JP3211732B2 - Distance measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a distance measuring device for determining the distance to an object such as a preceding vehicle by using a photodetector having an optical sensor array.
The present invention relates to a distance measuring device having a function of detecting a failure based on an operation failure due to fog or the like and a failure based on a defective optical sensor element in an optical sensor array.

[0002] In the drawings, the same reference numerals indicate the same or corresponding parts.

[0003]

2. Description of the Related Art First, a prior art of an inter-vehicle distance measuring device for measuring a distance from a preceding vehicle will be described. 2. Description of the Related Art As a conventional inter-vehicle distance measuring apparatus, an apparatus that electrically compares images formed by two right and left optical systems and measures a distance by triangulation is known. FIG. 16 is a diagram showing this type of conventional inter-vehicle distance measuring apparatus. In FIG. 1, the imaging lenses 1 and 2 are arranged at an optical axis interval B. Optical sensor array 3
Reference numerals A and 4A denote, for example, CCD linear sensor arrays, which are arranged at focal lengths f with respect to the imaging lenses 1 and 2, respectively. These optical sensor arrays 3A and 4A convert the images of the object 16A formed by the imaging lenses 1 and 2 into image signals 30A and 40A, respectively, and output them to the signal processing unit 5.

The signal processing unit 5 includes amplifiers 51 and 52, A / D
It comprises converters 53 and 54 and a storage device 55. The image signals 30A and 40A from the optical sensor arrays 3A and 4A are
, And is converted into digital data by A / D converters 53 and 54, and is converted into image data 31A and 41.
A is output to the storage device 55 as A. The distance detection circuit 6 provided on the output side of the signal processing unit 5 is configured by a microcomputer, compares the left and right image data 31A and 41A stored in the storage device 55, and compares the left and right image data 31A and 41A.
Is calculated and output to the outside as a distance signal 11.

Next, the principle of distance calculation will be described with reference to FIG.
I will tell. Horizontal axis with the midpoint of each imaging lens 1 and 2 as the origin O
X and the vertical axis Y are set, and the imaging position L₁, R₁The coordinates of
(-A_L1-B / 2, -f), (a_R1+ B / 2, -f) and
I do. Where a_L1, A_R1Is the optical sensor
This is the distance on the rays 3A and 4A. Center point of imaging lens 1
O_LIs (-B / 2,0), the center point of the imaging lens 2.
O_RIs (B / 2,0), and the point of the object 16A is
If the coordinates of M are (x, y), drop from point M to the X axis
The coordinates of the intersection N between the perpendicular and the X axis are (x, 0), and the point O_LOr
L of the perpendicular line dropped to the optical sensor array 3A₀Coordinates
Is (−B / 2, −f), point O_RTo light sensor array 4A
Position R of the perpendicular line₀Is (B / 2, -f)
is there. At this time, ΔMO_LN and ΔO_LL₁L₀, ΔMO
_RN and ΔO _RR₁R₀Are similar, so the following equation
1, 2 holds.

[0006]

(X + B / 2) f = (a _L1 + B / 2−B / 2) y

[0007]

(−x + B / 2) f = (a _R1 + B / 2−B / 2) y From Expressions 1 and 2, the following Expression 3 can be obtained. According to Equation 3, the distances a _L1 and a _{L with} respect to the imaging positions L ₁ and R ₁ are obtained.
_{If R1} is known, the distance y to the target 16A can be calculated.

[0008]

Y = B ・ f / (a _L1 + a _R1 ) Next, the operation of the distance detection circuit 6 will be described in detail. The distance detection circuit 6 compares the left and right image data 3AL and 4AR as shown by the solid line in FIG. 18 with respect to a separately set measurement window portion. If the images do not match, as shown by a broken line in FIG. Left image data 3AL to right, right image data 4
AR is sequentially shifted to the left, and the shift amount when the left and right image data match is detected.

The left and right image data 3AL, 4AR
The following evaluation function is used to determine the degree of coincidence. That is, the evaluation function is the left and right optical sensor arrays 3A and 4A.
Is the sum of the absolute values of the pixel data differences between the pixels (in this example, the CCD elements) located at the corresponding coordinates in the measurement window set for each pixel in the measurement window. The measurement windows are sequentially shifted, ie, the left measurement window is shifted to the left (therefore, the left image data 3AL is equivalently shifted to the right), and the right measurement window is shifted to the right (therefore, the right image). While the data 4AR is equivalently shifted to the left), the value of this evaluation function is examined, and when this function value is minimized, it is determined that the left and right image data match.

Since the distances a _L1 and a _{R1 with} respect to the left and right image forming positions L ₁ and R ₁ coincide with this shift amount, the distance detection circuit 6 sets the distance from the shift amounts a _L1 and a _R1 to the object 16A. The distance y is calculated by the above equation (3). FIG. 19 is a schematic diagram illustrating an image in a normal state in the detection of the inter-vehicle distance to the preceding vehicle 16. In the figure, a distance measurement range 23 is set in a measurement visual field 22, and a distance to an object in the distance measurement range 23, that is, a preceding vehicle 16 is detected as an inter-vehicle distance based on the above-described distance detection principle.

In the above-described conventional technique, the following inconvenience may occur. For example, as shown in FIG. 20, there are signs and pedestrian crossing patterns other than the preceding vehicle on the road, and the distance to these images appearing in the distance measurement range 23 is often measured. . Also, there are many signs and signs on the side of the road, and particularly when the road is curved, these can be seen in the center of the field of view of the sensor, so that it is easy to mistakenly identify the preceding vehicle. Therefore, as an inter-vehicle distance measuring device,
It is necessary to correctly recognize the preceding vehicle and to output only accurate distance information to the preceding vehicle.

