JPH10206118A - Object recognizing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent false recognition and to enable correct recognition by dividing images obtained from a multilevel line-type CCD along line rows and windows to measure a distance for each region, and calculating a distance for every line from distance data in every region to recognize an object. SOLUTION: At a distance measuring means 16, an image of a multilevel CCD 11 is plurally divided for every line in the direction of windows to measure a distance in each region. A line distance calculating means 26 calculates a distance in every line on the basis of a distance for every region measured by the distance measuring means 16, and a distance for every line is recognized at an object recognizing means 20. Next, a distance dead zone is set on every CCD line by a dead zone setting means 34, and distances belonging to this dead zone among distances for every line calculated buy the like distance calculating means 26 are not used for the recognition of objects at the object recognizing means 20. As a result of this, an object right behind is not falsely recognized as an object diagonally behind, and it is possible to correctly recognize objects such as passing vehicles.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an object recognizing apparatus, and more particularly, to an object recognizing apparatus which recognizes an object based on distance data obtained by a multi-stage line CCD.

[0002]

2. Description of the Related Art Conventionally, as this kind of object recognizing apparatus, for example, as shown in Japanese Patent Application Laid-Open No. Hei 7-225126, a vehicle is equipped with a stereo camera for photographing a running road surface, and images taken by this camera are taken. Is divided into multiple rows in the vertical direction and multiple columns in the horizontal direction to divide multiple windows, and recognize objects based on the distance in each window, and set the distance of each window in advance. Compared with the reference distance for each range, if the distance of the window is less than or equal to the reference distance for each range,
While the distance is used as distance data relating to the object, when the distance is larger than a reference distance for each range, a so-called range cut for excluding the distance from the data is known.

[0003]

As an example of such an object recognition apparatus mounted on a vehicle, a large number of CCDs arranged in a fixed direction are arranged in multiple stages in a direction orthogonal to the arrangement direction. It is conceivable to provide a multi-line CCD and recognize a specific object from two-dimensional distance data obtained based on the multi-line CCD. This multi-stage line CCD has a configuration in which CCDs are thinned out in one direction with respect to a sensor for a camera or the like in which a large number of CCDs are arranged vertically and horizontally, and the number of image data is reduced as the number of CCDs is reduced. As a result, the calculation speed is increased and the cost can be reduced.

Further, the multi-stage line CCD is disposed on the left and right sides at the rear of the vehicle to recognize objects such as other vehicles and two-wheeled vehicles (motorcycles) behind the vehicle and to sequentially transmit the information to the vehicle driver. If the notification is made, the operation of the driver checking the rear with a mirror or the like can be reduced, and the burden on the driver can be reduced.

[0005] By the way, for example, there is another lane beside the lane in which the vehicle is running, and when the vehicle or the like is running in this other lane, the vehicle or the like passes through the side of the vehicle. There is a high possibility of overtaking, and if this movement is not sufficiently recognized, it becomes an obstacle when the own vehicle changes lanes to another lane. For this reason, the direction of the multi-stage line CCD may be set to a direction obliquely rearward of the vehicle such that the image is centered on the object on the other lane.

However, in such a case, when a vehicle or the like approaches the vehicle immediately behind the vehicle on the lane of the vehicle, the image captured by the multi-stage line CCD indicates that the vehicle or the like does not exist in another lane. Even so, the vehicle or the like immediately behind the vehicle may be erroneously recognized as a vehicle or the like in another lane, and the vehicle or the like on another lane may not be accurately recognized.

[0007] Further, even if a vehicle or the like approaches a vehicle directly behind the vehicle in a state where the vehicle or the like in another lane is recognized from a distance, the image of the multi-stage line CCD immediately follows the vehicle. The vehicle etc. overlaps with the vehicle etc. on another lane, and the vehicle etc. directly behind it is erroneously recognized as a vehicle etc. on another lane,
A similar problem occurs.

The present invention has been made in view of the above points, and an object of the present invention is to limit the distance measurement range when an object is recognized by measuring the distance using a multi-stage line type CCD. An object of the present invention is to accurately recognize a distant object without overlooking it, and prevent an object immediately behind from being erroneously recognized as an obliquely-sided object, thereby enabling accurate recognition.

[0009]

In order to achieve the above object, according to the present invention, an image obtained by a multi-stage line type CCD is divided into a line direction and a window direction, and a distance is measured for each region. The object is recognized by calculating the distance for each line from the distance data for each area. Further, a dead zone is formed in the recognition range of the object by utilizing the concept of the conventional example, and the distance belonging to the dead zone among the detected distance data is not used as the distance data for the object recognition.

More specifically, according to the first aspect of the present invention, as shown in FIG. 1, CCD lines composed of a large number of CCDs arranged along a window direction are arranged in multiple stages in a line direction orthogonal to the window direction. An object recognition device comprising a multi-stage line type CCD 11 for capturing an image as luminance information, and recognizing a specific object from two-dimensional distance data obtained based on a luminance signal from the line type CCD 11 Is the target.

The CCD line at one end of the multi-stage line type CCD 11 in the line direction is a short distance side line for measuring a short distance side, while the CCD line at the other end is distance measuring at a long distance side. The far side line.

A distance measuring means 16 for dividing an image obtained by the multi-stage line type CCD 11 into a plurality of parts for each CCD line and in a window direction to measure a distance in each area, Line distance calculating means 26 for calculating the distance for each line based on the distances obtained for the respective areas, and object recognizing means 20 for recognizing an object based on the distance for each line calculated by the line distance calculating means 26. A dead zone setting unit 3 for setting, for each CCD line, a dead zone of a distance for each line which is not used for object recognition by the object recognition unit 20;
4 is provided.

With the above configuration, first, in the distance measuring means 16, the image of the multi-stage line type CCD 11 is divided into a plurality of lines for each line and in the window direction, and the distance is measured for each region. Next, in the line distance calculating means 26, the distance for each line is calculated based on the distance for each area measured by the distance measuring means 16, and the object recognizing means 20 is calculated.
In, the object is recognized from the distance for each line calculated by the line distance calculating means.

