JPS62162115A - Method for image inputting of image type unmanned vehicle - Google Patents

Abstract

PURPOSE:To allow an image pickup device to pickup the image of a traveling line constantly and accurately at all times by keeping the lightness of the position or nearby place of image pickup of the image pickup device to lightness suitable for taking picture for the device. CONSTITUTION:An illumination controller 19 controls the intensity of illumination of an illuminating lamp 5 by controlling signals of a CPU 11. By adjusting voltage V applied to the illuminating lamp 5 within the range of Vmin-Vmax by the illumination controller 19, the illumination of the illuminating lamp 5 can be adjusted linearly. At the time of initial setting, voltage is set to the intermediate voltage Vc of above-mentioned range. A drive controller 20 controls a motor for running and a steering mechanism, not shown in the figure, based on control signals from the CPU 11.

Description

[Detailed Description of the Invention] Purpose of the Invention (Field of Industrial Application) The present invention relates to an image input method for an unmanned vehicle, and more specifically, it is related to an image input method for an unmanned vehicle, and more specifically, to achieve reliability by making the image device fail under optimal conditions. The present invention relates to an image input method for an image-based unmanned vehicle that obtains highly accurate images.

(Prior art) Conventionally, this type of unmanned vehicle divides the image of the driving line that indicates the driving route taken by the imaging device into the left half and the right half, and the right half has the area covered by the white line and the left half. Find the area where the white line dances on each side, and calculate the difference between the two.

Then, based on the difference, the unmanned vehicle controls the steering mechanism to determine how to travel along the travel line, and calculates how far (!) the travel line is from the center of the screen.
The method of performing III was adopted.

(The problem that Kanmei is trying to solve) However, the captured image is greatly affected by the surrounding environment (brightness) at the time the image was captured. For example, depending on the brightness,
Noise was picked up, the traveling line could not be accurately imaged, and the area and deviation during image processing often showed values that were far different from the actual clay. Therefore, there is a problem in the reliability of the results of image processing of the image of the traveling line.

The purpose of the present invention is to solve the above-mentioned problems by using an image-based unmanned vehicle that is capable of obtaining highly reliable images of the traveling line taken by an imaging device and showing the traveling route under optimal conditions. Provides an image input method.

Structure of the Invention (Means for Solving Problems) This invention is a dance l! which is mounted on an unmanned vehicle to achieve the above object. Detects the illuminance at or near the imaging position of the device, adjusts the illuminance of the lighting lamp installed on the unmanned vehicle so that the illuminance is suitable for the image conveyed by the device, and instructs the unmanned vehicle's travel route. An image of an image-based unmanned vehicle whose driving line is conveyed by a retraction device (the gist is the input method).

(Function) Since the brightness at or near the imaging position of the imaging device is always maintained at a brightness suitable for conveying the image of the circuit imaging device, the imaging device can always convey images along the traveling line in a constant and accurate manner. .

(Example) Hereinafter, the input method for the pixel-type unmanned vehicle of the present invention will be described.
An embodiment of a four-unit unmanned ill control device will be described with reference to the drawings.

In FIG. 1, a support frame 2 is erected at the front upper center of the unmanned vehicle, and at the upper center of the frame 2 there is a Hil f/G and an Oct) (c) device.
A coupled device camera 3 is installed. The CCD camera 3 is mounted on the support frame 2 so as to move over pixels in an area 4a on a road surface 4 in front of the unmanned vehicle 1.

Therefore, in this embodiment, the CCD camera 3 is
The image 9 of the area 4a shown in FIG.
It is configured to consist of 256 pixels.

Illumination lamps 5 are installed below both sides of the CCD camera 3, and are illuminated to illuminate the area 4a.

On the road surface 4, as shown in FIG. 2, a travel line 6 is drawn with a certain line width to indicate the travel route of the unmanned vehicle. It is depicted in. Then, the CCD camera 3 photographs the traveling line 6 having a constant line width.

Note that the level of the signal from the CCD camera 3 that decoys the white running line 6 (hereinafter referred to as a pixel signal) is high, and on the contrary, the level of the pixel signal from the CCD camera 3 that decoys the dark road surface 4 is low. .

Next, the electrical configuration of the control ill device mounted on the unmanned vehicle 1 will be explained with reference to FIG.

