Image processing device and image processing method

1. An image processing device for detecting a parking section defined by T-markers, the image processing device comprising:

a camera that is configured to be mounted on a vehicle, the camera being configured to capture an image around the vehicle;

a marker detector that is configured to carry out an edge detection process on the image to detect the T-markers on a front end of the parking section, the front end being an end close to the vehicle when the vehicle enters the parking section;

a provisional parking frame setting part that is configured to detect an intersection of each of the detected T-markers on the front end and set a provisional parking frame based on the detected intersections of the T-markers on the front end, the provisional parking frame being set for assisting in parking the vehicle; and

a parking frame identifying part that is configured to identify a parking frame corresponding to the parking section,

wherein:

the marker detector is configured to detect the T-markers on a rear end of the parking section by carrying out the edge detection process in directions orthogonal to sidelines of the provisional parking frame along the sidelines of the provisional parking frame;

the parking frame identifying part is configured to identify the parking frame based on the detected intersections of the T-markers on the front end and intersections of the detected T-markers on the rear end; and

the T-markers on the front end and the T markers on the rear end of the parking section are not connected by a sideline.

2. The image processing device according to claim 1 , wherein the parking frame identifying part is configured to:

set a rear-end line by connecting corners of two of the detected T-markers on the rear end;

set sidelines respectively extending from the corners of the two of the detected T markers on the rear end, the sidelines being orthogonal to the rear-end line;

set a front-end line by connecting corners of two of the detected T-markers on the front end; and

identify the parking frame based on the set rear-end line, the set sidelines, and the set front-end line.

3. The image processing device according to claim 1 , wherein the marker detector is configured to determine a start point of the edge detection process by offsetting from an intersection between one of the sidelines of the provisional parking frame and a front-end line of the provisional parking frame toward a rear-end line of the provisional parking frame by a preset length along the one of the sidelines of the provisional parking frame.

4. The image processing device according to claim 1 , wherein the marker detector is configured to verify two of the detected T markers on the rear end based on at least one of following conditions (1) to (4), and

wherein the parking frame identifying part is configured to identify the parking frame based on the two T markers on the rear end verified by the marker detector;

(1) A distance between the two T-markers on the rear end is within a predetermined first threshold value;

(2) Each of a distance from a rear end of the vehicle to a first of the two T-markers on the rear end and a distance from the rear end of the vehicle to a second of the two T-markers on the rear end is within a predetermined second threshold value;

(3) A difference between a first position of the first of the two T markers on the rear end and a second position of the second of the two T markers on the rear end in a vehicle width direction is within a predetermined third threshold value when the vehicle is assumed to be parked at the provisional parking frame; or

(4) A difference between the first position of the first of the two T-markers on the rear end and the second position of the second of the two T-markers on the rear end in a vehicle longitudinal direction is within a predetermined fourth threshold value when the vehicle is assumed to be parked at the provisional parking frame.

5. An image processing method for detecting a parking section defined by T-markers, using an image around a vehicle captured by a vehicle-mounted camera, the image processing method comprising:

detecting the T-markers on a front end of the parking section by carrying out an edge detection process on the image, the front end being an end close to the vehicle when the vehicle enters the parking section;

detecting an intersection of each of the detected T-markers on the front end and setting a provisional parking frame based on the detected intersections of the T-markers on the front end, the provisional parking frame being set for assisting in parking the vehicle;

detecting the T-markers on a rear end of the parking section by carrying out the edge detection process in directions orthogonal to sidelines of the provisional parking frame along the sidelines of the provisional parking frame; and

identifying a parking frame corresponding to the parking section based on the detected intersections of the T-markers on the front end and intersections of the detected T-markers on the rear end,

wherein the T markers on the front end and the T-markers on the rear end of the parking section are not connected by a sideline.

6. The image processing method according to claim 5 , further comprising:

setting a rear-end line by connecting corners of two of the detected T-markers on the rear end;

setting sidelines respectively extending from the corners of the two of the detected T markers on the rear end, the sidelines being orthogonal to the rear-end line;

setting a front-end line by connecting corners of two of the detected T-markers on the front end; and

identifying the parking frame based on the set rear-end line, the set sidelines, and the set front-end line.

7. The image processing method according to claim 5 , further comprising:

determining a start point of the edge detection process by offsetting from an intersection between one of the sidelines of the provisional parking frame and a front-end line of the provisional parking frame toward a rear-end line of the provisional parking frame by a preset length along the one of the sidelines of the provisional parking frame.

8. The image processing method according to claim 5 , further comprising:

verifying two of the detected T-markers on the rear end based on at least one of following conditions (1) to (4); and

identifying the parking frame based on the two T markers on the verified rear end;

(1) A distance between the two T-markers on the rear end is within a predetermined first threshold value;

(2) Each of a distance from a rear end of the vehicle to a first of the two T-markers on the rear end and a distance from the rear end of the vehicle to a second of the two T markers on the rear end is within a predetermined second threshold value;

(3) A difference between a first position of the first of the two T markers on the rear end and a second position of the second of the two T markers on the rear end in a vehicle width direction is within a predetermined third threshold value when the vehicle is assumed to be parked at the provisional parking frame; or

(4) A difference between the first position of the first of the two T-markers on the rear end and the second position of the second of the two T-markers on the rear end in a vehicle longitudinal direction is within a predetermined fourth threshold value when the vehicle is assumed to be parked at the provisional parking frame.

9. The image processing device according to claim 1 , wherein the marker detector is configured to carry out the edge detection process from a position offset from the T markers on the front end along the sidelines of the provisional parking frame by a preset length.

10. The image processing method according to claim 5 , wherein the edge detection process is carried out from a position offset from the T markers on the front end along the sidelines of the provisional parking frame by a preset length.

11. The image processing device according to claim 1 , wherein the parking frame identifying part is configured to:

define a rear-end line by connecting the T markers on the rear end;

determine intersections between sidelines extending from the T-markers on the rear end in directions orthogonal to the rear-end line and a front-end line connecting the T markers on the front end; and

identify the parking frame based on the rear-end line, a straight line connecting the determined intersections, and the sidelines extending from the T markers on the rear end in directions orthogonal to the rear-end line.

12. The image processing method according to claim 5 , wherein the identifying the parking frame comprises:

defining a rear-end line by connecting the T-markers on the rear end;

determining intersections between sidelines extending from the T-markers on the rear end in directions orthogonal to the rear-end line and a front-end line connecting the T markers on the front end; and

identifying the parking frame based on the rear-end line, a straight line connecting the determined intersections, and the sidelines extending from the T markers on the rear end in directions orthogonal to the rear-end line.

13. The image processing device according to claim 1 , wherein the T-markers on the front end and the T markers on the rear end are defined by two white lines crossing each other.

14. The image processing method according to claim 5 , wherein the T-markers on the front end and the T markers on the rear end are defined by two white lines crossing each other.