JP2625716B2 - Automatic distance measuring device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic distance measuring device capable of obtaining a measured value of a distance to an object to be measured with high accuracy.

<Prior Art> Conventionally, there is a distance measuring apparatus as shown in FIG. 19, which is used for a camera or the like and optically measures a distance to an object to be measured. In this distance measuring device, a light beam from the object to be measured 52 that has passed through the main lens 51 and is incident on the stop mask 53 is converted into two light beams 54a and 54a symmetric with respect to the optical axis.
54b, re-focused by the separator lenses 55 and 56, connect the two images 58 and 59 of the measured object 52 on the CCD (charge coupled device) image sensor 57, and The distance between the images 58 and 59 is obtained using an electric signal. Since the position of each element of the optical system is fixed, there is a certain relationship between the distance to the measured object 52 and the distance between the images 58 and 58 of the measured object 52. Thus, the distance to the measured object 52 is determined based on the distance between the images 58 and 59.

<Problems to be Solved by the Invention> However, in the above-described conventional distance measuring device, the images 58 and 59 of the measured object 52 are optically formed on the CCD image sensor 57 and are obtained by photoelectric conversion. Since the distance to the measured object 52 is measured by detecting the distance between the images 58 and 59 using an electric signal, the intensity, contrast, and pattern of the reflected light from the measured object 52 are measured. For example, the reliability of the distance measurement result is significantly different depending on factors such as the output of the obtained distance value, and the reliability of the value is unknown.

Therefore, an object of the present invention is to provide an automatic distance measuring device capable of calculating a distance value to an object to be measured and calculating its reliability, and outputting the obtained distance value and its reliability. is there.

<Means for Solving the Problems> In order to achieve the above object, an automatic distance measuring apparatus according to the present invention includes an optical system, a photoelectric conversion element for receiving an object to be measured formed by the optical system, Distance measuring means for calculating the distance to the object to be measured based on the output of the conversion element, reliability calculating means for determining the reliability of the measured distance based on the output of the photoelectric conversion element, and being connected to another device. A connection means for outputting a signal corresponding to the measured distance and a signal representing the reliability of the distance to the outside of the apparatus, so that the distance to the measured object and the reliability thereof can be obtained. Features.

<Operation> The distance to the measured object is determined by the distance measuring means, and the reliability of the measured distance is calculated by the reliability calculating means. Then, when the connection means is connected to another external device, a signal corresponding to the measured distance and a signal representing the reliability of the distance are output outside the device. Therefore, in the above-mentioned other device, it is possible to cope with the reliability of the measured distance.

<Example> Hereinafter, the present invention will be described in detail with reference to an illustrated example.

FIG. 1 is a block diagram showing a schematic configuration of an embodiment of the present invention. The optical system 1 divides a light beam from the measured object 2 into two, and forms two images of the measured object 2 on a CCD linear image sensor (hereinafter referred to as CCD) 3.
The CCD control unit 4 includes a circuit for driving the CCD 3 and a circuit for A / D converting a signal obtained from the CCD 3 and outputting the signal to the internal signal line 5. The arithmetic unit 6 detects the positions of the two images of the measured object 2 from the data output from the CCD control unit 4 and calculates the distance between the two images, thereby obtaining the distance value to the measured object 2 and its reliability. And outputs the obtained distance value and its reliability to the external host computer 7 via the interface bus 8 (GP-IB is used in this embodiment).

Hereinafter, the above configuration will be described in more detail. Fig. 2
FIG. 1 is a diagram showing an optical system 1 in FIG. 1, which is composed of a main lens 1 ′, a condenser lens 10, a pair of re-imaging lenses 12 a and 12 b provided with an aperture mask 11, and the re-imaging lenses 12 a and 12 b. It is composed of the above-mentioned CCD3 and the like provided on the image forming plane 12b. The CCD 3 includes a photo sensor array, a charge-coupled device, and the like, and has a photoelectric conversion function.

The distance measurement in the optical system 1 is based on the following principle. That is, the light beam from the object to be measured 2 that has passed through the main lens 1 ′ is incident on the stop mask 11.
Are split into two light beams 13a and 13b symmetrical with respect to the optical axis, and are converged by the re-imaging lenses 12a and 12b.
The two images 14a and 14b of the measured object 2 are formed on the CD3. Here, when the position of each element of the optical system 1 is fixed,
As shown in FIG. 3, the distances (hereinafter referred to as image intervals) 15, 15 'between the images 14a, 14b of the measured object 2 change according to the distance from the main lens 1' to the measured object 2. I do. That is, the distance from the main lens 1 'to the object 2 to be measured and the image interval 15, 15' are associated with each other in a relationship unique to the optical system 1 (see FIG. 13 (c)). Image spacing
By obtaining 15,15 ', the distance to the measured object 2 is obtained.