The microcomputer 10 includes a central processing unit (hereinafter referred to as CPU) 11 and a program memory 1 consisting of a read-only memory (ROM) that stores control programs.
A readable and interchangeable memory (RAM
), a timer 14, and the like, and the CPU 11 executes various arithmetic processing operations for steering control using a control program stored in the program memory 12.

The CPU 11 controls the input/output interface 15 and the A at regular intervals based on the time measured by the Qima 14.
,/D converter 1G to scan and control the CCD camera 3, and convert the pixel signal from the CCD camera 3 into pixel data via the Δ/D converter 16 and bypass capacitor roller 17 to the working memory 13. to be memorized. Δ/D
When converting the pixel signals from the CCD camera 3 from analog 1 to digital values, the converter 16 determines whether each pixel signal described later is equal to or higher than a predetermined binarization level (threshold 1a) TH, In the case of a pixel signal of value conversion l novel Tl1 or higher, the pixel of the white running line 6 is set to "1",
On the other hand, less than i! 'ii For each Kishinyuki event, each pixel is set to "0" as the pixel of the dark road surface 4, and each pixel is successively made larger; It is stored in the working memory 13 via T117.

Therefore, the working memory 13 stores the image captured by the CCD camera 3 as 256 x 256 pixel data.

In addition, for the convenience of explanation, in this lesson/+I lesson, the scanning control of the COD camera 3 scans in the horizontal direction (X-axis direction), and a scanning method is adopted in which the scanning moves from the top of the screen to the bottom (Y-axis direction). Of course, it is also possible to use other scanning methods.

The binarization level controller 18 is the CPU
The binarization level Tl-1 for binarization of the A/D converter 16 is set based on the control signal from the A/D converter 1.
6 is output. In this embodiment, the binarization level controller 18 sets the binarization level Tl-1 in the range of TH0 to 1-1255 (256
It can be adjusted in stages). At the time of initial setting, the data is pitted at the binary level TH128, which is the middle of the range.

The illuminance controller 19 controls the illumination intensity of the illumination lamp 5! I is controlled by a control signal from the CPU 11. In this embodiment, the illumination intensity of the illumination lamp 5 is directly adjusted as shown in FIG. It is now possible to do so. And η
When setting J light/dark, it is set to 7R VC, which is in the middle of the above range.

In addition, the drive controller 1-○-ra 20 has a driving morph and a steering □M? J'21 as well as CPU1
Control is based on the ft1ll l wholesale signal from 1.

Next, the processing operation of the CI'''tJ11 will be explained.

Now, when you activate the illumination device, the contents of the CPIJll work menu 12 will be cleared and initial settings will be made.
Step 1). The initial setting is 211! when one illuminance controller binarizes the initial applied voltage Vc of the voltage V applied to the illumination lamp 5 and the pixel signal. The initial level T812B of the binarization level T)-1 which the I conversion level controller 1-roller 18 outputs to the A, -'D converter 16 is set. At this time 1. The CPU 11 calculates the correction value ΔT1- of the binarization level TH.
1 and the correction value ΔV of the applied voltage V. In this embodiment, the correction value ΔTH is TH128,
The corrected tuffΔV has a value of (VmaX − Vmin)/2.

Then, based on these 6 ships, the cpu il is 2
Value level controller 18 and illuminance controller 1
9 to output a control signal to set the initial binarization level T.
A at H (=TH128) and applied voltage V (=Vc)
/D converter 16 and illumination lamp 5 are operated and controlled.

When the initial settings are completed, the CPU II starts the illuminance adjustment processing operation. First, when the cpuii operates the CCD camera 3 under the illuminance state of the illumination lamp 5 that is lit with the applied voltage Vc, the CCD camera 3 captures an image on the area 4a in front that includes the running line 6, and A as a signal
/D converter 16 (step 2). The A/D converter 16 compares the level of the pixel signal output from the CCD camera 3 with the initially set binary level TH128, converts it into a binary value, and outputs the pixel data via the bus controller 17. The data is stored in the user memory 13 (step 3).

Then, CPU'+i executes an n ratio processing operation for the width of the travel line 6 in the captured image based on the 2'56 x 256 pixel data stored in the working memory 13 (
Step 4). In this dedicated processing, in this embodiment, C
PU11 is "1" out of 256x256 pixel data
The width of the traveling line 6 in the image marked with R is determined by finding the total sum N of the pixel data.

