JP2002170103A - Parking space map creating device and display device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To facilitate recognition of parking space by accurately determining whether or not there exists parking space and a depth position, and creating an ambient map. SOLUTION: The distance to a subject is measured using ultrasonic sensors 1 and 2 to determine whether or not there exists parking space. If parking space is determined to exist, a range within which its depth position is detected is determined based on images taken by electronic cameras 3 and 4. An integrated value of brightnesses is determined for each row in a horizontal direction within this detecting range and the absolute value of a difference in the integrated value between the rows is calculated. If the absolute value is equal to or greater than a threshold T2, a determination is made that parking space exists. In this case, the size of the parking space is also detected to determine whether or not one's own vehicle can be parked. If the absolute value is not less than a threshold T1 but less than T2, then a determination is made that there is the possibility of an obstacle existing, and a further determination is made as to whether or not an obstacle really exists. If the absolute value is less than T1, then a determination is made that there is no depth. Based on these pieces of information, a map showing the parking space is created and displayed in a display 12.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a parking space map making apparatus and a display apparatus using a distance measurement result using a non-directional detection signal.

[0002]

2. Description of the Related Art A method in which a laser mounted on a vehicle is scanned at a wide angle to detect the presence or absence of a parking space, its size, and the presence or absence of an obstacle in the parking space to determine whether or not it is possible to park in the parking space. (Japanese Patent Laid-Open No. 9-18 / 1990)
No. 0100).

[0003]

However, in this parking space determination method, since the laser is scanned at a wide angle, it takes some time until a determination result is obtained. In addition, the laser has a drawback that short distance accuracy is poor. Furthermore, it can be applied only when the back of the parking space has a surface having a certain expanse such as a wall or a vehicle.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a parking space map creating device and a display device which have a high distance measuring accuracy at a short distance and facilitate the recognition of a parking space situation by creating a parking space map.

[0005]

FIG. 1 shows an embodiment of the present invention.
The present invention will be described with reference to FIG. (1) A parking space map creation device according to the present invention includes distance measuring means 1 and 2 for measuring a distance to an object based on a non-directional detection signal, and an image for acquiring an image for determining the presence or absence of a parking space. Acquiring means 3, 4, a feature extracting means 8 for extracting a feature representative of the depth of the image based on the image acquired by the image acquiring means 3, 4, and a feature extracted by the feature extracting means. Calculating the distance of the vehicle based on the position on the acquired image and the measurement results of the distance measuring means 1 and 2, and parking at least based on the distance to the characteristic portion calculated by the position calculating means 8. The above object is achieved by providing the map creating means 8 for creating a map representing a space. (2) The invention according to claim 2 is the parking space map creating device according to claim 1, wherein the map creating means includes the calculated distance to the characteristic portion and the distance to the object measured by the distance measuring means 1 and 2. When the difference from the vehicle length is equal to or longer than the own vehicle length, a map indicating that there is a parking space is created, and the distance between the calculated characteristic portion and the distance to the target measured by the distance measuring means 1 and 2 is calculated. When the difference is less than the vehicle length, a map indicating that there is no parking space is created. (3) The invention according to claim 3 is the parking space mapping device according to claim 1 or 2, wherein the distance to the object measured by the distance measuring means 1 or 2 is at least equal to or greater than the entire length of the vehicle, or the distance is measured. The feature extracting means 8, the position calculating means 8, and the map creating means 8 are driven in a non-detection state where no result is obtained. (4) The invention of claim 4 is the parking space map creating device according to any one of claims 1 to 3, wherein the characteristic portion extracting means 8 is provided.
A luminance histogram creating means for obtaining an integrated value of the luminance values of each row arranged in the horizontal direction in the acquired image to create a luminance histogram; a difference calculating means for calculating a difference between the luminance histograms between adjacent rows; And a horizontal edge determining means for determining that a horizontal edge exists when the size of the horizontal edge is greater than or equal to a first predetermined value. (5) The invention according to claim 5 is the parking space map creating device according to claim 4, wherein the characteristic portion extracting means further comprises: when the magnitude of the difference is equal to or more than a first predetermined value and equal to or less than a second predetermined value.
A histogram creation unit that creates a brightness histogram by calculating an integrated value of brightness values of each column arranged in the vertical direction in the acquired image; a difference calculation unit that calculates a difference between brightness histograms between adjacent columns; Vertical edge determining means for determining that a vertical edge exists when the size is equal to or greater than a third predetermined value. (6) The invention according to claim 6 is the parking space map creating device according to claim 5, wherein the map creating means creates a map indicating that there is no parking space when it is determined that a vertical edge exists. Features. (7) A parking space map display device according to the invention of claim 7 includes the parking space map creation device according to any one of claims 1 to 6, and a display unit 12 that displays the created parking space map. Achieve the above objectives.