Then, a dead zone of the distance is set for each CCD line by the dead zone setting means 34, and of the distances for each line calculated by the line distance calculating means 26,
The distance belonging to the dead zone is not used for the object recognition by the object recognition means 20. For this reason, for example, by arranging a multi-stage line CCD so as to recognize a vehicle or the like as an object obliquely behind the vehicle, and setting the dead zone to include an object such as a vehicle immediately behind the vehicle,
The object is recognized based on the distance data excluding the distance data relating to the object immediately behind the object. As a result, the object immediately behind is not erroneously recognized as an obliquely rearward object, and an object such as a vehicle that may be overtaking and that is traveling diagonally behind and in another lane is accurately recognized. can do. In addition, since the dead zone is set so as to include the object immediately behind the vehicle, even if the vehicle is behind the vehicle, it is possible to accurately recognize a distant object that is not directly behind the vehicle and does not overlook the object. .

In the invention of claim 2, the dead zone setting means 34 is configured to set a range in which the distance on the long distance side line is equal to or less than a predetermined value as a dead zone. In the present invention, similarly to the first aspect of the present invention, when a multi-stage line CCD is disposed on a vehicle so as to recognize the vehicle or the like as an object obliquely behind the vehicle, the distant object behind the vehicle is not overlooked. By preventing erroneous recognition of the object immediately behind the object obliquely behind the object, it is possible to accurately recognize an object such as a vehicle that may pass on another lane.

According to the third aspect of the invention, similarly to the first aspect of the invention, a large number of Cs arranged along the window direction are provided.
A multi-stage CCD is provided in which CCD lines composed of CDs are arranged in multiple stages in a line direction orthogonal to the window direction to capture an image as luminance information, and obtained based on a luminance signal from the multi-line CCD. As an object recognizing device for recognizing a specific object from two-dimensional distance data, a CCD line at one end in a line direction of a multi-stage line type CCD is set as a short distance side line, and a CCD at the other end is used.
The line is a long distance side line, and the distance measuring means 1
6. A line distance calculation means 26 and an object recognition means 20 are provided.

The body weight center direction detecting means 3 for obtaining the center of gravity direction of the object recognized by the object recognition means 20
5, an object distance detecting means 36 for obtaining a distance to the object recognized by the object recognizing means 20, and a dead zone of a distance for each line not used for object recognition by the object recognizing means 20, And a dead zone setting means for setting the object distance and the object distance determined by the means and the object distance detecting means, respectively.

According to the present invention, the distance is measured for each divided area of the multi-stage line type CCD 11 by the distance measuring means 16, and the distance for each line is calculated by the line distance calculating means 26 based on the distance for each area.
The object is recognized by the object recognition means 20 from the distance for each line.

Further, the direction of the center of gravity of the object to be recognized by the object recognizing means 20 is obtained by the object body weight direction detecting means 35, the distance to the recognized object is obtained by the object distance detecting means 36, and the dead zone setting means 34 The dead zone of the distance is set according to the direction of the center of gravity of the object and the object distance. Of the distances for each line calculated by the line distance calculating means 26, the distance belonging to this dead zone is determined by the object recognition means 20. Not used for Therefore, the present invention can provide the same functions and effects as those of the first aspect.

In this case, in the invention of claim 4, the dead zone setting means 34 is configured to make the dead zone variable in accordance with the number of lines included in the recognition object. The means is configured to reduce the dead zone as the number of lines included in the recognition object decreases.

As a result, the dead zone changes according to the number of lines included in the recognition object, and is set smaller as the number of lines decreases. Therefore, for example, the dead zone can be changed according to the size of the width of the object directly behind the vehicle, and when the object immediately behind the vehicle is a wide vehicle, the number of lines included in the recognition object increases and the dead zone is increased. On the other hand, when the object immediately behind is a motorcycle having a small width, the number of lines included in the recognized object is reduced and the dead zone is reduced. That is, an object as a two-wheeled vehicle has a high possibility that its distance data is used for object recognition due to a decrease in the dead zone, and accurately identifies the two-wheeled vehicle as a vehicle and has a high possibility of overtaking. The wheeled vehicle can be easily recognized as an object.

On the other hand, in the invention of claim 6, the dead zone setting means 34 is configured to make the dead zone variable in accordance with the relative speed with respect to the recognition object. In the invention of claim 7, the dead zone setting means 34 The dead zone is reduced as the relative speed with respect to the recognition object increases.

In this case, the dead zone changes according to the relative speed with respect to the recognition object, and is set smaller as the relative speed increases. Therefore, for example, the dead zone can be changed according to the relative speed with the object directly behind the vehicle, and the dead zone becomes large when the relative speed with the object directly behind the vehicle is low, and becomes small when the relative speed is high. . In other words, when the relative speed with the object directly behind is high, if the object with the relative speed is a vehicle, it is considered that the distance between the vehicles is such that it can collide with the own vehicle, and the object is determined as a two-wheeled vehicle. be able to. Therefore, according to the present invention, the two-wheeled vehicle can be easily recognized as an object by accurately identifying the two-wheeled vehicle.

According to the eighth aspect of the present invention, the object recognizing means 2
0 is configured to output a signal such as an alarm based on the recognition result of the object. Thus, the recognition object can be easily known.

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION

(Embodiment 1) FIG. 2 shows a vehicle C (automobile) equipped with an object recognition device according to Embodiment 1 of the present invention. The object O (shown in FIGS. 5, 11, and 12) is recognized.

In FIG. 2, reference numeral 1 denotes a vehicle body of the vehicle C, 2 denotes a vehicle room formed at a substantially central portion in the front and rear direction of the vehicle body 3, 3 denotes an engine room formed at a front end of the vehicle body 1, and 4 denotes a vehicle room. An instrument panel 5 at the front end is a package tray at the rear end of the vehicle compartment 2, and 6 is a rear window glass. Then, as shown in FIG. 3, the object recognition device receives the left and right rear side detection sensors 10 and 10 for measuring the distance to the object O, and the output signals of the respective detection sensors 10. Controller 15 which receives a signal from the controller 15
The display device 31 includes a display device 31 for displaying by RT, liquid crystal, or the like, and an alarm device 32 for alarming the degree of danger of the object O. Then, as shown in FIG.
Are attached and fixed to both left and right end portions of the package tray 5 in a state of facing obliquely rearward, respectively. The controller 15 is provided at the rear end of the engine room 3, and the display device 31 and the alarm device 32 are provided at the instrument panel 4.