After the total sum Nt has been used, the CPU 11 checks how many times the illuminance adjustment processing operation has been repeated, that is, up to 8 times in this embodiment (step 5).
. In this embodiment, the number m is stored in advance in a predetermined storage area of the working memory 13, and its contents are stored each time the In operation is performed at a3 in step 2. is incremented by 1.

At this time, since it is only the first time, CPtJll judges whether the width of the running line 6 is within an allowable range based on the calculated total Nt (step 6).

This judgment is made within a predetermined range (Ntlli) by the CPU 11.
This is a judgment as to whether or not the total sum Nt exists (n≦N and ≦NtllIax). The minimum value N tmin and the maximum value NtllaX are calculated using the true value of the number of pixel data obtained by imaging the running line 6 when the CCD camera 3 images the area 4a including the running line 6 with the line width under an optimal environment. The allowable value for the true value has been determined in advance through experimentation, and if it is within that range, the recognition of the traveling line 6 imaged at the appropriate illuminance and binarization level will be accurate based on the pixel data. It is 1:f that can be performed.

If the sum N is less than the minimum value Ntmin, the CPU 11 determines that the running line 6 is too narrow, that is, too bright, and changes the applied voltage V (=VC) to lower the illuminance of the illumination lamp 5. do.

This process is performed using the correction value △V set before the initial settings.
After calculating a new correction value △V that is halved (step 7), the calculated correction value △V is used to lower the illuminance.
and the previous applied voltage V (=VC) to create a new applied voltage V
Find it using the following procedure (Step 8).

V=Vc-Δ■ Therefore, the illumination lamp 5 is controlled to be turned on using the calculated new applied voltage V, and illuminates the area 4a with a slightly lower illuminance than the previous illuminance 1α.

On the other hand, if the total sum Nt is greater than or equal to the maximum value Ni1naX, CPtJll determines that the running line 6 is too thick, that is, it is too turned on, and changes the inlet voltage (=VC) to increase the illuminance of the illumination lamp 5. do. This process creates a new correction value Δ which is half of the correction value △■ set in the initial settings.
After calculating (step 9), the i+Ii positive value ΔV and the applied voltage V (=Vc
) and perform the following a1μ to find the new applied voltage V (
Step 10).

V=VC+ΔV Therefore, in this case, the illumination lamp 5 is controlled to turn on using the calculated new applied voltage V, and does not illuminate the first rear 4a with an illuminance slightly brighter than the previous illuminance.

In this way, CI]U11 has a consonance Nt in the range (
The processing operations of Steps 2 to 10 (in this embodiment, 8
(up to 10 times). That is, in steps 2 to 10, 3
The illuminance adjustment processing operation of the illumination lamp 5 in step 3 is a processing operation for finding an illuminance within a permissible range when the CCD camera 3 carries an image in a so-called binary search.

Then, if the total number Nt falls within the allowable range by the time the number m of repetitions of the processing operation is 8 times (step 6), the CPI
Jll is the @image position of CCD camera 3, that is, area 4a
It was determined that the assembly was illuminated by lighting lamp 5 at an illuminance that did not cause any problems, and the illuminance adjustment process was terminated. Control for travel control will be performed based on the image.

On the other hand, if the illuminance adjustment processing operation is repeated eight times and the total number Nt is not within the allowable range, CPIJll is set to the binarization level TH of the binarization of the A/'D converter 16.
is inappropriate, and it is determined that JliTl of the CCD camera 3 is not sufficient just by adjusting the illuminance of the illumination lamp 5 (step 5).
), when performing the first shift (binarization level) adjustment processing operation, first, the CPU 11 clears the number of repetitions m stored in a predetermined storage area of the working memory 13 (step 11), and then turns on the illumination lamp 5. The voltage V is set to the applied voltage VC set in step 1, and the correction value ΔV is similarly set (step 12).