In the section of the means for solving the above-mentioned problems, the present invention is associated with FIG. 1 of the embodiment for easy explanation of the present invention, but the present invention is limited to the embodiment. Not something.

[0007]

As described above, according to the present invention, the following effects can be obtained. (1) According to the first to seventh aspects of the present invention, the distance is measured using the non-directional detection signal, and the parking space map is created by using the measurement result and the image processing together.
The time required for distance measurement can be reduced, and a highly reliable parking space map can be created at a high speed while the vehicle is running. (2) According to the invention of claim 3, when it is determined that there is no parking space based on the distance measurement result, unnecessary image processing calculations are not performed, so that the calculation speed can be improved, It is possible to create a parking space map at a higher speed. (3) According to the fourth aspect of the present invention, since the characteristic portion is extracted by using the brightness histogram of the image, the extraction time of the horizontal edge can be reduced, and the parking space map can be created more quickly. Become. (4) According to the fifth aspect of the present invention, since the obstacle is extracted by using the brightness histogram of the image, the time for extracting the obstacle is reduced, and a parking space map with high speed and high reliability can be obtained. It can be created. (5) According to the invention of claim 6, when an obstacle is detected, it is possible to reliably display a map indicating that there is no parking space.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described. FIG. 1 is a block diagram showing a configuration of an embodiment. The parking space mapping device and display device are:
Ultrasonic sensors 1 and 2 which are respectively installed laterally at the rear portions of both sides of the vehicle, and optical sensors 1 and 2 whose optical axes are
The ultrasonic sensors 1 and 2
Electronic cameras 3 and 4 which are also respectively installed facing the rear of the vehicle so as to overlap with the detection areas of
Memories 5, 6 for storing images taken by the electronic cameras 3, 4 as digital data, a vehicle behavior detecting device 7,
An arithmetic unit 8 including a PU, a RAM, a ROM, and the like;
And a display 12 for displaying a parking space map created based on the detection result of the surroundings of the vehicle. The vehicle behavior detecting device 7 includes a shift position sensor 9 for the vehicle.
, A wheel speed sensor 10 for detecting the left and right wheel speeds of the rear wheels of the vehicle, a computing device 11, and the like. The arithmetic unit 11 calculates the position, traveling direction, direction, moving distance, and the like of the vehicle from output values of the shift position sensor 9 and the wheel speed sensor 10. The digital data stored in the memories 5 and 6 is luminance value data having, for example, 0 to 256 gradations according to the luminance of each pixel constituting the image.

FIG. 2A is a diagram showing an output example of the measurement results when the measurement is performed using the parking space map creating apparatus. In FIG. 2A, the vehicle 15 equipped with the parking space map creation device and the display device detects the parking space while traveling forward at a low speed. FIG. 2 (a)
Are the wall 13 and the parked vehicle 14. These objects are scanned by the ultrasonic sensor 1 or 2,
When the output values of the detected distances are plotted, they become a hollow dot sequence 17 as shown in FIG.