As shown in FIG. 5, each of the detection sensors 10
Is a pair of upper and lower Cs arranged in a vertical direction at a predetermined distance.
The image pickup apparatus includes CD chips 11, 11 and lenses 12, 12 arranged corresponding to the CCD chips 11, 11.
Each of the CCD chips 11 is composed of a multi-stage line type CCD in which CCD lines composed of a large number of CCDs arranged in a vertical window direction are arranged in multiple stages in a line direction (horizontal direction) orthogonal to the window direction. Each of the CCD chips 11 passes through the lens 12 through the rear window glass 6 of the vehicle C through the lens 12 and has a range of an angle θ1 in the vertical direction and a range of an angle θ2 in the horizontal and horizontal directions (FIGS. 10 and 12).
) Is captured as luminance information.

As shown in FIG. 4, each of the detection sensors 10 is connected to a distance measuring circuit 16 (distance measuring means) in a controller 15. Each of the distance measuring circuits 16 is connected to both CCs.
A parallax calculating unit 17 that calculates the parallax (phase difference) of the object image at the D chips 11 and 11 and a distance calculating unit 18 that calculates a distance to the object O based on a signal from the parallax calculating unit 17 are provided. . Then, as shown in FIGS. 6 and 7, each distance measuring circuit 16 converts the image captured by each CCD chip 11 into n lines for each CCD line in the line direction (horizontal direction).
While dividing each line into m windows in the window direction (vertical direction)
Approximately the entire image is composed of m × n areas E, E,..., And the parallax between the same areas E, E in the images obtained by both CCD chips 11 is obtained. To calculate the distance to the object O.

That is, the images captured by both CCD chips 11, 11 are as shown in FIG. 6, but the images of both CCD chips 11, 11 are at the same line position (line i in the illustrated example). As shown in FIG. 8, the two CCD chips 11 and 11 are displaced in the vertical direction to generate parallax, and the distance to the object O is measured using the parallax. This principle will be described with reference to FIG. 9 by referring to the triangles P.O1.Q and O1.P1 in FIG.
・ Q1 and triangles P · O2 · Q and triangle O2
Since P2 and Q2 have similar shapes, the distance from the detection sensor 10 (lens 12) to the object O is a,
The distance between the centers of both lenses 12 and 12 is B (constant), the focal length of lens 12 is f (constant),
The displacement amount of the object image from the lens center at 11 is x
Assuming that 1, x2, a · x1 / f = Ba−x2 / f, and from this equation, a = B · f / (x1 + x2) is obtained. That is, the distance a to the object O can be measured by the parallax (phase difference) of the object image between the two CCD chips 11, 11.

G (white point) in FIGS. 6 and 7 is
The distance measuring points (ranging points) are arranged in a vertical and horizontal lattice so as to correspond to the CCD of the CCD chip 11, and the distance measuring points G are included in each area E. The window of each CCD line is divided so that a part of the window overlaps with the adjacent window, and the same distance measuring points G, G, G, and G are located in regions E and E adjacent in the vertical direction (window direction). …It is included. O 'is an image of the object.

Further, as shown in FIG.
A plurality of lines formed by dividing the image captured by the chip 11 line by line are arranged such that the line position for measuring a short distance outside the vehicle C has a smaller number, while the center position in the vehicle width direction has a smaller number. The line position for measuring a long distance is set to a large number, and the number is sequentially increased from the outer line toward the center line in the vehicle width direction.

As shown in FIG. 4, a specific object is provided to the controller 15 based on two-dimensional distance data in the vertical and horizontal directions obtained based on the sensor 10, that is, a signal from each distance measuring circuit 16. An object recognition unit 20 that recognizes O;
An object selecting section 21 for selecting whether or not the object O is a new object based on an output signal of the object recognizing section 20;
And a danger determining unit 22 for determining whether the object O selected in Step 1 is a dangerous object for the vehicle C (own vehicle). The object recognizing unit 20 displays a display signal based on the recognition result of the object O. Is output to the display device 31 and an alarm signal is output to the alarm device 32 through the object selection unit 21.

In addition, the controller 15 includes a range cut section 24 for eliminating an unnecessary range where the object O cannot be originally located in recognizing the object O, a distance data for each measured area, and a surrounding area. An eight-neighbor point processing unit 25 as effective point number giving means for comparing the eight adjacent areas (eight-neighbor point processing) and giving an effective point number, and a line distance calculating unit 26 for calculating the distance for each line , A guardrail determination unit 27 for determining a guardrail, and a grouping unit 28 for grouping distance data for each object O.

Here, both CCs are
FIG. 1 shows a processing operation from distance data for each area E of an image captured by D chips 11 to object recognition.
5 will be described along the flowchart.

First, in step S1, the effective range (range cut range) of the distance data is set by the range cut unit 24.
Set. That is, FIG.
4 shows a range cut range Z1 in the vertical direction, and FIG. 12 shows a range cut range Z2 in the horizontal direction. These range cut ranges Z1 and Z2 are shown in FIG.
Is detected based on the angle of the line and the distance at that position. In FIG. 12, F is the road surface of the vehicle C, M is a white line on the road surface F for setting the vehicle traveling lane on the road, F1 is a road side zone installed on both sides of the road, and H is an implant.

Next, in step S2, distance data d (i, j) for each area E (i, j) divided into the number n of lines and the number m of windows is read, and in step S3, the distance data d (i) is read. , J) are within the valid range. When this determination is YES for “being within the effective range”, the distance data d (i, j) is registered as valid in step S4, while NO
In step S5, the distance data d
Clear (i, j) as invalid data. After these steps S4 and S5, the process proceeds to step S6, where the line number i (1 to n) and the window number j (1 to m)
Is determined for each area E, and steps S3 to S6 are repeated until all areas E, E,. If the determination in step S6 is YES, the process proceeds to step S7, where a representative distance l (i) (line distance data) for each line is obtained from the distance data d (i, j).
That is, in this step S7, the eight adjacent points processing unit 25 assigns an effective point number to the distance data of each measured area E by the eight adjacent points processing unit 25, and the line distance calculating unit 26 executes the eight adjacent points processing routine. Is calculated.