In order to respond to this demand, there is an object recognition method based on image processing in which the edges of the image data are extracted and binarized using the grayscale image information of the CCD sensor, and then the area is divided to perform object recognition. . However, such image processing requires time for pre-processing such as removal of noise by filtering processing of a grayscale image up to binarization and extraction of an edge due to image difference and the like. However, even if processing is performed by using the method, an increase in cost and an increase in size of the apparatus are inevitable, and practical use is difficult.

As an example of a technique for solving this problem, there is a proposal of Japanese Patent Application Laid-Open No. Hei 8-210848 filed by the present applicant as a prior application (referred to as a first prior application for convenience). Next, the technical contents of the first prior application will be briefly described. In this earlier application, the distance to the preceding object, such as the distance between the preceding vehicle,
An object of the present invention is to stably, accurately and easily obtain a distance measuring device using a photodetector having an optical sensor array, thereby reducing the cost of a distance measuring device.

FIG. 6 shows an example of a configuration in which the technology of the first prior application is applied to an inter-vehicle distance measuring apparatus. A preceding vehicle 16 as a preceding object whose inter-vehicle distance is to be measured is illustrated. You are traveling in the same lane as your own vehicle. The imaging lenses 1 and 2 are arranged at an optical axis interval B, and the light receivers 3 and 4 are arranged at a focal length f (not shown for convenience). The light receiver 3 is composed of m light sensor arrays 31 to 3m arranged in parallel in a plane perpendicular to the optical axis. Similarly, the light receiver 4 is located in a plane perpendicular to the optical axis. It comprises m optical sensor arrays 41 to 4m arranged in parallel, and 31 and 41, 3i and 4i, and 3m and 4m are juxtaposed so as to have the same field of view.

The image of the object formed by the imaging lens 1 is converted into image signals 301 to 30 m by the optical sensor arrays 31 to 3 m of the light receiver 3, and similarly, the object formed by the imaging lens 2. The image of the object is stored in the light sensor array 4 of the light receiver 4.
The image signals are converted into image signals 401 to 40 m by 1 to 4 m and output to the signal processing unit 5 respectively. The signal processing unit 5 includes an amplifier 51
1 to 51 m and 521 to 52 m, A / D converter 531
53 m and 541 to 54 m and a storage device 55, and image signals 301 from the optical sensor arrays 31 to 3 m of the light receiver 3.
To 30 m are amplified by amplifiers 511 to 51 m.
The data is converted into digital data by the D converters 531 to 53m and stored in the storage device 55 as image data 311 to 31m.

Similarly, the optical sensor arrays 41 to 41 of the light receiver 4
The image signals 401 to 40 m from the 4 m are amplified by the amplifiers 521 to 52.
m, and is converted into digital data by A / D converters 541 to 54m, and the image data 411 to 41m
Is stored in the storage device 55. The distance detection circuit 6
The left and right image data 31 stored in the storage device 55 is constituted by a microcomputer as in FIG.
The distance from 1,411, 31i, 41i, 31m, 41m to the object within the distance measurement range within the measurement visual field of each optical sensor array 31, 41, 3i, 4i, 3m, 4m is calculated.

As shown in FIGS. 10 and 11, the distance block diagram extraction unit 7 sets the distance measurement range 23 in the measurement visual field 22 to m.
× n (m: natural number indicating the number of optical sensor arrays on one side, n:
A distance block diagram 24 is created in which the blocks are addressed as (a natural number indicating the number of measurement windows in the optical sensor array) distance blocks and measurement distance information in each block is collected.
This distance block diagram can be said to be a collection of m × n distance measurement data.

Here, as an example, the number m of optical sensor arrays is 7, and the number n of measurement windows in the longitudinal direction of the optical sensor array is 1.
The case of No. 2 will be described with reference to FIGS. In these figures, the optical sensor arrays A1 to A7 are arranged in order from the top of the distance measuring range 23, the measuring points in the longitudinal direction of the optical sensor array are set to W1 to W12 in order from the left, and the measuring distance at the measuring point Wj on the optical sensor array Ai is measured. Is represented by Lij. In the illustrated example, 7 × 12 distance detection in the measurement visual field 22 is possible, and as a result, a distance block diagram 24 as shown in FIG. 11 is extracted.

Here, the principle of distance measurement of a plurality of points in the longitudinal direction of the optical sensor array will be described with reference to FIG. The configuration of the distance measuring device in this case is exactly the same as that of FIG. 16 except that each sensor array is divided into a plurality of regions (measurement windows). FIG. 12 shows an example in which the optical sensor array is divided into three regions. In such a measurement window area, a part of the optical sensor elements (in this example, the CCD elements) constituting the optical sensor array overlap between the measurement window areas (that is, some of the CCD elements (Belonging to the area of two adjacent measurement windows).

The objects O ₁ , O ₂ , and O ₃ for distance measurement are separated in three directions indicated by alternate long and short dash lines, that is, the center line direction and the direction of the angle α on both sides of the distance L _1. , L
It shall be located at a _2, L _3. The regions of each of the optical sensor arrays 3 and 4 that are paired with each other correspond to the objects O ₁ , O ₂ and O ₃ , respectively. In other words, to simultaneously imaged on a pair of regions of the optical sensor arrays 3 and 4 relates to the object O ₁ in the direction of the left angle α of the center line, as well as a pair of regions, at the same time The images are formed on the objects O ₂ and O ₃ located in the direction of the center line and the right angle α, respectively. Then, the distances L ₁ , L ₂ , L _{3 to} the respective objects O ₁ , O ₂ , O ₃ are represented by the following equations (4) to (6). The distances B, f, U11, U12, U13, U2 in these equations
1, U22 and U23 are as shown in FIG.