The ultrasonic sensors 1 and 2 are non-directional, and detect the distance from the sensor to an object in a detection area. As for the detection range characteristics of the ultrasonic sensors 1 and 2, for example, when the output value is 3 m, an object within a range of ± 40 cm from the central axis of the sensor is detected. Therefore, the shape of the object 3 m away is expressed with a maximum error of 80 cm in the horizontal direction, and is recognized as having a width wider than the actual width. Also, a parked vehicle 1 as shown in FIG.
A concave portion having a distance from the sensor different from that of the adjacent object, such as a space 16 existing between the sensors 4, is not detected or is smaller than the actual width because it enters the detection range of the adjacent object. Is recognized as something.

Therefore, when considering a parking space, a narrow area such as a concave portion between adjacent vehicles is covered by the ultrasonic sensors 1 and 2.
In the case where there is a space that is not detected and is likely to be parked, the distance to the deepest part of the space is measured. Here, the distance to the deepest part is the distance from the ultrasonic sensors 1 and 2 to the deepest part, not the distance of the parking space itself. Black dot sequence 1 shown in FIG.
When the distance to the deepest portion is equal to or longer than the maximum detection distance L0 of the ultrasonic sensor as in 8, the distance is not detected and the state is a non-detection state. The distance to the deepest part is measured, and if the value is equal to or longer than the vehicle length L or is in the non-detection state, a process of detecting the depth position of the parking space is performed.
If the distance to the deepest part is less than or equal to the vehicle length L, it is determined that parking is not possible.

In order to detect the depth position of the parking space, the parking space is photographed using the electronic camera 3 or 4, and an image is obtained. A method of determining a region for detecting a depth position from an acquired image will be described with reference to FIG. FIG. 3 shows a case where a parked vehicle is present on the left side and a parking space is present on the right side thereof. The parking space is an area surrounded by white division lines 21.

The coordinate system of the image is set as shown in FIG. That is, the horizontal axis is the i-axis, the vertical axis is the j-axis, and the direction of the arrow in FIG. 3 is the positive direction. In addition, as a reference point of the coordinate system, infinite point coordinates (IV,
JV) is obtained in advance.

The distance from the ultrasonic sensor is R, and FIG.
In the image coordinate system shown in (1), the horizontal coordinate i and the vertical coordinate j of the coordinates (i, j) of the point at a position away from the sensor center axis by the distance P in the positive direction of the i-axis are respectively expressed by the following equations.

I = IV + x (P, R) (1)

## EQU2 ## j = JV + y (R) (2)

The range for detecting the depth position is i = i
It is an area surrounded by s, i = ie, j = js, and j = je. The values of is, ie, js, and je are determined as follows. The output value of the distance between the ultrasonic sensor and the target at time t is D (t). Where D (t)
Is a value corrected in consideration of the vehicle behavior. This is because, as shown in FIG. 2B, when the vehicle 15 travels obliquely with respect to the object, such as moving away from the object wall 13 or the parked vehicle 14, the output value point sequence 19 Is performed to prevent the output from being performed obliquely with respect to the object. When the output value D (t) after the correction is plotted, a point sequence 20 is obtained. Further, as described above, the distance measurement of the ultrasonic sensor involves an error. It is assumed that the distance output characteristic error is E, the width of the host vehicle is W, and the maximum detection distance of the ultrasonic sensor is L0.

First, values is and ie for determining a boundary line in the i-axis direction are obtained. The detection range is from -E to the vehicle width W based on the IV on the central axis of the sensor. Therefore, is and ie are represented by the following equations from equation (1).