FIG. 4 is a diagram showing the CCD control section 4 in FIG. 1. The CCD drive circuit 16 for driving the CCD 3 and a CCD signal input from the CCD 3 in accordance with the luminance of the object 2 to be measured are shown in FIG. An AGC (automatic gain control) circuit 17 that amplifies at any magnification of 1, × 2, × 4, and × 8 and always outputs a CCD signal at an appropriate level;
An A / D converter 18 for sequentially converting an analog signal output from 7 into a 1-byte digital signal; a CCD data converted to a digital signal and a gain when the CCD signal is amplified by the AGC circuit 17 (hereinafter referred to as AGC). , Etc.) to the internal signal line 5. The signals transmitted by the internal signal line 5 include the following. That is, the power supply voltage V _CC supplied to the CCD control unit 4, the reset signal RESET, which resets the circuit in the CCD control unit 4, and further performs the initial operation of the CCD3.
A start signal START for notifying the start of transmission of CCD data and AGC data from the control unit 4 to the outside, a synchronization signal SYNC for transmitting the above data, and a transmitted data signal DATA
1, and each signal is transmitted at the timing shown in FIG. Note that the above C that constitutes the CCD data to be transmitted
Pixel data from each pixel of CD3 is 8 bits, the data of the 8 bits, as shown in FIG. _{5, b 7, b 6, ...} b 2, b
₁ , b ₀ , b ₆ , b ₄ , b ₂ , b ₀ (hereinafter referred to as even-numbered bits) are connected to the data line 1 of the internal signal line 5 and b ₇ , b ₅ , b ₃ , b ₁
(Hereinafter, referred to as odd bits) are output to the data line 2.

FIG. 6 is a diagram showing the arithmetic unit 6 in FIG. 1. The arithmetic unit 6 is composed of a timer circuit 26, an I / O port 22, a CPU 23, a memory 24, and an interface circuit 25. They are interconnected by a bus. The timer circuit 26 has a time-out error when receiving data transmitted from the CCD control unit 4 (described later).
Signal TFLG, which is set to “L” by a set signal SET output from the output port 22g of the I / O port 22 and becomes “H” after a certain period of time, is input to the input port of the I / O port 22. Output to 22h. I / O port 22 is transmitted output port 22a for supplying a power supply voltage V _CC to the CCD controller 4, the output port 22b for sending a signal RESET for resetting the CCD control unit 4, the CCD controller 4 Signal START, SYNC, DAT
Ports 22c, 22d, 22e and 22 for inputting A1 and DATA2
f, an output port 22g for outputting a set signal SET for setting the timer circuit 26, and the signal T from the timer circuit 26
Each input / output port of the input port 22h for receiving FLG is provided. The CPU 23 receives the CCD input by the I / O port 22.
The distance and reliability are calculated based on the data, and the result is stored in the memory 24. The interface circuit 25 includes a CP
This is a circuit for transmitting the result calculated by U23 to the host computer via the interface bus 8.

An outline of an algorithm for calculating an average distance to the measured object 2 in the automatic distance measuring device having the above configuration will be described with reference to a flowchart of FIG.

In Step S _1, carried out after power setting of the input and output mode of the I / O port 22 of the operation unit 6, the initialization of the interface circuit 25, and the initial setting such as setting the address of the interface bus 8.

In step S _2, waiting waiting for input of the measurement start trigger signal, the trigger signal is input goes to step S _3.

In step S _3, performs the initial operation of the automatic distance measuring device (hereinafter, referred to as the device). That is, the output port 22a
To the CCD controller 4 to turn on the power, and then reset the reset signal RESET output from the output port 22b to “H”.
Then, the CCD controller 4 is initialized and the CCD 3 is initialized. After that, the CCD 3 performs photoelectric conversion, and the CCD controller 4
Then, the CCD data composed of N pixel data corresponding to the number of pixels of the CCD 3 is transmitted to the arithmetic unit 6 together with the AGC data.

In Step S _4, and inputs the respective data outputted from the CCD control unit 4.

In step S _5, checks the hardware, the process proceeds to step S ₆ if there is abnormal, the host computer 7
To inform.

In step S _7, if there is no abnormality in the hardware, based on the CDD data entered correctly executes the processing of the original calculation of the distance to the object to be measured 2 as described below.

In step S _8, it is calculated by a method described later reliability of distance value obtained in step S _7.