[0021]

L ₁ = B · f / (U21−U11)

[0022]

L ₂ = B · f / (U22 + U12)

[0023]

L ₃ = B · f / (U13−U23) Each shift amount (U21, U11, U22, U12, U13, U2) based on the image data of each of the optical sensor arrays 3 and 4.
Since 3) is obtained by the distance detection circuit 6, Equation 4
The distances L ₁ , L ₂ , and L ₃ can be obtained by Expressions (6).

The distance signal 12 thus obtained from the distance block diagram extracting unit 7 is output to the distance selecting unit 8 in FIG. The distance selection unit 8 obtains the distance frequency distribution of the distance signal 12, selects only the distance to the preceding vehicle 16 from the distance signal 12, and outputs the distance signal 13 to the moving average processing unit 9. Next, the operation principle of the distance selection unit 8 will be described with reference to FIG. In FIG. 13, a distance class K having a class width ΔL is provided on the horizontal axis, and a frequency value Y belonging to each class is shown on the vertical axis. Thus, the distance signal 12
Sort. Here, as shown in FIG. 14, the size occupied by the preceding vehicle 16 in the distance measurement range 23 is
6, the score of the measurement data considered to measure the distance to the preceding vehicle 16 also differs depending on the inter-vehicle distance.

Generally, when the inter-vehicle distance becomes n times, the size occupied by the preceding vehicle 16 in the measurement range becomes (1 / n) ² from a similar relationship. That is, it is easily estimated that the number of points of the measurement data for measuring the distance to the preceding vehicle 16 is also (1 / n) ² . Therefore, the score of the measurement data measuring the distance to the preceding vehicle 16 at a certain distance, that is, the frequency value is given by the following equation (7).

[0026]

Y = a / K ² (Y: numeric value, a: constant, K:
Distance class (distance)) The constant a is determined by the size of the preceding vehicle 16 itself and the shape of the distance measuring device, and a description of a specific method for obtaining the constant a is omitted here. This curve Y is shown in FIG.
, A median L (= (K2 + K3) / of a distance class K having a frequency value extending to an upper region of the curve Y.
2) is set as a candidate value of the measured distance to the preceding vehicle 16 and output to the moving average processing unit 9 as the distance signal 13.

The moving average processing section 9 performs moving average processing within the range of the size of the 7 × 12 distance block described above based on the distance signal 13, and calculates the distance average value, standard deviation / standard deviation at each moving average position. The moving average processing result 14 of the distance average value is
This is transmitted to the preceding vehicle recognition unit 10 as the preceding object recognition unit. The operation principle of the moving average process will be described with reference to FIG. First, the size of a distance block of i × j (i: a natural number equal to or less than m, j: a natural number equal to or less than n), which is a moving average, is determined from the distance signal 13 (median value L of the distance class). The size of this distance block is determined by the distance
The distance is determined by the distance between the vehicles, the size of the preceding vehicle 16 itself, and the shape of the distance measuring device.

Hereinafter, the moving average processing will be described with reference to the 7 × 12 distance block diagram described above as an example. Here, when the size of the distance block for taking the moving average is 3 × 6,
The results of taking the moving average of FIG. 15A are shown in FIGS.
(D). Assuming that Lij is a measurement distance at a j-th window position on the i-th sensor array, Aij, Sij, and Aij in FIGS. 15B, 15C, and 15D are used.
Dij is represented by Equations 8 to 10 below.

[0029]

Aij = { ^{i + 2, j + 5} } _{i, j} (Lij)} / (3
× 6) = AVG

[0030]

Sij = [９ ^{i + 2, j + 5} Σ _{j, j} (Lij−AV
G) ² ｝ / (3 × 6)] ^1/2

[0031]

Dij = Sij / Aij In Expression 8, AVG is an average distance value. For example, in FIG. 15A, the average value of the distance in the hatched portion is shown in FIG. 15B, the standard deviation is shown in FIG. 15C, and the standard deviation / average distance value is shown in the hatched portion in FIG. Area.

A preceding vehicle recognition unit 1 as a preceding object recognition unit
0 is connected to the moving average processing unit 9 and based on the distance average value, the standard deviation, and the standard deviation / distance average value as the moving average processing result 14 transmitted from the moving average processing unit 9,
It is determined whether or not the measurement object ahead is the preceding vehicle 16. For this determination, the value of Dij with the upper left corner of the distance block taking the moving average being Lij is used. If this value is smaller than a certain standard value b, the distance average is calculated at the moving average position of Lij in the measurement visual field. It is determined that the preceding vehicle 16 exists at the inter-vehicle distance represented by the value Aij,
Of the preceding vehicle and the inter-vehicle distance to the preceding vehicle 16
5 is output to an external alarm device or the like.

FIG. 7 shows a flowchart of the processing in the above-described example. That is, the distance detecting circuit 6 calculates m × n pieces of measured distance information using the data in the storage device 55 of the signal processing unit 5 (S1), and based on this, the distance block diagram extracting unit 7 Is created (S2). afterwards,
The distance selection unit 8 obtains a distance frequency distribution (S3), and extracts a median L of the distance class K (S4).