As shown in FIG. 13, the eight adjacent point processing by the eight adjacent point processing unit 25 is performed on the distance data of a certain area E (i, j) and the eight adjacent areas R1 surrounding the distance data.
This is to judge the correlation between the distance data of R8 to R8 and is specifically performed as shown in FIG. That is, in the eight adjacent point processing routine, in the first step S101, the number of lines n
And the distance data d (i, j) for each area E (i, j) divided into the number m of windows are read, and the next step S1
02, the number of effective points P (i, j) of each area E (i, j)
j) is initialized to P (i, j) = 0. The number of effective points P (i, j) is set for each area E (i, j). The larger this value is, the higher the validity of the distance data of the area is, and it is determined that the area data has reliability and credibility. Is done. In the next step S103, the number of effective points to the area (lattice point) located at the left and right ends and the upper and lower ends of all the areas is increased, and the effective point number P (i, j) is increased by +1 to the surrounding area.
And the number of effective points P (i, j) is increased by +2 in the four corner areas. Thereafter, in step S104, it is determined whether or not to perform the adjacent point processing. If this determination is NO,
In step S111, the number of valid points P (i, j)
Is set to P (i, j) = 8, and then the process proceeds to step S112. If the determination is YES, the process proceeds to step S105.
To set a distance threshold value d0 for determining the number p of assigned points.

After step S105, step S106
The distance data d (Ri) of the adjacent region Ri is read, and in the next step S107, the region E and the adjacent region R are read.
Distance difference dx = | d (i, j) −d (Ri) from 1 to R8
Is calculated. Thereafter, in step S108, the magnitude difference between the distance difference dx and the distance threshold value d0 is determined, and if this determination is NO where dx ≧ d0, the process proceeds to step S108.
On the other hand, when it is determined that dx <d0 (YES), the number p of points to be provided is set in step S109.

After such step S109, in step S110, the number of valid points P
(I, j) is added with the number p of points to be given, and a new number P (i, j) of effective points = P (i, j) + p is set.
The process proceeds to step S112. In this step S112, it is determined whether or not the processing of steps S106 to S110 has been completed for each of the eight adjacent regions R1 to R8. If this determination is NO, the process returns to step S106, and the same applies to the other remaining regions. Is performed. On the other hand, when the determination is YES, the process proceeds to step S113,
Number of effective points P for all areas E, E, ...
It is determined whether the setting of (i, j) (the processing of steps S106 to S110) has been completed. If this determination is NO, the process returns to step S104, and the setting of the number of effective points P (i, j) is repeated for another area E. on the other hand,
If the determination is YES, the process proceeds to the next distance calculation processing for each line (see FIG. 17).

On the other hand, FIG.
And calculates the distance for each line based on the number of effective points P (i, j) assigned by the eight adjacent point processing unit 25 (effective point number assigning means). In the line distance calculation routine, first, in step S201, distance data d (i, j) for each area E divided into the number n of lines and the number m of windows is read, and the area E given by the above-described eight adjacent point processing is read.
The number of valid points P (i, j) is read for each line, and in the next step S202, the number of line representative valid points PI
(I) is initialized to PI (i) = 0. The line representative effective point number PI (i) is set for a line at the time of distance calculation for each line, and the larger this value is, the higher the validity of the distance data of the line is, and the more reliable and reliable it is. Is determined.

In the next step S203, a line representative threshold value P0 corresponding to the line representative effective point number PI (i) is set. After step S203, the process proceeds to step S204, where the number of effective points P for each of the above areas is set.
It is determined whether (i, j) is greater than the line representative threshold value P0. This determination is NO for P (i, j) ≦ P0
, The process proceeds directly to step S206, but if the determination is P (i, j)> P0,
At 205, the representative distance l (i) for each line is updated for averaging, and the number of valid points P (i, j) for each area is added to the line representative valid point number PI (i).
, And the line representative effective point number PI (i) is updated, and then the process proceeds to step S206. That is, the line distance calculation unit 26 performs the distance calculation for each line in an area where the number of effective points P (i, j) given by the eight adjacent point processing unit 25 is larger than the line representative threshold value P0. ing.

The updating of the representative distance l (i) for each line is performed by the following equation. l (i) = [l (i) × PI (i) + d (i, j) ×
{P (i, j) -PO + 1}] {PI (i) + P
(I, j) -PO + 1}

In step S206, it is determined whether or not the processing has been completed for all the window numbers (area E) of the line, and steps S203 to S205 are repeated for each area E of the line until the determination becomes YES. If the determination in step S206 is YES, step S2
07, it is determined whether or not the processing has been completed for all the line numbers. Step S2 is performed until the determination becomes YES.
02 to S206 are repeated. If the determination in step S207 is YES, the process proceeds to the next object recognition process (see FIG. 19).

It should be noted that the above-described distance data d (i, j) for each area E is subjected to 8-neighboring point processing to give the number of effective points P (i, j), and thereafter, the representative distance for each line FIG. 14 shows a specific example of the process of calculating l (i), and FIG. 14A shows distance data d (i, j) for each region.
And FIG. 14B shows the number of effective points P for each area.
(I, j), and FIG. 14 (c) shows a representative distance l (i) for each line.

Referring again to FIG. 15, after calculating the representative distance l (i) for each line in step S7, the process proceeds to step S8, where the effective range for the representative distance l (i) for each line is set, and In step S9, a dead zone is set for each line. FIG. 18 shows a map in which the effective range and the dead zone are respectively set. The effective range is set so that the distance of the effective range becomes larger as the position of the line becomes larger and the line becomes longer. In this effective range, a range in which the distance on the long distance side line is equal to or less than a predetermined value is set as a dead zone. According to the characteristics shown in FIG. In the characteristic shown in FIG. 18B, the range of the long-distance line where the line position is equal to or greater than a predetermined value and the line representative distance l (i) is equal to or less than a predetermined value is defined as a dead zone.

