Image processing device and image processing method

Patent metadata

Field	Value
Publication number	US-11250290-B2
Application number	US-201916574462-A
Country	US
Kind code	B2
Filing date	Sep 18, 2019
Priority date	Dec 14, 2018
Publication date	Feb 15, 2022
Grant date	Feb 15, 2022

Abstract

An image processing device includes: an extraction unit which extracts, from image data taken by shooting a neighborhood of a vehicle, edge points that are to be used for detecting partition lines of a parking frame and whose edge intensity values are larger than an edge threshold value; and a correction unit which corrects the edge threshold value based on a density of the edge points extracted by the extraction unit.

First claim

What is claimed is:

1. An image processing device comprising:
a computer including a hardware processor configured to function as an extraction unit and a correction unit, wherein
the extraction unit extracts, from image data taken by shooting a neighborhood of a vehicle, edge points that are to be used for detecting partition lines of a parking frame and whose edge intensity values are larger than an edge threshold value; and
the correction unit corrects the edge threshold value only when a density of the edge points extracted by the extraction unit is outside of a prescribed range.

2. The image processing device according to claim 1 , wherein the extraction unit extracts edge points again using the edge threshold value that has been corrected by the correction unit and supplies the edge points extracted again to a downstream unit.

3. The image processing device according to claim 1 , wherein the correction unit corrects the edge threshold value based on the density of edge points in a target region where a resolution of the image data of a road surface is higher compared to other regions of the image data.

4. The image processing device according to claim 1 , wherein the correction unit corrects the edge threshold value so that a number of the edge points extracted by the extraction unit becomes equal to a prescribed number.

5. The image processing device according to claim 1 , wherein the density of edge points is a number of the edge points that have been extracted by the extraction unit from all or a portion of the image data.

6. The image processing device according to claim 1 , wherein:
the hardware processor is further configured to function as an edge line detection unit, and a parking frame detection unit;
the extraction unit re-extracts edge points from the image data using the edge threshold value that has been corrected by the correction unit when the density of the edge points previously extracted by the extraction unit is outside of the prescribed range, so that the density of the edge points re-extracted using the edge threshold value that has been corrected by the correction unit is within the prescribed range;
the edge line detection unit detects edge lines by connecting either (i) the edge points that have been extracted by the extraction unit when the density of the edge points extracted by the extraction unit is within the prescribed range, or (ii) the edge points that have been re-extracted by the extraction unit when the density of the edge points previously extracted by the extraction unit is outside of the prescribed range; and
the parking frame detection unit detects a parking frame based on the edge lines detected by the edge line detection unit and that have been designated as partition line candidates.

7. The image processing device according to claim 2 , wherein the correction unit corrects the edge threshold value based on the density of edge points in a target region where a resolution of the image data of a road surface is higher compared to other regions of the image data.

8. The image processing device according to claim 2 , wherein the correction unit corrects the edge threshold value so that a number of the edge points extracted by the extraction unit becomes equal to a prescribed number.

9. The image processing device according to claim 3 , wherein the correction unit generates a histogram of edge intensity values of respective pixels in the target region and corrects the edge threshold value based on the histogram.

10. The image processing device according to claim 3 , wherein the correction unit corrects the edge threshold value so that a number of the edge points extracted by the extraction unit becomes equal to a prescribed number.

11. The image processing device according to claim 7 , wherein the correction unit generates a histogram of edge intensity values of respective pixels in the target region and corrects the edge threshold value based on the histogram.

12. The image processing device according to claim 7 , wherein the correction unit corrects the edge threshold value so that a number of the edge points extracted by the extraction unit becomes equal to a prescribed number.

13. An image processing method comprising:
extracting, by a computer having a hardware processor, from image data taken by shooting a neighborhood of a vehicle, edge points that are to be used for detecting partition lines of a parking frame and whose edge intensity values are larger than an edge threshold value; and
correcting, by the computer having the hardware processor, the edge threshold value only when a density of the extracted edge points is outside of a prescribed range.

14. The image processing method according to claim 13 , wherein the density of edge points is a number of the edge points that have been extracted from all or a portion of the image data.

15. The image processing method according to claim 13 , further comprising:
re-extracting, by the computer having the hardware processor, edge points from the image data using the edge threshold value that has been corrected by the correcting step when the density of the edge points previously extracted by the extracting step is outside of the prescribed range, so that the density of the edge points re-extracted using the edge threshold value that has been corrected is within the prescribed range;
detecting, by the computer having the hardware processor, edge lines by connecting either (i) the edge points that have been extracted by the extracting step when the density of the edge points extracted by the extracting step is within the prescribed range, or (ii) the edge points that have been re-extracted by the re-extracting step when the density of the edge points previously extracted by the extracting step is outside of the prescribed range; and
detecting, by the computer having the hardware processor, a parking frame based on the edge lines that have been detected and that have been designated as partition line candidates.

16. An image processing device comprising:
a computer including a memory and a hardware processor, the memory storing an edge threshold value, the hardware processor configured to function as an extraction unit, a correction unit, an edge line detection unit, and a parking frame detection unit, wherein
the extraction unit extracts, from image data taken by shooting a neighborhood of a vehicle, edge points that are to be used for detecting partition lines of a parking frame and whose edge intensity values are larger than the edge threshold value that is stored in the memory;
the correction unit corrects the edge threshold value that is stored in the memory only when a density of the edge points extracted by the extraction unit is outside of a prescribed range, the density of the edge points being a number of the edge points that have been extracted from all or a portion of the image data;
the extraction unit re-extracts edge points from the image data using the edge threshold value that has been corrected by the correction unit when the density of the edge points previously extracted by the extraction unit is outside of the prescribed range, so that the density of the edge points re-extracted using the edge threshold value that has been corrected by the correction unit is within the prescribed range;
the edge line detection unit detects edge lines by connecting either (i) the edge points that have been extracted by the extraction unit when the density of the edge points extracted by the extraction unit is within the prescribed range, or (ii) the edge points that have been re-extracted by the ex

Assignees

• Denso Ten Ltd

Inventors

• Okada Yasutaka
• Sano Hiroaki
• Yamamoto Tetsuo
• Yoshihara Atsushi
• Kanetake Jun
• Yoshimura Ryo
• Shidori Tomoki

Classifications

• G06V20/586 Primary

  of parking space · CPC title

• G06V10/44

  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 title

• G06K9/4647

  Physics · mapped topic

• G06K9/00812

  Physics · mapped topic

• G06K9/4604

  Physics · mapped topic