The moving average processing section 9 determines the size i × j of the distance block for which the moving average is to be calculated based on the median L (S5), and moves the m × n area by the i × j area. An average is obtained (S6). Next, a distance average value, a standard deviation, and a standard deviation / distance average value at each moving average position are calculated and transmitted to the preceding vehicle recognition unit 10 (S7).

The preceding vehicle recognizing section 10 compares the standard deviation / average distance value with the standard value b (S8). As a result, when it is determined that there is a preceding vehicle (S91), the moving average position is determined by the preceding vehicle. Is output as the inter-vehicle distance to the preceding vehicle (S10). If it is determined that there is no preceding vehicle (S92), the process ends. by the way,
In order to realize the above-described device, it is desirable that the measurement field of view be as wide as possible in order to reliably catch the preceding vehicle 16.
For this purpose, it is conceivable to lengthen the optical sensor array in the longitudinal direction, or to arrange a large number of pairs of optical sensor arrays, but in such a method, the device becomes large, and the optical sensor array is connected at the end portion of the light receiver. There is a problem that the blurring of the image is increased due to the aberration of the image lens, the characteristics are deteriorated, and the processing circuits such as an amplifier and an A / D converter are complicated.

Accordingly, in the configuration example of FIG. 8, the distance measuring device main body 01 is configured to swing so that the optical axis thereof takes a radial direction. In FIG. 8, a control circuit 18 is connected to the distance measuring device swing motor 17 and the signal processing unit 5 and controls signals 19 and 20 by the motor 17 and the signal processing unit 5.
Is sending to. The motor 17 is mechanically connected to the distance measuring device main body 01, and swings the distance measuring device main body 01 based on the control signal 19 so as to take the optical axis in a radial direction.

In the configuration example of FIG. 9, the distance measuring device is
The distance measuring device main body 01 is fixed, and the reflecting mirror 21 swings so that light from the radial direction is incident on the distance measuring device main body 01. The control circuit section 18 is connected to the reflection mirror driving motor 17 and the signal processing section 5, and outputs control signals 19 and 20 to the motor 17 and the signal processing section 5. The motor 17 is mechanically connected to the reflection mirror 21, and swings the reflection mirror 21 based on the control signal 19 so that light from the radial direction is incident on the distance measuring device main body 01.

As a technique for specifying the existing area of the preceding vehicle in the field of view of the light receivers 3 and 4, in addition to the technique of the first prior application described above, the prior application of the present applicant (for convenience, Second
Japanese Patent Application Laid-Open No. 7-280563. To briefly describe the technology of the second prior application, an image at a position where the light amount distribution on each optical sensor array of at least one of the light receivers takes a maximum value by the line detection unit is
It is detected as a line related to the lane. Next, based on the signal indicating the position of the line from the line detection unit, the possible position range of the preceding vehicle in the horizontal direction is detected by the distance measurement range detection unit. Next, based on the image formation position of the preceding vehicle within the possible position range, the distance detection unit calculates and calculates the inter-vehicle distance to the preceding vehicle using triangulation.

The following three methods are used as the line detection method. The first is a method of detecting an image as a line when the image forming position having the maximum value of the light quantity distribution is maintained within a set range for a set time, and the second is a method of detecting two optical sensor arrays in each light receiver. As a line, the image when the straight line connecting the image forming positions that take the maximum value of the light amount distribution above passes through the range set around the image forming position that takes the maximum value of the light amount distribution on the other optical sensor array is the center. The third way to detect
Is a method of detecting the image as a line when the width of the object obtained from the distance to the object obtained from the imaging position having the maximum value of the light amount distribution and the width of the image is within the set range. It is. The line referred to here includes a white line and other types of lines, for example, a yellow line.

[0040]

However, in the distance measuring apparatus as described above, the visibility is increased due to fogging or fouling of a transparent cover glass or lens for protecting the front surface of the lens, or fog. If the optical sensor element becomes defective, the distance measurement becomes inaccurate or impossible. If the optical sensor element (CCD element in this example) as a pixel in the optical sensor array becomes defective, the defective element is detected. The value of the evaluation function using the output of (1) becomes abnormal, and there is a problem that accurate distance calculation cannot be performed.

Accordingly, an object of the present invention is to solve the above-mentioned problems and to provide a distance measuring device capable of performing accurate distance calculation.

[0042]

In order to solve the above-mentioned problems, a distance measuring apparatus according to a first aspect of the present invention includes an optical system having a pair of optical axes parallel to each other (a cover glass CG, an imaging lens 1,
2), one or a plurality of photosensors (31 to 3m, 41 to 4m) arranged in parallel at predetermined intervals on the image plane corresponding to each optical axis of (3, 3). 4) are arranged in the longitudinal direction of the optical sensor array, and the image data (311 to 31m, 411 to 4m) of the optical sensor array of this pair of light receivers.
1m), a plurality of measurement windows set in the optical sensor array in a distance measurement device for obtaining a distance from a preceding object (such as the preceding vehicle 16) (via the vehicle area detection / ranging unit 91). low contrast state of the output of the optical sensor element of the inner is, the alarm signal is when a predetermined time or longer
Signal (102) is output by the poor visibility detection means (101).
A plurality of measurement windows indicating the low contrast state.
The number of windows exceeds a predetermined value, and
When a measurement window is present on a pair of receivers,
The poor visibility detection means transmits the alarm signal to an optical system.
A signal to warn of malfunction due to mist (F) or fog
You.