## EQU00003 ## is = IV + x (-E, L0) (3)

## EQU00004 ## ie = IV + x (W, L0) (4)

Next, values js and je for determining a boundary line in the j-axis direction are obtained. The boundary line of the detection range on the host vehicle side is
The position is the position of the maximum detection distance L0 of the ultrasonic sensors 1 and 2, and the depth side is the position of the distance obtained by adding the length L of the own vehicle to the distance from the own vehicle to the object detected by the ultrasonic sensor before the time Δt. I do. However, the value of Δt is determined according to the vehicle speed V detected by the wheel speed sensor 10. That is, when the vehicle speed V is high, Δt is decreased, and when the vehicle speed V is low, Δt is reduced.
To increase. The distance from the host vehicle to the target detected by the ultrasonic sensor before the time Δt can be expressed as D (t−Δt). In the example of FIG. 3, this corresponds to the distance to the parked vehicle. Therefore, js and je are represented by the following equations from equation (2).

Js = JV + y (D (t−Δt) + L) (5)

Je = JV + y (L0) (6)

However, when formula (5) is obtained, time Δt
When the distance from the preceding vehicle to the target object cannot be detected and the state is in the non-detection state, js = JV. This is because the depth position is near the ground surface, as shown in FIG.
This is because it is always located below the point at infinity (positive direction of the j-axis).

When obtaining the detection range of the depth position,
The value for determining the boundary is calculated using Equations (3) to (6), but is described as one embodiment.
The present invention is not limited to this.

FIG. 4 is a diagram for explaining a method of detecting the depth position after determining the detection range. FIG.
(A-1) is a diagram when there is no obstacle in the parking space, and FIG. 4 (b-1) is a diagram when there is an obstacle in the parking space. In order to detect the depth position, it is only necessary to detect a lane marking, a car stop, a boundary between a wall and a road surface, a lateral edge existing in a vehicle behind, and the like.

FIG. 3 stored in the memory 5 or 6
Perform the following processing on the image data as shown,
Extract horizontal edges (features). First, from row je to row j
The luminance values in the horizontal direction are integrated for each row up to s. Coordinate
Let the luminance value at (i, j) be g _ijThen this coordinate
The integrated value of the luminance value in the horizontal direction in the row containing (i, j)
Σ_ig_ijIt is represented by Everything I asked in this way
4A based on the integrated value of the luminance value for the row of FIG.
2) or integrated histogram as shown in FIG. 4 (b-2)
Create a system. Next, the integration of the brightness values calculated for each row
Based on the value, an adjacent row (for example, FIG. 4A-2)
Row j_kAnd row j_{k + 1}) Is calculated, and the difference
Absolute value Hh_jIs determined by the following equation.

Equation 7] _{Hh j = | Σ i g ij} -Σ i g i (j + 1) | (7) is represented by Figure the absolute value Hh _j of the difference, FIG. 4 (a-3)
Alternatively, the result is as shown in FIG. Hh _j obtained by the equation (7) corresponds to the strength of the horizontal edge. At a position where a long horizontal edge such as a lane mark exists, Hh _{j is obtained.}
Becomes large.

Next, the value of Hh _j obtained by the equation (7) is compared with the threshold value T1 in the j-axis direction from je to js. As described above, at a position where a long horizontal edge exists, the value of Hh _j suddenly increases. Therefore, a position j1 exceeding a certain threshold T1 is set as a depth position.

However, even when an obstacle as shown in FIG. 4 (b-1) exists, as shown in FIG. 4 (b-3),
At the position jp where the obstacle exists, the value of Hh _j is equal to the threshold T.
May exceed 1. In this case, since the horizontal edge exists only at the place where the obstacle exists, the peak value of Hh _j is not as large as the peak value at the position j1 where the lane marking shown in FIG. 4A-3 exists. Thus, the T1 or higher, and set the threshold value T2 in a range of less than the peak value of the Hh _j positions the presence of partition lines, if the peak value of the Hh _j is less than T1 or T2, there are obstacles It is possible that

The above results are summarized as follows. If the peak value of Hh _j is equal to or greater than T2, there is no obstacle, and the position j at which the value of Hh _j exceeds a certain threshold value T1
Let 1 be the depth position. When the peak value of Hh _j is equal to or more than T1 and less than T2, there is a possibility that an obstacle exists. If the value of Hh _j in the detection range is equal to or less than T1, there is no depth such as a dividing line or a wall. This corresponds to, for example, a place like a passage shown in FIG.