In this embodiment, in step S9, the dead zone of the distance for each line not used for object recognition in the object recognition unit 20 is set for each CCD line, and the distance on the long distance side line is set to a predetermined value. A dead zone setting means 34 for setting the following range as a dead zone is configured.

After step S9, the process proceeds to step S10, where the line representative distance l (i) is input. At step S11, the line representative distance l (i) falls within the effective range set at step S8. Is determined. When this determination is YES for “being within the effective range”, the process proceeds to step S12, and this line representative distance l
It is determined whether or not (i) is within the dead zone set in step S9. When the determination in step S11 is NO for "not in the effective range" and when the determination in step S12 is NO for "in the dead zone", the process proceeds to step S13, and the line representative distance l (i) is entered. Are initialized as data values, and the process proceeds to step S14.

On the other hand, when it is determined as YES in the above-mentioned step S12, "No in the dead zone",
Proceed to 14. In this step S14, the above step S
It is determined whether or not the processing of 11 to S13 has been completed for all the lines i (1 ≦ i ≦ n), and if not completed, steps S11 to S14 are repeated. When all lines have been completed and the determination in step S14 is YES, the process proceeds to step S15, in which the object recognition unit 20 of the controller 15 recognizes the object O from the line representative distance l (i) and determines the object number k. Of the object distance L (k) (object distance data).

Specifically, FIG. 19 shows the processing operation of the object recognition unit 20. In the object recognition unit 20, based on the representative distance l (i) for each line calculated by the line distance calculation unit 26, To recognize the object O. That is, the object number k is set in step S301,
At 302, the object distance L (k) and the number of effective object points P
Both K (k) and the number of data in the object N (k) are set to 0, and the primary storage data set is reset.

In the next step S303, it is determined whether or not valid unregistered line data is registered. If this determination is NO, the flow advances to step S308. If the determination in step S303 is YES, step S30
At 4, the front / rear position XD (i) and the horizontal position YD (i) of the line data are set. Thereafter, step S305
It is determined whether or not the object distance L (k) has already been defined. If this determination is NO, the process proceeds to step S306, where the object distance L (k) is set to L (k) =
XD (i), and the number of effective object points PK (k) to P
K (k) = PI (i) and the number of data in the object N
After setting (k) to N (k) = 1 and setting a primary storage data set, the process proceeds to step S308.

On the other hand, the determination in step S305 is Y
In the case of ES, the process proceeds to step S307, where the object distance L
Let (k) be L (k) = ｛PK (k) × L (i) + P (i)
× XD (i)} / {PK (k) + P (i)} and the number of object effective points PK (k) as PK (k) = PK (k)
+ PI (i), and the number of data in the object N (k) by N
After setting (k) = N (k) +1 to update the primary storage data set, the process proceeds to step S308.

In step S308, it is determined whether or not the processing has been completed for all the line numbers.
Steps S303 to S307 are repeated until ES is reached. If the determination in step S308 is YES, a determination is made in steps S309 to S311 as to whether registration of the object O is possible. First, in step S309, it is determined whether the number N (k) of data in the object is equal to or greater than a predetermined value. Note that the predetermined value can be variably set so as to decrease as the distance increases. If the determination in step S309 is NO, it is considered that the distance data is due to noise or the like, and in step S310, the object O
Is not performed, the object data of the object number k is initialized, and the process ends. On the other hand, the determination in step S309 is YES
If, the process ends after the object O is registered in step S311. That is, the object recognition unit 20
Registers an object corresponding to the distance of each line as a new object only when the number N (k) of distance data per line calculated by the line distance calculation unit 26 is equal to or greater than a predetermined value.

After the processing operation in the object recognition unit 20,
A display process for displaying an object on the display device 31 and a warning process for a warning on the warning device 32 are performed.

Therefore, in this embodiment, when an image is captured as luminance information by the left and right rear side detection sensors 10, 10, first, each distance measuring circuit 1 of the controller 15
At 6, the image of each detection sensor 10 is divided in the line direction and the window direction, and the distance d (i, j) is measured for each region E. Next, the number of effective points of the distance data for each region E is calculated by the eight adjacent point processing unit 25 based on the measured distance d (i, j) for each region E and the difference dx between the distances of the adjacent regions R1 to R8. P (i, j) is added, and the line distance calculation unit 26 calculates the distance l (i) for each line based on the number of effective points P (i, j), and the object recognition unit 20 calculates the distance l (i). The object O is recognized from the distance l (i) for each line calculated by As described above, the validity of the distance data for each region E is determined as the number of valid points P (i, j) from the relationship with the adjacent regions R1 to R8, and based on the number of valid points P (i, j). Since the distance O (i) for each line is obtained and the object O is recognized based on the distance l (i), even if there is a variation or noise in the distance measurement data, it is possible to reduce the influence as much as possible. As a result, accurate distance calculation can be performed, and accurate object recognition can be performed.

In the 8-neighboring point processing unit 25, the distance difference dx between the area E and the adjacent areas R1 to R8 is equal to the threshold value d0.
When the number is smaller than the effective point number P (i, j) (the number of assigned points p), the area E
Only when the distance difference dx from the adjacent regions R1 to R8 is smaller than the threshold value d0 can be determined to be valid by providing the effective point number P (i, j), and the accuracy of the distance calculation can be improved.

The line distance calculating section 26 calculates the distance for each line in an area where the number of effective points P (i, j) is larger than the line representative threshold value P0. It is possible to calculate accurately.

Further, as shown in FIG. 19, the object recognizing unit 20 determines the distance for each line only when the number of data of the distance for each line calculated by the line distance calculating unit 26 is equal to or more than a predetermined value. Is registered as a new object O, it is possible to restrict the registration of the new object O, and it is possible to suppress noise or the like other than the object O from being erroneously registered as the object O.

Since the object recognizing section 20 is configured to output a signal such as an alarm based on the recognition result of the object O, the recognized object O can be easily known.