An optical system having a pair of optical axes (parallel to each other )
Bar glass CG, imaging lenses 1, 2, etc.)
One or more light sensor arrays (3
1-3 m, 41-4 m) are arranged in parallel at a predetermined interval.
(3, 4) of photodetectors are in the longitudinal direction of the optical sensor array
And the image data of the optical sensor array of this pair of receivers.
(311 to 31m, 411 to 41m)
The object ahead (via the leading vehicle 16 via the output / ranging unit 91)
Distance measurement device to determine the distance to
Within each measurement window set in the array
Low contrast state of the output of the optical sensor element, or multiple
State of the evaluation function obtained for each measurement window
If the state continues for more than a predetermined time, an alarm signal (1
02) that outputs poor visibility.
A plurality of measurement windows indicating the low contrast state
Is the total number of measurement windows in the photosensor array.
And such a plurality of measurement windows is 1
When only one receiver is present, or
Multiple measurement windows less than the total number of measurement windows in a ray
The evaluation function obtained for the window is
When the poor visibility detection means outputs the alarm signal,
A signal to warn of an operation failure due to contamination of the optical system (SP)
I do.

Further, the distance detecting devices according to the third aspect of the present invention can
An optical system having a pair of parallel optical axes (imaging lenses 1 and 2
) On the image plane corresponding to each optical axis.
Sensor arrays (31-3m, 41-4m) at predetermined intervals
One pair (3, 4) of photo detectors arranged in parallel is an optical sensor
The optical sensors of this pair of receivers are
Subarray image data (311 to 31m, 411 to 41
m) and each pair of lights (via distance detection circuit 6A etc.)
For each pair of measurement windows set in the sensor array
Calculation of the evaluation function (FE) of the
In a distance measuring device that detects the distance to the car 16 etc.)
Prior to the detection of the distance,
From the data, the corresponding light sensor element in the light sensor array
(P _k ) and an optical sensor element adjacent to the optical sensor element
(P _{k + 1} or (and) P _k-1 )
The absolute value of (| P _{k + 1} −P _k | or (and) | P _k −P
_k-1 |) exceeds a predetermined threshold value (THa).
The sensor element (Pk) is detected as a defective pixel, and at least
The defective pixel position information (non-
Defective pixel detecting means (20) for outputting good pixel information 202)
1) shall be provided.

Further, the distance detecting device according to claim 4 is characterized in that
Based on good pixel position information, this defective pixel is detected by the distance detection.
Means to exclude from the calculation of the evaluation function when
Road 6A). The distance measurement according to claim 5
The distance measuring device according to claim 4, wherein
The defective pixel detecting means detects the defective pixel by this distance.
When the measuring device is switched on, and / or
It is performed every predetermined period.

Further, the distance measuring device of claim 6 is
Or in the distance measuring device according to 5, wherein the defective pixel
Detecting means for detecting the presence of the defective pixel together with the defective pixel position information;
Output alarm signal (203) to warn you
I do. Further, the distance measuring device according to claim 7 is based on claims 1 to
6. The distance is detected by the distance measuring device according to any one of 6.
The object ahead of the vehicle is a vehicle.

An optical system having a pair of optical axes parallel to each other
Each optical axis of (cover glass CG, imaging lenses 1, 2, etc.)
One or more optical sensor arrays on the imaging plane corresponding to
(31-3m, 41-4m) are arranged in parallel at predetermined intervals.
The pair of light receivers (3, 4) is the length of the optical sensor array
Image of the light sensor array of this pair of receivers arranged in the direction
From the data (311 to 31m, 411 to 41m)
Target object (preceding vehicle) via the area detection / ranging unit 91
16 etc.) in a distance measuring device
Multiple measurement windows set in the sensor array
The difference between the maximum value and the minimum value of the output of the optical sensor
Below threshold or set in photosensor array
Optical sensors in each of the multiple measurement windows
If the sum of the absolute values of the differences
Due to the low contrast state, or multiple each
Abnormal state of evaluation function found for measurement window
Is longer than a predetermined time, an alarm signal (10
2) equipped with poor visibility detecting means (101) for outputting 2)
And

[0048]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram of a main part of an inter-vehicle distance measuring apparatus according to an embodiment of the present invention, and corresponds to FIG. In FIG. 1, reference numeral CG denotes a transparent cover glass provided on the front surface of the imaging lenses 1 and 2 for protecting an optical system;
Detects an area where a preceding vehicle can exist from the image data stored in the storage device 55, detects a distance of an object estimated to correspond to the preceding vehicle in this area, and outputs a distance signal 96 indicating the distance. The vehicle area detection / distance measuring section 91 includes the means 6 to 10 shown in FIG. The distance signal 96 corresponds to an inter-vehicle distance represented by a distance average value at the moving average position when it is determined that there is a preceding vehicle in FIG. That is, the distance signal 96 in FIG. 1 is included in the preceding vehicle information 15 in FIG. Reference numeral 101 denotes a visibility defect detection unit which is a main component of the present invention.

FIG. 2 is a view for explaining a first embodiment (hereinafter referred to as a first embodiment) of the operation of the visibility defect detecting means 101.
FIG. 4A is a diagram showing a distribution of outputs of CCD elements in an optical sensor array when visibility is poor, wherein FIG. 4A shows an example of a state in which a cover glass CG is clouded, and FIG. It is cloudy based on adhesion of water droplets and the like. FIG. 3B shows the optical sensor arrays (31 to 31) of the left and right light receivers 3 and 4 when the cover glass CG or the imaging lenses 1 and 2 are fogged or the view becomes invisible due to fog. 3m, any one of pairs of 41 to 4m), the output distribution of each CCD element, that is, the output of the CCD element (vertical axis) and the pixel number as an array coordinate of the CCD element (abbreviated as pixel NO., Horizontal axis) The following shows an example of the relationship with. That is, both the left and right optical sensor arrays have a small output difference between the CCD elements as a whole, and are in a so-called low contrast state.