If it is determined from the peak value of Hh _{j that} an obstacle may be present, the presence or absence of the obstacle is detected by detecting a vertical edge. The vertical edge detection method is
The same method as the method for detecting a horizontal edge is used. That is, the luminance values in the vertical direction are integrated for each of the columns from the column is to the column ie, and an integrated histogram as shown in FIG. Based on the integrated value of the calculated luminance values, it calculates the difference between the integrated value between adjacent columns and calculate the absolute value Hv _i from the following equation.

Equation 8] _{Hv i = | Σ j g ij} -Σ j g (i + 1) j | (8) horizontal axis i, is represented by FIG vertical axis as the absolute value of the difference,
The result is as shown in FIG. Compare them with a predetermined threshold value T3 in order from is a Hv _i obtained by the equation (8) in the i-axis direction until ie, if it is detected values above the threshold value T3, the vertical edges there It is determined that an obstacle exists within the detection range.

If there is no obstacle, the actual distance R from the ultrasonic sensor mounted on the vehicle to the depth position j1
Calculate _j1 . First, from equation (2), y (R
_j1 ) is calculated.

Y (R _j1 ) = j1−JV (9) The distance R _j1 to the actual depth position is calculated from y (R _j1 ) calculated by equation (9). This is given by equation (2).
It can be obtained by a procedure reverse to the procedure for obtaining y (R) from R.

The size S of the parking space is R _j1 and time Δ
Using the distance D (t−Δt) from the host vehicle to the target object detected by the ultrasonic sensor t before, it is obtained by the following equation.

S = R _j1 −D (t−Δt) (10) Instead of D (t−Δt) in equation (10), time Δt to n
Each distance D (t detected before Δt (n is an integer of 2 or more)
−Δt) to D (t−nΔt) may be used.
When D (t−Δt) is not detected and is in the non-detection state, the value of D (t−Δt) is set to a predetermined value (for example, 2 m) equal to or less than the own vehicle length L. If S calculated by the formula (10) is equal to or longer than the own vehicle length L, it is determined that there is a parking space for stopping the own vehicle, and if not more than L, it is determined that there is no space for stopping the own vehicle.

When the next parked vehicle is smaller than the host vehicle, such as a minicar, the distance D (t-Δ) to the parked vehicle is determined.
Since t) increases, the value of S calculated by the equation (10) becomes smaller than the actual size of the parking space. Therefore, the distance from the vehicle before the time Δt to the object (the next parked vehicle or the like) is larger than the predetermined value (for example, 2 m) in the non-detection state, and the total length difference between the own vehicle and the mini vehicle (for example, If it is smaller than a value obtained by adding a predetermined value in the non-detection state (about 1 m), it is determined that the vehicle is a small vehicle such as a minicar. At this time, the value of D (t−Δt) is set to a predetermined value (for example, 2 m) equal to or less than the own vehicle length L as in the case of the non-detection state.

Based on the information obtained so far, FIG.
A surrounding map which is a parking space map as shown in FIG. FIG. 5A shows a case where no obstacle exists and a depth such as a lane marking or a wall exists. This is created by detecting a parked vehicle or the like with an ultrasonic sensor and determining the distance R _j1 to the depth of the parking space. FIG. 5B shows a case where an obstacle is present, and FIG. 5C shows a case where no obstacle is present but depth is not detected. In the cases of FIGS. 5B and 5C, it is determined that there is no parking space. To create a map.

Based on the principle explained so far,
Detects the presence or absence of parking space and the distance to the depth,
The procedure up to creation of the surrounding map will be described with reference to the flowchart of FIG. The processing procedure in the flowchart is
The calculation is performed by the arithmetic unit 8.