Further, in this embodiment, a dead zone for the line representative distance l (i) for each line calculated by the line distance calculating unit 26 is set for each CCD line, and this dead zone is located behind the vehicle C. The distance on the corresponding long distance side line is set to a range that is equal to or less than a predetermined value.
Then, of those line representative distances l (i), the distance belonging to the dead zone is not used for object recognition in the object recognition unit 20. For this reason, even if there is an object O such as a vehicle directly behind the vehicle C, that is, at a distance equal to or less than the predetermined value,
The object O belongs to the dead zone, and the object O is determined by the distance data excluding the distance data of the object O immediately behind.
Is recognized. As a result, the object O directly behind the vehicle C is not erroneously recognized as the obliquely rearward object O, and the vehicle O, which is obliquely behind the vehicle C and is running on another lane, may be overtaken. The object O can be accurately recognized.

Further, since the dead zone is set so as to include the object O right behind the vehicle C, the object O located not behind and immediately behind the vehicle C can be accurately recognized. There is no oversight.

(Embodiment 2) FIGS. 20 to 23 show Embodiment 2 of the present invention (note that the same parts as those in FIG. 15 are denoted by the same reference numerals and detailed description thereof is omitted). In 1, the effective range and the dead zone of the line representative distance l (i) are set for each line. On the other hand, the dead zone is set according to the direction of the center of gravity of the object O and the distance L (k). The variable is made according to the number of lines included in the object O.

That is, FIG. 20 shows the processing operation from the distance data for each area E of the image to the object recognition in the controller 15, and steps T1 to T7 are the same as those in the first embodiment.
Steps S1 to S7 and step T8 are the same as step S15, respectively (see FIG. 15).

Then, after detecting the detection distance L (k) (object distance data) of the object k in step T8, in step T9
To set an effective range for the object distance, and set a dead zone right behind the vehicle C in the next step T10. Specifically, FIG. 21 shows the direction of the center of gravity of the object and the object distance L.
(K) indicates an effective range and a dead zone set in accordance with (k),
The effective range is set so that the distance of the effective range increases as the direction of the object center of gravity ObjCenter becomes farther. And, in this effective range, the object weight center Ob
A range in which jCenter is on the long distance side and the object distance L (k) is equal to or less than a predetermined value is defined as a dead zone. In the characteristic shown in FIG. 21A, the dead zone is determined as the direction of the center of gravity of the object moves toward the long distance side. 21B, on the other hand, in the characteristic shown in FIG. 21B, a range in which the direction of the object center of gravity ObjCenter is on the far side beyond the predetermined direction and the object distance L (k) is equal to or smaller than the predetermined range is a dead zone.

After step T10, step T11 is executed.
At the center of gravity ObjectCenter (k) of the object number k
(The direction) is input, and then the process proceeds to step T12, where the process proceeds to a processing routine when the object O is located immediately behind the vehicle C. In this processing routine, as shown in FIG. 22, in the first step T101, the object distance L (k) and the object center of gravity O obtained in steps T8 and T11, respectively.
It is determined whether or not the direction of bjCenter is within the effective range set in step T9. If the determination is NO for “not within effective range”, the process proceeds to step T102, where the object distance L (k) is initialized and canceled, and then the process proceeds to step T109.

On the other hand, if the determination in step T101 is "YES in the effective range", the process proceeds to step T10.
Then, it is determined whether or not the object distance L (k) and the direction of the physical center of gravity are within the dead zone set in step T10. If this determination is YES for "none in dead zone", the process proceeds to step T108,
After registering the recognition object O as an object, the process proceeds to step T109.

If it is determined as NO in the dead zone in step T103, the process proceeds to step T104,
Number of lines ObjLine included in the recognition object O
(K) is input, and in the next step T105, a threshold value ObjLimit of the line number ObjLine (k) is set. FIG. 23 shows the characteristics of the threshold value ObjLimit.
The threshold value ObjLimit is set to decrease proportionally (FIG. 23A) or stepwise (FIG. 23B) toward the far side.

After step T105, step T1
06, the number of lines Obj included in the recognition object O
Line (k) is compared with its threshold value ObjLimit, where ObjLine (k) ≧ ObjLi
When mit is NO, the process proceeds to step T102 to initialize the object distance L (k) and the dead zone threshold, and then proceeds to step T109, while objLine
(K) If <ObjLimit is YES, the dead zone (the threshold value) of the line representative distance set in step T10 is reduced in step T107, and the process proceeds to step T108 to register an object. Then, the process proceeds to step T109.

In step T109, the processing in steps T101 to T108 is repeated for all object numbers k (1 ≦ k
It is determined whether or not the processing has been completed for ≦ R). If not completed, steps T101 to T108 are repeated, whereas if the determination in step T109 is YES, the processing is ended.

In this embodiment, the object distance detecting means 36 for obtaining the distance to the object O recognized by the object recognizing unit 20 is constructed by the step T8.

In step T11, the body weight center direction detecting means 35 for obtaining the center of gravity direction of the object O recognized by the object recognition unit 20 is configured.

Further, steps T10, T104 to T1
07, the dead zone setting means 34 for setting the dead zone of the distance for each line not used for object recognition in the object recognition unit 20 according to the direction of the center of gravity of the object O and the object distance L (k) to the object O. The dead zone setting means 34
The dead zone is made variable according to the number of lines ObjLine included in the recognition object O, and the smaller the number of lines ObjLine is, the smaller the dead zone is set.

In step T102, when the object distance L (k) is out of the effective range, the dead zone is initialized.

Therefore, in this embodiment, when the object O is recognized by the object recognition unit 20, the recognized object O
And the distance L (k) to the recognition object O are obtained, and the dead zone of the line representative distance l (i) is set according to the direction of the center of gravity of the object O and the object distance L (k). Of the line representative distances l (i) calculated by the line distance calculation unit 26, the distances belonging to this dead zone are not used for object recognition in the object recognition unit 20.
Therefore, in this embodiment, the same operation and effect as those of the first embodiment can be obtained.