In the first embodiment, the poor visibility detecting means 101
Is the CC in each of a plurality of measurement windows equal to or more than a predetermined number of the aforementioned measurement windows set in the optical sensor array.
The low contrast state of the output of the D element is the left and right light receivers 3,
4, and if this state continues for a predetermined time or longer, it is determined that the operation is defective due to fogging or fog, and an alarm signal 102 indicating this is output.

Note that the low contrast state of the output of the CCD element in the measurement window means the following state. (A) The difference between the maximum value and the minimum value of the output of the CCD element in the measurement window is equal to or less than a predetermined threshold. (B) As shown in Equation 11, CC in the measurement window
The sum of the absolute values of the differences between the outputs of the D elements is equal to or less than a predetermined threshold.

That is, the pixel value of the i-th pixel (CCD element output) is P _i , the number of the head pixel of the measurement window is _i 0,
Assuming that the width of the measurement window represented by the number of pixels is W and the threshold for this determination is C, Equation 11 is expressed as follows.

[0053]

Equation 11] _{^{i0 + W-1 Σ i =}} i0 | P i + 1 -P i | ≦ C Figure 3 illustrates the second embodiment of the operation of the visibility detecting means 101 (hereinafter referred to as Example 2) FIG. 7A is a diagram showing the distribution of the CCD element output in the optical sensor array when the visibility is poor. FIG. 7A shows an example in which the dirt SP is attached to a part of the cover glass CG, and FIG. ) Is this dirt SP
, Which is on the left side of the light receiver 3 in this example.
The pixel NO. Another distribution and the right receiver 4 paired with this photosensor array
The pixel NO. This is shown in comparison with another distribution. As shown in this figure, a dirt image SPa in the CCD element output distribution of the left optical sensor array.
Is in the low-contrast state described above.

When dirt SP exists in the optical system,
Since the stain SP generally does not exist in the same form in the left and right optical systems, the left and right images including the stain image SPa are different. As described above, when the left and right images are different, or in the case of the above-described low contrast based on dirt, the above-described evaluation function for determining the degree of coincidence between the left and right images becomes abnormal. Here, "the evaluation function is abnormal" means the following case. (A) The evaluation function has no minimum value (that is, the evaluation function monotonously increases or decreases). (B) The evaluation function has a plurality of minimum values, and the minimum value cannot be determined. (C) The slope near the minimum value of the evaluation function is smaller than a predetermined threshold. (D) The minimum value of the evaluation function is larger than a predetermined threshold (that is, the coincidence between the left and right images is poor).

In the second embodiment, poor visibility detecting means 101
The low contrast state of the output of the CCD element in each of the plurality of measurement windows less than the total number of measurement windows in the optical sensor array exists in only one of the left and right light receivers 3 and 4, and When this state continues for a predetermined time or more, or when the above-mentioned abnormal state of the evaluation function obtained for each of a plurality of measurement windows less than the total number of measurement windows in the optical sensor array continues for a predetermined time or more Determines that the operation is defective due to dirt, and outputs an alarm signal 102 indicating that. This can prompt the operator or the like to clean the cover glass CG.

FIG. 4 is a block diagram of a main part of an inter-vehicle distance measuring apparatus as one embodiment of the present invention, and corresponds to FIG.
In FIG. 4, the distance detection circuit is replaced with 6A in FIG. 6, and the defective pixel detection means 201 is provided in parallel between the storage device 55 and the distance detection circuit 6A. The defective pixel detecting means 201 is provided with the right and left light receivers 3 in the storage device 55.
The defective pixel is detected in advance from the image data 311 to 31m and 411 to 41m of the photosensor arrays 31 to 3m and 41 to 4m, and defective pixel information 202 including its coordinates is sent to the distance detection circuit 6A, and an alarm signal is output. 203 is output to warn an operator or the like of the presence of a defective pixel.

The distance detection circuit 6A has basically the same function as the distance detection circuit 6 of FIG. 6, and uses the image data 311 to 31m and 411 to 41m in the storage device 55 to detect the distance between the left and right. The evaluation function is determined by the difference between the image data of the corresponding pixels of each measurement window of the pair set in each optical sensor array, and the distance of the object related to the image in each measurement window is detected. The only difference from the distance detection circuit 6 in FIG. 6 is that the evaluation function is obtained using the difference between the image data of the corresponding pixels excluding the defective pixel based on the defective pixel information 202.

FIG. 5 is an image distribution diagram for explaining the operation of the defective pixel detecting means 201. FIG. 5A is a diagram showing an example in which the left and right photodetectors 3 and 4 are set in a pair of optical sensor arrays. 5 shows an example of an image distribution within one pair of measurement windows. The horizontal axis of the two image distribution diagrams in FIG. 9A is the pixel number (abbreviated as pixel number, i is a parameter representing the number value) as the coordinates of the CCD element, and the vertical axis PLi of the left image distribution diagram is PLi. Pixel value (image data) of the i-th pixel of the left optical sensor array
Similarly, PRi on the vertical axis of the image distribution diagram on the right side is the CCD element output as the pixel value (image data) of the i-th pixel of the right photosensor array. In this example, a case is shown where a defective pixel Pk having an extremely small pixel value happens to be present at the peak portion of the image of the right optical sensor array.

