Image processing device and image processing method
An image processing device is equipped with a marker detector, an oblique line detector, and a parking frame detector. The marker detector extracts, from an image, positive and negative edge segments that are aligned next to each other at a predetermined interval as a pair, and detects a parking space division marker based on the pair. The oblique line detector detects an edge on one side of an edge segment (single-edge segment) that is not extracted as the pair, and detects a presumed straight line based on the single-edge segment connected with an oblique line. The parking frame detector detects neighboring edge segments based on the positive and negative edge segments as well as the single-edge segment of the presumed straight line, and detects a parking space based on a distance between the neighboring edge segments.
1. An image processing device comprising a processor, the processor being configured to:
scan an image generated by an image signal in a predetermined direction, the image signal being captured by an imaging device for imaging a road surface around a vehicle,
detect a first edge segment at which the image signal increases by a predetermined value or more and a second edge segment at which the image signal decreases by a predefined value or more,
filter the detected first edge segment and the detected second edge segment with a reference length,
extract the filtered first edge segment and the filtered second edge segment that are aligned next to each other at a predetermined interval as a pair,
detect a parking space division marker based on the extracted pair,
select the filtered first edge segment and the filtered second edge segment that are not extracted as the pair,
define the selected edge segment as a single-edge segment,
scan the image from a scanning start point on one side of the single-edge segment in a direction along the single-edge segment to determine an existence of an oblique line,
detect a straight line based on the single-edge segment connected with the oblique line,
assign the detected straight line as a presumed straight line that forms a marker having the oblique line,
detect neighboring edge segments aligned next to each other based on the first edge segment and the second edge segment of the parking space division marker as well as the single-edge segment of the presumed straight line, and
detect a parking space based on a distance between the detected neighboring edge segments.
2. The image processing device according to claim 1 , wherein the processor is configured to determine that the oblique line exists when detecting, at a predetermined interval, an edge segment at which the image signal increases by the predetermined value or more and an edge segment at which the image signal decreases by the predefined value or more at least once.
3. The image processing device according to claim 1 , wherein the processor is configured to scan an area in which the image signal is higher with respect to the single-edge segment when detecting the oblique line.
4. The image processing device according to claim 1 , wherein the processor is configured to detect the edge segment based on a luminance pattern or a density pattern of the image signal.
5. The image processing device according to claim 1 , further comprising a memory that is configured to store the reference length and a line width of the parking space division marker, an angle indicating an extending direction of the parking space division marker, as well as a width of the parking space.
6. An image processing method for detecting a parking space, the method comprising:
scanning an image generated by an image signal in a predetermined direction, the image signal being captured by an imaging device for imaging a road surface around a vehicle,
detecting a first edge segment at which the image signal increases by a predetermined value or more and a second edge segment at which the image signal decreases by a predefined value or more,
filtering the detected first edge segment and the detected second edge segment with a reference length,
extracting the filtered first edge segment and the filtered second edge segment that are aligned next to each other at a predetermined interval as a pair,
detecting a parking space division marker based on the extracted pair,
selecting the filtered first edge segment and the filtered second edge segment that are not extracted as the pair,
defining the selected edge segment as a single-edge segment,
scanning the image from a scanning start point on one side of the single-edge segment in a direction along the single-edge segment to determine an existence of an oblique line,
detecting a straight line based on the single-edge segment connected with the oblique line,
assigning the detected straight line as a presumed straight line that forms a marker having the oblique line,
detecting neighboring edge segments aligned next to each other based on the first edge segment and the second edge segment of the parking space division marker as well as the single-edge segment of the presumed straight line, and
detecting a parking space based on a distance between the detected neighboring edge segments.
7. The image processing method according to claim 6 , further comprising determining an existence of the oblique line when detecting, at a predetermined interval, an edge segment at which the image signal increases by the predetermined value or more and an edge segment at which the image signal decreases by the predefined value or more at least once.
8. The image processing method according to claim 6 , further comprising scanning an area in which the image signal is higher with respect to the single-edge segment when detecting the oblique line.
9. The image processing method according to claim 6 , further comprising detecting the edge segment based on a luminance pattern or a density pattern of the image signal.