizon;

calculating an intensity of the edges in the first edge extraction area as a first edge intensity based on a distribution of the edges extracted from the first edge extraction area, when the assessment has been made that currently daytime exists;

conducting exposure control of the image capturing device based on the first edge intensity, when the assessment has been made that currently daytime exists;

extracting edges of a subject in a second edge extraction area, including a road edge reference area that references the road edge, when an assessment has been made that currently nighttime exists;

calculating an intensity of the edges in the second edge extraction area as a second edge intensity based on a distribution of the edges extracted from the second edge extraction area, when an assessment has been made that currently nighttime exists; and

conducting exposure control of the image capturing device based on the second edge intensity, when an assessment has been made that currently nighttime exists.

22. A three-dimensional object detection method in which

converting viewpoints of captured images from an image capturing device into bird'"'"'s-eye view images;

detecting edge information based on the bird'"'"'s-eye view images;

detecting a presence of a three-dimensional object based on the edge information;

assessing whether currently daytime or nighttime exists;

extracting edges of a subject in a first edge extraction area, including at least a horizon reference area that references the horizon near a vanishing point of a road, when the assessment has been made that currently daytime exists, the horizontal reference area being area including the horizon as well as an area from which extraction of a predetermined quantity of edges is expected in an image area corresponding to the horizon;

calculating an intensity of the edges in the first edge extraction area as a first edge intensity based on a distribution of the edges extracted from the first edge extraction area, when the assessment has been made that currently daytime exists;

conducting exposure control of the image capturing device based on the first edge intensity, when the assessment has been made that currently daytime exists;

extracting edges of a subject in a second edge extraction area, including a road edge reference area that references the road edge, when an assessment has been made that currently nighttime exists;

calculating an intensity of the edges in the second edge extraction area as a second edge intensity based on a distribution of the edges extracted from the second edge extraction area, when an assessment has been made that currently nighttime exists; and

conducting exposure control of the image capturing device based on the second edge intensity in the second edge extraction area, when an assessment has been made that currently nighttime exists.