The dead zone is variable according to the number of lines ObjLine included in the recognition object O, and the smaller the number of lines ObjLine included in the recognition object O, the smaller the dead zone. Therefore, for example, the dead zone can be changed in accordance with the width of the object O immediately behind the vehicle C. When the object O immediately behind the vehicle C is a wide vehicle, the number of lines ObjLine included in the recognized object O
Increases, and the dead zone increases, whereas when the object O directly behind is a small-sized two-wheeled vehicle, the number of lines ObjLine included in the recognition object O decreases and the dead zone decreases. As a result, there is a high possibility that the distance data of the object O such as a motorcycle is used for object recognition due to a decrease in the dead zone. A tall two-wheeled vehicle can be easily recognized as the object O.

(Embodiment 3) FIGS. 24 and 25 show Embodiment 3 in which the change of the dead zone in Embodiment 2 is described.
This is performed in consideration of not only the number of lines ObjLine included in the object O but also the relative speed with respect to the object O.

That is, in this embodiment, as shown in FIG. 24 (the same steps as those in FIG. 22 are denoted by the same reference numerals and detailed description thereof is omitted), and
01 to T109 correspond to steps T101 to T101 of the second embodiment.
It is the same as T109 (see FIG. 22).

Then, in step T106, ObjLin
If it is determined that e (k) <ObjLimit is YES,
Proceeding to step T110, the relative speed Objv (k) of the recognition object O with respect to the own vehicle C is input, and the next step T11 is performed.
The threshold value ObjvL of the relative speed Objv (k) at 1
Set the limit. This threshold ObjvLimit
Are set as shown in FIG. 25, and the threshold value ObjvLimit is set to increase proportionally (FIG. 25A) or stepwise (FIG. 25B) as the distance increases. Is done.

After step T111, step T1
At 12, the relative velocity Objv with respect to the recognition object O
(K) and the threshold value ObjvLimit are compared, and Nj of Objv (k) ≦ ObjvLimit is obtained.
In the case of O, the process proceeds to step T102 and the object distance L
(K) is initialized. On the other hand, if the determination in step T112 is YES, that is, Objv (k)> ObjvLimit, in step T107, the dead zone (the threshold) of the line representative distance is changed to be small, and the process proceeds to step T108 to register the object. .

In this embodiment, steps T10,
The dead zone setting unit 34 is configured by T104 to T112, and the dead zone setting unit 34 changes the dead zone in accordance with the relative speed Objv with the recognition object O, and sets the dead zone to be smaller as the relative speed Objv is larger. I have.

Therefore, in this embodiment, the dead zone of the line representative distance is set smaller as the relative speed Objv with respect to the recognition object O increases, so that, for example, according to the relative speed Objv with the object O directly behind the vehicle C. The dead zone can be changed. When the relative speed Objv is low, the dead zone becomes large, whereas when the relative speed Objv is high, the dead zone becomes small. In other words, the object O
Is higher, the relative speed Obj
If the object O having v is a vehicle, it is considered that the inter-vehicle distance is such that the vehicle O collides with the own vehicle C, so that the object O can be determined as a two-wheeled vehicle. Therefore, according to the present invention, the two-wheeled vehicle can be easily recognized as the object O by accurately identifying the two-wheeled vehicle.

In the above-described embodiment, the object O is recognized from the side or obliquely behind the vehicle C. However, the present invention can be applied to the case of recognizing an object outside these ranges. The present invention can also be applied to a case where an object is recognized inside a building or outdoors, in addition to the vehicle C.

[0082]

As described above, according to the first aspect of the present invention, the image of the multi-stage line type CCD is divided into a plurality of lines for each line and in the window direction, and the distance is measured for each region. An object is recognized by calculating a distance for each line from the distance data, and a dead zone of a distance for each line that is not used for object recognition is set for each CCD line. Further, in the invention of claim 2, a range in which the distance on the long distance side line is equal to or less than a predetermined value is set as the dead zone. Further, in the third aspect of the present invention, similarly to the first aspect of the present invention, the image of the multi-stage line type CCD is divided into a plurality of lines and a plurality of lines in the window direction, and the distance is measured for each area. An object is recognized by calculating a distance for each line from the distance data, and a dead zone of a distance for each line that is not used for object recognition is set according to the direction of the center of gravity of the recognized object and the distance to the object.

Therefore, according to these inventions, for example, a multi-stage line CCD is arranged in a vehicle so as to recognize an object such as a vehicle obliquely behind the vehicle, and the dead zone is set so as to include an object such as a vehicle immediately behind the vehicle. Therefore, it is possible to prevent the vehicle immediately behind the vehicle from being erroneously recognized as a vehicle obliquely behind, and to avoid overlooking a distant vehicle behind the vehicle, and obstruct the vehicle on another lane diagonally behind the vehicle. It is possible to accurately recognize a running vehicle or the like that may pass.

In the invention of claim 4, the dead zone is made variable according to the number of lines included in the recognition object. In the invention of claim 5, the dead zone is made smaller as the number of lines is smaller. According to these inventions, for example, the dead zone can be changed according to the size of the width of the object directly behind the vehicle, the two-wheeled vehicle can be accurately distinguished from the two-wheeled vehicle, and there is a high possibility of overtaking. A car can be easily recognized as an object.

According to a sixth aspect of the present invention, the dead zone is made variable in accordance with the relative speed with respect to the recognition object. In the seventh aspect of the present invention, the dead zone is reduced as the relative speed increases, for example, just behind the vehicle. The dead zone can be changed according to the relative speed with respect to the object, and the two-wheeled vehicle can be accurately identified and the two-wheeled vehicle can be easily recognized as the object.

According to the eighth aspect of the present invention, a signal such as a warning is output based on the recognition result of the object, so that the recognized object can be easily known.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of the present invention.

FIG. 2 is a perspective view showing positions of respective components of the object recognition device according to the first embodiment of the present invention in a vehicle.

FIG. 3 is a block diagram illustrating a schematic configuration of an object recognition device.

FIG. 4 is a block diagram illustrating a detailed configuration of the object recognition device.

FIG. 5 is a side view showing a concept of measuring a distance to an object by a detection sensor.

FIG. 6 is a diagram showing an image captured by a CCD chip.