When the CPU II is turned on with an applied voltage Vc and operates the CCD camera 3 under the illuminance of the illumination lamp 5, the CCD camera 3 detects the area 4a in front of the running line 6 where it meets.
The upper image is reimaged and output as a pixel signal to the A/D converter 16 (step 13). A/D converter 1G is CCD
The level of the pixel signal output from the camera 3 is compared with the binarization level TH128 initially set in step 1, and the binarized signal is stored as pixel data in the working memory 13 via the bus controller 17 (step 14). Then, the CPU 11 executes the calculation operation of the width disturbance and the total sum Nt of the traveling line 6 in the rain imaged in the same manner as in step 4 (step 15).

After calculating the total sum Nt, the CPU 11 checks how many times the binarization level adjustment processing operation has been repeated, that is, whether the number n has been repeated up to 8 times in this embodiment (step 16). In this embodiment, the number n is stored in advance in a predetermined storage area of the working memory 13, and each time the imaging operation is performed in step 13, the content is changed to 1. It is added in increments. At this time, it is only the first time, so CP
U11 judges whether the width of the running line 6 is within an allowable range based on the total sum Nt calculated on the surface (step 17).

This judgment is made by the CPU 11 in the same way as in step 6 as to whether or not there is a total sum Nt of the above-mentioned nodes (Ntmin≦Nt≦NtIllax).

If the total sum Nt is less than the minimum faNtmin, the CPU 11 determines that the binarization level TH is high, that is, the running line 6 is too thin, and sets the binarization level T I-
The dichotomous level TH128 is changed to lower it by 1. This process is performed using the correction field ΔT H (=
T I-1128) is halved fIi and complement 1fi
CIΔT H(= T l-164> wo request/jtu
(Sj"t 718), the desired meta i + U positive first detective ΔT I-I G4 by lowering the binarization level
Destination (1) 21+ff Convert L/Hell T H (=TH
128) and a new 2! The IiI level is determined and H is determined using the following formula (step 19).

Tl-1=TH128-ΔT t1=T
H64 Then, this binarization level T I-I 64, that is, the 21 digitalization level T H is 64 in the stage of O to 255.
CPtJll outputs a control signal to the binarization level controller 18 so that the pixel signal is binarized at a level corresponding to the step. 2 (The lO conversion level controller 18 responds to this and binarizes the pixel signal input to the A/D converter 16 at the binary conversion level TH64.
,' controls the D converter 16.

On the other hand, if the total sum N is greater than or equal to the maximum value N tma
It is determined that the 1-terrain 6 is too thick, and the binarization level TH128 is changed to lower the same level. In this process, after calculating a new correction value ΔTH64 in the same way as in step 18 (step 20), in order to increase the binarization level TH, the calculated correction value ΔTH(=TH(34)
and the previous binarization level TH (=TH12'8), a new binarization level T I-1 is calculated by the following calculation (Sj
・Whelk 21).

T1=TH128+ΔTH=TH192 And this binarization level TH192, that is, the binarization level T
The CPU 11 uses 21i so that the pixel signal is binarized at a level corresponding to 192 levels of H from 0 to 255.
A control signal is output to the ti level controller 18.

In response to this, the A/D conversion level controller 18 converts the pixel signal input to the A/D converter 16 into a binary signal at a binary conversion level TH192.
Control the converter 16.

When the binarization level TH is adjusted in this way, C
The PIJll performs the illuminance adjustment processing operation again under the adjusted binarization level of the line E (steps 2 to 1).
0). Under the condition of the newly adjusted 211a level TH, the applied voltage V is varied (the brightness of the illumination lamp 5 is varied), and the imaging area 4a of the CCD camera 3 is illuminated with an illuminance that does not interfere with imaging. The brightness of the illumination lamp 5 is adjusted and controlled so as to

Then, when the brightness of the illumination lamp 5 reaches an illuminance that does not interfere with the adjustment process, the process is terminated, and the image capturing process is performed under this favorable imaging environment in the same manner as described above.
Control for traveling control is performed based on the +1 image.

On the other hand, if a good illuminance is not found even after the illuminance adjustment operation is repeated eight times under the condition of the adjusted binarization level TH, the CPU 11 determines that the binarization level TH is not sufficient. (Step 5), and performs the binarization level adjustment processing operation again (Steps 11 to 21).
.

Then, by performing the same processing operation as above, a new 21-level conversion level Tl-1 is set, and the illuminance adjustment processing operation is performed again at this new binary conversion level TH (step 2-10). Thereafter, the above processing operation is repeated until a good imaging state is achieved. Then, once the binarization level TH for good imaging and the brightness applied voltage V) of the illumination lamp 5 are determined, control for driving control is started.