In step S1, the distance to the object measured by the ultrasonic sensors 1 and 2 is read. If the distance is measured, the process proceeds to step S2, where the distance measured in step S1 is corrected based on information such as the position, traveling direction, direction, and moving distance of the vehicle obtained by the vehicle behavior detection device 7. , The distance D (t) to the object is calculated. D (t)
Is calculated, the process proceeds to step S3. If the distance cannot be measured in step S1 and the state is not detected, step S2
To step S3 via.

In step S3, a determination is made as to the possibility of a parking space. D calculated in step S2
If (t) is equal to or longer than the own vehicle length L or is in the non-detection state, it is determined that there is a possibility that a parking space exists, and the process proceeds to step S4. If D (t) is equal to or less than the vehicle length L, it is determined that there is no parking space, and the process proceeds to step S10.

In step S4, the electronic camera 3 or 4
A range for detecting the depth position is obtained in the image photographed in step S1 and stored in the memory 5 or 6. As described above, the detection range is i = is, i = ie, j = j
This is an area surrounded by s and j = je. These is, i
The values of e, js, and je are given by equations (3), (4),
(5) and (6). However, in the above-described embodiment, an example in which the vehicle moves forward is described. However, when the vehicle moves backward, that is, the traveling direction of the vehicle is the negative direction of the horizontal axis of the camera coordinate system. If they match, is and ie are respectively calculated from the following equations.

Is = IV + x (E, L0) (11)

## EQU12 ## ie = IV + x (-W, L0) (12)

After the detection range is obtained in step S4, the process proceeds to step S5. In step S5, the luminance values in the horizontal direction are integrated to obtain an integrated histogram (FIGS. 4A-2 and 4B-2).
2)), and based on this integrated value, the difference between the integrated values between adjacent rows is calculated (FIG. 4 (a-
3), (b-3)), and its absolute value Hh _j is obtained from equation (7). If Hh _j is calculated, the process proceeds to step S6.

In step S6, it is determined whether or not there is a depth in the parking space. Hh _j is sequentially scanned from the row je to js of the host vehicle, and the threshold value T1
If there is a value that is equal to or greater than that value, the position is set to the depth position j1
And proceeds to step S7. If there is no value that is equal to or greater than T1, it is determined that there is no parking space, such as a road with no depth, and the process proceeds to step S10.

In step S7, it is determined whether there is an obstacle in the parking space. If the peak value of Hh _j when the threshold value T1 is exceeded is equal to or greater than the threshold value T2, it is determined that there is a parking space, and the process proceeds to step S9. When the peak value of Hhj is equal to or smaller than the threshold value T2, it is determined that there is a possibility that an obstacle exists in the parking space, and the process proceeds to step S8.

In step S8, it is determined whether or not an obstacle actually exists in the parking space by detecting a vertical edge. Obtains an integrated histogram of the vertical direction of the brightness values of the detection region (Fig. 4 (c-1)), the absolute value Hv _i of the difference in luminance value between adjacent columns from equation (8) (FIG. 4 (c
-2)). Threshold T the Hv _i in the order from the column is to ie
3, and if there is a value that is equal to or greater than the threshold value T3, a vertical edge is detected and it is determined that an obstacle exists in the parking space, and the process proceeds to step S10. If a value equal to or greater than the threshold value T3 is not detected, it is determined that there is no obstacle and there is a parking space, and the process proceeds to step S9.

In step S9, when it is determined that there is a parking space, it is determined whether or not the size S of the parking space is large enough to stop the host vehicle. This is performed by calculating the value of S from equation (10) and comparing it with the host vehicle length L. If the value of S is equal to or greater than L, it is determined that the own vehicle can be stopped. If the value of S is equal to or less than L, it is determined that the own vehicle cannot be stopped.