FIG. 7 is a diagram showing a concept of dividing a line in an image captured by a CCD chip in a window direction to divide an area.

FIG. 8 is an explanatory diagram showing a state in which images obtained by upper and lower CCD chips are shifted on the same line and parallax occurs.

FIG. 9 is a diagram showing a principle of measuring a distance to an object by upper and lower CCD chips.

FIG. 10 shows C in an image obtained by a CCD chip.
It is a top view which shows the ranging direction of a CD line.

FIG. 11 is a side view showing a range cut area in a vertical direction.

FIG. 12 is a plan view showing a horizontal range cut area.

FIG. 13 is a diagram showing an arrangement of eight adjacent regions adjacent to the region.

FIG. 14 is a diagram showing a specific example from the processing of eight adjacent points to the calculation of the distance for each line.

FIG. 15 is a flowchart illustrating a distance data processing operation performed by the controller according to the first embodiment.

FIG. 16 is a flowchart illustrating an eight adjacent point processing operation.

FIG. 17 is a flowchart illustrating a distance calculation processing operation for each line.

FIG. 18 is a characteristic diagram of a map showing an effective range of a line representative distance and a dead zone according to a line position.

FIG. 19 is a flowchart showing an object recognition processing operation.

FIG. 20 is a flowchart illustrating a distance data processing operation performed by the controller according to the second embodiment of the present invention.

FIG. 21 is a characteristic diagram showing an effective range and a dead zone according to the direction of the center of gravity and the distance of an object.

FIG. 22 is a flowchart illustrating a processing routine when an object is located immediately behind in the second embodiment.

FIG. 23 is a characteristic diagram for setting a threshold value of the number of lines included in an object.

FIG. 24 is a diagram corresponding to FIG. 22 showing the third embodiment.

FIG. 25 is a characteristic diagram for setting a threshold value of a relative speed with respect to an object.

[Explanation of symbols]

 C vehicle 10 rear side detection sensor 11 CCD chip (multi-stage line type CCD) 15 controller 16 distance measuring circuit (distance measuring means) 20 object recognition unit (object recognition means) 21 object identification unit 25 8 adjacent point processing unit (effective point) Numbering means) 26 Line distance calculating unit (Line distance calculating means) 31 Display device 32 Alarm device 34 Dead zone setting means 35 Body weight center direction detecting means 36 Object distance detecting means E, E (i, j) areas R1 to R8 Adjacent areas d (i, j) measured distance dx distance difference from adjacent area P (i, j) number of effective points P0 line representative threshold l (i, j) line representative distance L (k) object distance ObjCenter (k) object Center of gravity ObjLine (k) Number of lines included in object Objv (k) Relative speed with object O Object O 'Object image

──────────────────────────────────────────────────の Continuation of the front page (51) Int.Cl. ⁶ Identification code FI G06T 7/60 G06F 15/64 M 15/70 350J (72) Inventor Toru Yoshioka 3-1, Fuchu-cho, Shinchu, Aki-gun, Hiroshima Mazda Inside (72) Inventor Hiroki Uemura 3-1, Shinchi, Fuchu-cho, Aki-gun, Hiroshima Mazda

Claims (8)

[Claims] 1. A multi-stage line type CCD comprising a plurality of CCD lines arranged along a window direction and arranged in multiple stages in a line direction orthogonal to the window direction to capture an image as luminance information, An object recognizing device for recognizing a specific object from two-dimensional distance data obtained based on a luminance signal from said multi-stage line type CCD, comprising:
The line is a short distance side line for measuring the short distance side, while the CCD line at the other end is a long distance side line for measuring the long distance side. The above CC
Distance measuring means for measuring a distance for each area by dividing the area for each D line into a plurality of pieces in the window direction, and a line for calculating the distance for each line based on the distance for each area measured by the distance measuring means Distance calculating means, object recognizing means for recognizing an object based on the distance for each line calculated by the line distance calculating means, and a dead zone for each line not used for object recognition by the object recognizing means. An object recognizing device, comprising: a dead zone setting means for setting each CCD line. 2. The object recognition apparatus according to claim 1, wherein the dead zone setting means is configured to set a range in which the distance on the long distance side line is equal to or less than a predetermined value as a dead zone. Recognition device. 3. A multi-stage line CCD in which CCD lines composed of a large number of CCDs arranged along a window direction are arranged in multiple stages in a line column direction orthogonal to the window direction to capture an image as luminance information. An object recognizing device for recognizing a specific object from two-dimensional distance data obtained based on a luminance signal from the multi-stage CCD, wherein one end of the multi-stage CCD in a line direction is provided. CC
The D line is a short distance line for measuring the short distance side, while the CCD line on the other end is a long distance line for measuring the long distance side. Image is CC
Distance measuring means for measuring a distance for each area by dividing the area for each D line into a plurality of pieces in the window direction, and a line for calculating the distance for each line based on the distance for each area measured by the distance measuring means Distance calculating means, object recognizing means for recognizing an object based on the distance for each line calculated by the line distance calculating means, and a body weight center direction detecting means for obtaining the center of gravity of the object recognized by the object recognizing means An object distance detecting means for obtaining a distance to an object recognized by the object recognizing means; a dead zone for each line not used for object recognition by the object recognizing means; A dead zone setting unit configured to set in accordance with the direction of the body weight and the object distance respectively obtained by the detecting unit; Recognition device. 4. The object recognition apparatus according to claim 3, wherein the dead zone setting means is configured to change the dead zone in accordance with the number of lines included in the recognition object. 5. The object recognition apparatus according to claim 4, wherein the dead zone setting means is configured to make the dead zone smaller as the number of lines included in the recognition object is smaller. 6. The object recognition apparatus according to claim 3, wherein the dead zone setting means is configured to change the dead zone according to a relative speed with respect to the recognized object. Recognition device. 7. The object recognition apparatus according to claim 6, wherein the dead zone setting means is configured to reduce the dead zone as the relative speed with respect to the recognition object increases. 8. The object recognition apparatus according to claim 1, wherein the object recognition means is configured to output a signal such as a warning based on a recognition result of the object. Recognition device.