Incidentally, even if the number of times n of the binarization level adjustment processing operation is performed up to 8 times, the binarization level T +- for a good image will still remain.
If the brightness (V) of the illumination lamp 1 and the illumination lamp 5 and the applied voltage V) are not determined, it is assumed that there are no good image carrying conditions and the subsequent processing operation is stopped.

In this way, in this embodiment, when changing the running line 6, the binarization level and illuminance are always adjusted, so that the running line can always be imaged in the optimal situation.
It is possible to produce 17 highly reliable images. Therefore, unmanned I! Even if the environment in which the CCD camera 1 travels, for example, the brightness, etc., changes, the CCD camera 3 will always be able to take images in the optimal condition, and therefore optimal control can be performed.

Note that this invention is not limited to the above-mentioned eggplant example,
For example, in the above embodiment, the binarization level T)-1 is set in advance for each! '3, the illuminance of the illumination lamp 5 was made to vary under that condition, but this was reversed and the illuminance of the illumination lamp 5 was kept constant and the illuminance level was varied as shown in Figure 8. Good positive image focusing may be set by doing so.

Further, in the embodiment C, the illuminance of the illumination lamp 5 was adjusted in relation to the binarization level TH, but as shown in FIG. It may be implemented so that only adjustment is performed. Also, in the above embodiment, the CCD camera 3 is I8f? (I L The image (I judged the quality of the illuminance based on the width of the running line 6 during pleasure, but this can be determined by detecting the area 4a or its vicinity with a light meter instead of directly using the CCD camera 3. The brightness of the illumination lamp 5 may be controlled based on the result.Furthermore, in the above-mentioned practical example, the illuminance is determined by the width of the running line 6, and the width is determined by the NJ of the pixel data. The sum of N
Although the judgment was made based on tl,:, the judgment may be made by other methods.

Further, in the above embodiment, the CCD camera 3 was used as the R imaging device, but even if other imaging devices are used, J:
<Also, in the above embodiment, the pixel configuration (resolution) of the image captured by the CCD camera 3 was set to 256 x 256 pixels.
It is not limited to this, for example, 512X512
The number of pixels, 1024×1024 pixels, etc. may be changed as appropriate.

Effects of the Invention As described in detail above, according to the present invention, the R-imaging device can image and transmit the traveling line indicating the decoyed traveling route under the optimum conditions. It is possible to obtain images with high Il 4'I, and has excellent effects as an image input method for image-based unmanned vehicles.

[Brief explanation of drawings]

Figure 1 is a side view of an unmanned vehicle embodying this invention;
The figure is also a plan view, Figure 3 is a 1. Figure 6 is a diagram to explain the relationship between applied voltage and illuminance, Figure 7 is a flowchart diagram to explain the operation of the 11111311 device, and Figures 8 and 9 are diagrams to explain the relationship between applied voltage and illuminance. 1 is a flowchart for explaining the invention in detail. In the figure, 1 is an unmanned vehicle, 3 is a CCD camera, 4 is a road surface, and 4a
is the area, 5 is the illumination lamp, 6 is the running line, 9 is the image, 10 is the microcomputer, and 11 is the central processing unit (
12 is a program memory, 13 is a working memory, 14 is a timer, 16 is an A/D converter, 18 is a binary level controller, 19 is an illuminance controller, 2
0 is a drive control roller, and 21 is a steering mechanism. Patent Applicant: Toyota Industries Co., Ltd. (Representative of Boru Seisakusho, Patent Attorney On 1) Hiroki 6th Vmin Vc Vrmx■9

Claims (1)

[Claims] 1. The illumination lamp mounted on the unmanned vehicle detects the illuminance at or near the imaging position of the imaging device mounted on the unmanned vehicle, and adjusts the illumination lamp mounted on the unmanned vehicle so that the illuminance is suitable for imaging by the imaging device. An image-based image input method for an unmanned vehicle, which adjusts illuminance and uses an imaging device to image a driving line indicating the driving route of the unmanned vehicle. 2. The illuminance is detected by processing the image of the traveling line taken by the imaging device, and detecting the illuminance based on the width of the traveling line.
Image input method for image-based unmanned vehicles described in Section 1.