The evaluation function is defined as a value obtained by adding the absolute value of the difference between the pixel values of the corresponding pixels of the left and right measurement windows for all the pixels in the measurement window. Assuming that the numbers of the pixels located are ls and rs, and the width of the measurement window represented by the pixel number is W, the evaluation function FE is expressed by the following equation 12.

[0060]

FE = ^W- 1Σi _{= 0} | PL _{ls + i} -PR _{rs + i} | Therefore, if the pixel value of the defective pixel Pk is incorporated into the evaluation function FE, it is impossible to detect the correct distance to the object. It is clear. FIG. 5B shows the absolute value | PR _{i + 1} −PR _i | of the pixel value difference between adjacent pixels in the right photosensor array corresponding to the image distribution on the right side of FIG. Is shown. As shown in this figure, the absolute value of the difference between the pixel values of adjacent pixels usually indicates a protrusion value that is discontinuously larger than a predetermined threshold value THa in a defective pixel portion. Therefore, the defective pixel detecting means 201 of the present invention obtains the absolute value of the difference between the pixel values of adjacent pixels in the optical sensor array in advance when the power of the distance measuring device is turned on, and obtains the following Expression 13 or (and)
A defective pixel is detected by the operation of Expression 14.

That is, the pixel value of the defective pixel is generally Pk, Is k, the absolute value of the difference between the pixel value of the defective pixel and the adjacent pixel is equal to or larger than the predetermined threshold value THa as shown in Expressions 13 and 14, and the defective pixel P
k is specified.

[0062]

| P _{k + 1} −P _k | ≧ THa

[0063]

| P _k −P _k−1 | ≧ THa The defective pixel detection means 201 supplies the defective pixel information 202 including the coordinates k of the defective pixel to the distance detection circuit 6A and outputs an alarm signal 203. To warn an operator or the like of the presence of a defective pixel.

Based on the defective pixel information 202, the distance detection circuit 6A removes the defective pixel as described above.
The evaluation function is calculated for each corresponding pixel in the left and right measurement windows.

[0065]

According to the present invention, the state where the output of the CCD element in the plurality of measurement windows set in the optical sensor array of the light receiving device of the distance measuring device is low contrast, or the evaluation obtained for the plurality of measurement windows. When an abnormal state of the function continues for more than a predetermined time or when a defective pixel is detected, an alarm signal is output.
A warning can be given to an operator or the like, and an operation corresponding to a malfunction or a defective pixel can be performed. Further, when the number of the plurality of measurement windows in the low contrast state is equal to or more than a predetermined number, and such a measurement window is present in both of the paired light receivers, the alarm signal is clouded or a malfunction due to fog is warned. And the number of the plurality of measurement windows in the low contrast state is less than the total number of measurement windows of the optical sensor array, and the plurality of measurement windows in the low contrast state is one.
If there is only one photodetector, or if the evaluation function obtained for a plurality of measurement windows less than the total number of measurement windows of the optical sensor array is abnormal, the alarm signal is used as a signal to warn of malfunction due to dirt. Therefore, it is possible to cause the operator to perform an operation corresponding to each operation failure.

Using the image data of the photosensor arrays of the left and right photodetectors, an evaluation function is calculated for each pair of measurement windows set in each pair of photosensor arrays, and When detecting the distance to the object ahead, the absolute value of the difference between the image data of adjacent pixels in the photosensor array is obtained from the image data of the photosensor array prior to the detection of the distance, and the absolute value of the difference is calculated. A pixel whose value exceeds a predetermined threshold value is detected as a defective pixel, the defective pixel is excluded from the calculation of the evaluation function at the time of the distance detection, and the presence of the defective pixel is warned to an operator or the like by an alarm signal. Therefore, in order to accurately detect the distance to the object ahead, repair of a defective pixel, for example, cleaning of an optical system can be prompted.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a main part as one embodiment of the present invention.

FIG. 2 is a diagram showing an example of a CCD element output distribution in an optical sensor array when visibility is poor due to cloudiness or fog of a cover glass of a light receiving device.

FIG. 3 is a diagram showing an example of a CCD element output distribution in an optical sensor array when visibility is poor due to dirt on a cover glass of a light receiver.

FIG. 4 is a block diagram showing a configuration of a main part as one embodiment of the present invention.

FIG. 5 is an image distribution diagram for explaining the operation of the defective pixel detection means in FIG. 1;

FIG. 6 is a diagram showing a configuration example of an inter-vehicle distance measuring device based on the technology of the first prior application;

FIG. 7 is a flowchart showing processing of the apparatus in FIG. 6;

FIG. 8 is a configuration diagram illustrating a different example of the imaging mechanism unit in FIG. 6;

FIG. 9 is a configuration diagram illustrating another different example of the imaging mechanism unit in FIG. 6;

FIG. 10 is an explanatory diagram of a measurement visual field and a distance measurement range in FIG. 6;

FIG. 11 is a schematic diagram showing a distance block diagram of FIG. 6;

FIG. 12 is a diagram illustrating a principle of distance measurement of a plurality of points in the longitudinal direction of the optical sensor array in FIG. 6;

FIG. 13 is a diagram illustrating the operation principle of the distance selection unit in FIG. 6;

FIG. 14 is a diagram showing the size of a preceding vehicle in the measurement visual field of FIG. 6;

FIG. 15 is a diagram showing the operation principle of the moving average processing unit in FIG. 6;

FIG. 16 is a configuration diagram of an inter-vehicle distance measuring device based on a technology prior to the prior application.

FIG. 17 is a diagram showing the principle of distance calculation in FIG. 16;

FIG. 18 is a diagram showing the operation principle of the distance detection circuit in FIG.