In step S10, a surrounding map is created based on information such as the presence / absence of an obstacle and the presence / absence of a depth obtained so far. Steps S1 to S10 are repeatedly performed while the vehicle is traveling, and a plurality of pieces of information such as the distance R _j1 are obtained. Therefore, a parking space map (thick black line in FIG. 5) is created based on a plurality of pieces of information. If there is a space where the vehicle can be stopped, the distance R _j1 to the depth position calculated based on the detection of the parked vehicle or Expression (9)
Based on such information, a map as shown in FIG. 5A is created. When it is determined that parking is not possible, such as when the depth cannot be detected, when an obstacle exists, when the parking space does not have enough space to stop the own vehicle, etc., FIG.
As shown in (c), a map is created by drawing a straight line from the host vehicle to a position at a fixed distance L1 equal to or less than the host vehicle length L. The created map is displayed on the display 12.

According to the parking space creating device and the display device of the embodiment described above, the following effects can be obtained. (1) The distance is measured using an ultrasonic sensor which is a non-directional detection signal, and a parking space map is created based on the measurement result. A highly reliable parking space map can be created at a high speed while driving. (2) Unnecessary image processing calculation is not performed when it is determined that there is no parking space based on the distance measurement result (when step S3 is negatively determined), so that the calculation speed can be improved. It is possible to create a parking space map even faster.

(3) A row-direction luminance histogram of an image as shown in FIGS.
As shown in (a-3) and (b-3), the difference Hh between the rows
_{Since the} horizontal edge (characteristic portion) is extracted when _j is equal to or larger than the threshold value T2, the extraction time of the horizontal edge is reduced, and a parking space map can be created at a higher speed. (4) When the difference Hh _j between rows is equal to or more than the threshold value T1 and less than T2, a column-direction luminance histogram of an image as shown in FIG. 4C-1 is created, and is shown in FIG. 4C-2. like,
Since the difference Hv _i between columns is so as to extract a longitudinal edge (obstacle) when above the threshold T3, reduces the extraction time of the obstacle, even faster, and reliable parking space map Can be created.

Although a conventional distance sensor using a directional laser or the like can detect an object having a certain surface such as a wall, it can detect an elongated obstacle as shown in FIG. Because it could not be detected, the reliability of the parking space display was low. However, if image processing is used in combination as in the above-described embodiment, a narrow obstacle can be reliably detected, and as a result, the reliability of the parking space map is improved.

(5) As shown in FIGS. 5 (b) and 5 (c), when an obstacle is detected or when a passage is detected, a map display such that there is no parking space is performed. As a result, the reliability of the parking space map is improved.

In the above embodiment, the ultrasonic sensor is used as the non-directional distance sensor. However, a distance sensor using another medium may be used. It should be noted that the present invention is characterized in that a parking space map is created by using both distance measurement using a non-directional detection signal and image processing. Therefore, as long as the device can realize such a characteristic method, the configuration of the device, image processing, and the like are not limited to the above-described embodiment.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of an embodiment of a parking space map creation device and a display device according to the present invention.

2A is a diagram illustrating an output result when surroundings are detected using the parking space map creating device according to the present invention, and FIG. 2B is a diagram illustrating a state where a vehicle travels obliquely with respect to an object; Diagram showing output results in case

FIG. 3 is a diagram for explaining a method of determining a range for detecting a depth position from a captured image;

FIG. 4A is a diagram when an obstacle does not exist in a parking space; FIG. 4A is a diagram showing an integrated histogram of luminance values in a horizontal direction when an obstacle does not exist; a
-3) is a diagram showing the value of Hh _j when no obstacle is present, (b-1) is a diagram when an obstacle is present in the parking space, and (b-2) is a diagram where an obstacle is present (B-3) is a diagram showing an integrated histogram of luminance values in the horizontal direction in the case of
Is a diagram showing the value of Hh _j when an obstacle is present, (c)
-1) is a diagram showing the integrated histogram in the vertical direction of the brightness value, (c-2) is a diagram showing a value of Hv _i

5A is a map created when there is no obstacle and a parking space is present, FIG. 5B is a map created when an obstacle is present, and FIG. Map created if not obtained

FIG. 6 is a flowchart showing a procedure until a surrounding map is created.

[Explanation of symbols]

1,2 ... ultrasonic sensor, 3,4 ... electronic camera, 5,6 ...
Memory, 7: Vehicle behavior detecting device, 8: Computing device, 9: Shift position sensor, 10: Wheel speed sensor, 11: Computing device in the vehicle behavior detecting device, 12: Display, 1
3 ... wall, 14 ... parked vehicle, 15 ... vehicle, 16 ... space between parked vehicles, 17 ... point sequence showing output when measuring an object with an ultrasonic sensor, 18 ... point sequence showing non-detection part , 19
... A point sequence indicating an output when the vehicle travels obliquely with respect to the target, 20.
… Parking lot demarcation line

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G01B 11/00 G01B 11/00 H G06T 7/60 250 G06T 7/60 250Z 11/60 300 11/60 300 // G01S 15/08 G01S 15/08 F term (reference) 2F065 AA01 BB05 CC11 DD03 FF04 JJ19 QQ24 QQ27 QQ29 QQ31 5B050 AA07 BA17 EA07 FA02 FA13 5B057 AA16 BA11 BA24 CC03 CH01 DA07 DA16 DC16 DC19 50508 A05 AC05 EA26 5L096 BA04 CA04 FA06 FA35 FA66 FA67 FA69 LA05

Claims (7)

[Claims] 1. A distance measuring means for measuring a distance to an object based on a non-directional detection signal; an image acquiring means for acquiring an image for judging the presence or absence of a parking space; A feature extracting means for extracting a feature representative of the depth of the image based on the image; and a distance to the feature extracted by the feature extracting means, a position on the acquired image and the distance. A position calculating unit that calculates a result of the measurement by the measuring unit; and a map creating unit that creates a map representing a parking space based on at least the distance to the characteristic unit calculated by the position calculating unit. And a parking space mapping device. 2. The parking space map creating device according to claim 1, wherein the map creating means is configured to calculate a difference between the calculated distance to the characteristic portion and the distance to the object measured by the distance measuring means. If the vehicle length is equal to or longer than the own vehicle length, a map indicating that there is a parking space is created, and the difference between the calculated distance to the characteristic portion and the distance to the object measured by the distance measuring means is equal to the own vehicle length. A parking space map creation device, which creates a map indicating that there is no parking space when the number is less than or equal to. 3. The non-detection device according to claim 1, wherein the distance to the object measured by the distance measuring means is at least equal to or greater than the entire length of the vehicle, or a distance measurement result is not obtained. In the state, the parking space map creating device drives the characteristic portion extracting means, the position calculating means, and the map creating means. 4. The parking space map creating apparatus according to claim 1, wherein said characteristic portion extracting means obtains an integrated value of luminance values of respective rows arranged in a horizontal direction in the acquired image to obtain a luminance. Brightness histogram creating means for creating a histogram; difference calculating means for calculating a difference between the brightness histograms between adjacent rows; and a horizontal side for determining that a horizontal edge exists when the magnitude of the difference is equal to or greater than a first predetermined value. A parking space map creating device, comprising: an edge determining unit. 5. The parking space map creating device according to claim 4, wherein said characteristic portion extracting means further obtains when the magnitude of said difference is equal to or more than said first predetermined value and equal to or less than said second predetermined value. A histogram creation unit that creates a brightness histogram by calculating an integrated value of brightness values of each column arranged in a vertical direction in the image; a difference calculation unit that calculates a difference between the brightness histograms between adjacent columns; A vertical edge determining unit that determines that a vertical edge exists when the size of the parking space is equal to or greater than a third predetermined value. 6. A parking space map creating apparatus according to claim 5, wherein said map creating means creates a map indicating that there is no parking space when it is determined that said vertical edge exists. Parking space mapping equipment. 7. A parking space map display device, comprising: the parking space map creating device according to claim 1; and display means for displaying the created parking space map.