FIG. 19 is a diagram showing an image in a normal state in the following distance detection in FIG. 16;

FIG. 20 is a diagram showing inconvenience in the following distance detection in FIG. 16;

[Explanation of symbols]

1, 2 imaging lens 3, 4 light receiver 31 to 3m, 41 to 4m optical sensor array 5 signal processing unit 6A distance detection circuit 7 distance block diagram extraction unit 8 distance selection unit 9 moving average processing unit 10 preceding vehicle recognition unit 11 １３13 Distance signal 14 Moving average processing result 15 Information on preceding vehicle 16 Information on preceding vehicle 301 to 30m, 401 to 40m Image signal 311 to 31m, 411 to 41m Image data (light amount distribution data) 531 to 53m, 541 to 54m A / D Converter 55 Storage device 91 Vehicle area detection / ranging unit 96 Distance signal 101 Visibility detection means 102 Alarm signal 201 Defective pixel detection means 202 Defective pixel information 203 Alarm signal FE Evaluation function CG Cover glass F Cloudy SP Dirty SPa Dirty image

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-7-280563 (JP, A) JP-A-58-161816 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01C 3/06 G01B 11/00 G01J 1/02 G01M 11/00

Claims (8)

(57) [Claims] 1. A photodetector comprising: an optical system having a pair of optical axes parallel to each other; one or a plurality of optical sensor arrays arranged in parallel at predetermined intervals on an image plane corresponding to each optical axis; In a distance measuring device in which one pair is arranged in the longitudinal direction of the optical sensor array, and a distance to an object in front is obtained from image data of the optical sensor array of the pair of photodetectors, a plurality of pairs set in the optical sensor array low contrast state of the output of the optical sensor elements in each measurement window is, where
If an alarm continues for more than the specified time,
The low-contrast state.
If the number of measurement windows exceeds a predetermined value and
Multiple measurement windows such as
When present, the poor visibility detection means outputs the alarm
Signals warning of malfunction due to cloudy or fog of optical system
A distance measuring device, which is a signal . 2. An optical system having a pair of optical axes parallel to each other.
One or more optical sensor antennas are respectively provided on the image plane corresponding to the optical axis.
A pair of light receivers, in which rays are arranged in parallel at predetermined intervals,
It is located along the length of the sensor array and
From the image data of the optical sensor array, determine the distance to the object ahead.
In the distance measurement device to be sought, a plurality of measurement windows set in the optical sensor array
(C) a low contrast state of the output of the light sensor element in
Differences in the evaluation function obtained for each of the multiple measurement windows
If the normal state continues for more than the specified time, an alarm signal
Signal detection means for outputting a low contrast signal.
The number of measurement windows indicating the
Less than the total number of measurement windows in the optical sensor array, and
Such multiple measurement windows can be
Only when present or all measurements in the relevant photosensor array
For multiple measurement windows less than the number of windows,
When the evaluation function obtained in the above is an abnormal state,
The poor visibility detection means converts the alarm signal into a dirty optical system.
A distance measuring device, which serves as a signal to warn of a malfunction caused by the distance measurement. 3. A photodetector having one or a plurality of optical sensor arrays arranged in parallel at a predetermined interval on an image plane corresponding to each optical axis of an optical system having a pair of optical axes parallel to each other. One pair is arranged in the longitudinal direction of the optical sensor array, and using the image data of the optical sensor array of this pair of photodetectors, calculates an evaluation function for each pair of measurement windows set in each optical sensor array. And a distance measuring device for detecting a distance to a forward object, wherein, prior to the detection of the distance, the image data of the optical sensor array is used to determine the distance between the relevant optical sensor element in the optical sensor array and the optical sensor element. A light sensor element whose absolute value of the difference between the image data and the detected light sensor element exceeds a predetermined threshold value is detected as a defective pixel, and at least defective pixel position information for specifying the position of the defective pixel is output. Bad picture A distance measuring device comprising element detection means. 4. The distance measuring device according to claim 3, wherein said defective pixel is determined based on said defective pixel position information.
Means to exclude from the calculation of the evaluation function when detecting the distance.
A distance measuring device characterized by the following. 5. The distance measuring apparatus according to claim 3, wherein said defective pixel detecting means detects said defective pixel by using said distance.
When the power of the remote measuring device is turned on, or (and) the power is turned on
The distance measuring device is performed every predetermined period. 6. A distance measuring apparatus according to claim 3, wherein
In the above, the defective pixel detecting means is provided together with the defective pixel position information.
Output an alarm signal to warn of the presence of a defective pixel
A distance measuring device characterized by the above-mentioned. 7. The distance measurement according to claim 1,
In the fixed device, the object in front of which the distance is detected is a vehicle.
Characteristic distance measuring device. 8. An optical system having a pair of optical axes parallel to each other.
One or more optical sensor antennas are respectively provided on the image plane corresponding to the optical axis.
A pair of light receivers, in which rays are arranged in parallel at predetermined intervals,
It is located along the length of the sensor array and
From the image data of the optical sensor array, determine the distance to the object ahead.
In the distance measurement device to be sought, a plurality of measurement windows set in the optical sensor array
C The difference between the maximum and minimum output of the optical sensor element in
Below the threshold value or installed in the optical sensor array.
Adjacent light sensors in each of the specified measurement windows.
The sum of the absolute values of the output differences of the
Low contrast conditions due to a, or a plurality of each
Abnormal state of evaluation function found for measurement window
However, an alarm signal is output if it continues for a predetermined time or more.
Distance measurement device comprising
Setting device.