In step S _9, calculating the m most recent distance values determined in step S _7, the average distance value by performing an average processing described later by the m reliability obtained in step S _8.

As described above, it is necessary to output the data of the average distance value obtained by performing the averaging process at an arbitrary timing in response to a request from the host computer 7. Therefore, in this embodiment, the following is performed.

In step S _10, the respective data obtained by one of average distance calculation loop, once memory in the arithmetic unit 6
Write to 24 RAMs and update data before exceeding m.

In step S _11, the host computer 7, when there is an output request via the interface bus 8, to interrupt the above CPU 23, the process proceeds to the output subroutine. After the data output is completed, the process returns to the original main routine. Thereby, the host computer 7
When an output request is made, the data of the latest average distance value stored in the RAM 24 can be immediately obtained.

Hereinafter, main steps in the algorithm of the average distance calculation will be described in detail.

Check FIG. 7 step S _4, the data in the S ₅ inputs and hardware will be described in detail with reference to the flowchart of Figure 8. The arithmetic unit 6 monitors the start signal START output by the CCD control unit 4. Here, the start signal START is a signal that notifies the start of data transmission at the level “L”. After receiving the start signal START,
The calculation unit 6 starts inputting data.

Step T _1, a timer is set and outputs a set signal SET to the timer circuit 26.

In step T _2, reads the start signal START.

Step T _3, read start signal START determines whether a level "L". As a result, when it is determined that "L", the process proceeds to step T _6.

Step T _4, at step T _3, when the start signal START is judged not to be "L", the timer circuit 2
Read the signal TFLG output from 6.

Step T _5, determines whether or not the signal TFLG is "H", determines whether a sufficiently long period set by the timer circuit 26 has elapsed. As a result, if "H" is regarded as an abnormal start signal START of the disconnection or the CCD controller 4 of the line, the process proceeds to step T ₅₁ of the abnormality processing routine.
Also, if the signal TFLG is "L" back to step T _2, the signal TFLG from step T ₂ start signal START routine through Step T ₅ becomes "L" or the timer circuit 26, "H" Repeat until. Here, the start signal START
The reason why the hardware is considered to be abnormal when the signal and the signal TFLG are both “H” is as follows. As shown in FIG. 6, since the start signal START has been pulled up to the level “H”, there is a disconnection of the bus line 5 between the CCD control unit 4 and the arithmetic unit 6 or an abnormality of the CCD control unit 4. In this case, the start signal START does not go to the level “L”. Therefore, even if the signal TFLG becomes “H” after a sufficiently long time has elapsed since the output of the set signal SET, the fact that the start signal START does not become “L” means that the start line is disconnected or the CCD controller 4 is determined to be abnormal.

Step T _6, the set signal SET to the timer circuit 26 again
And set the timer.

In step T _7, reads the start signal START.

Step T _8, determines whether or not the start signal START returns to "H" (see FIG. 5). As a result, the process proceeds to step T ₁₁ If "H", the process proceeds to the input processing of the CCD data and AGC data is output in synchronization with the synchronization signal SYNC.

In step T _9, when the remains of the start signal START is "L", reads the signal TFLG.

In step T _10, determines whether or not the signal TFLG is "H", if the result is "H", the other words, if not after a certain period of time to the start signal START is "H", the internal bus line It is determined that only the CCD control unit 4 is abnormal irrespective of the presence / absence of the disconnection of 5, and the processing shifts to the abnormality processing routine. In the case of the signal TFLG is "L" back to step T _7.

In step T _11, the total number of data to be input, ie, C
A total of N + 1 pixel data numbers of CD3 and one AGC data are set in the register BYTC.

In step T _12, sets the input number of cycles required to enter a single data register BITC. In the case of this embodiment, one data is composed of 8 bits as described above, and the number of input bits in one cycle is an even number or an odd number of 2 bits, so the number of input cycles is 4.

Step T ₁₃ through T _15, the start signal similarly to the above STAR
The line T is checked for disconnection, and if the start signal START is at level "L", it is determined that the CCD control unit 4 is abnormal, and the process proceeds to the abnormality processing routine.

Monitors the synchronization signal SYNC in step T _16, T _17, waits for the synchronization signal SYNC is at a level "L". At that time, similarly to the above, to check the time-out error at the step T ₁₃ through T _19, the presence or absence of disconnection of the synchronizing signal SYNC line or CC
The D control unit 4 is checked for an abnormality, and if there is an abnormality, the process proceeds to an abnormality processing routine.

At steps T ₂₁ and T ₂₂ , the level of the synchronizing signal SYNC goes to “L” in step T ₁₇ , followed by the level “H”.
Wait to become. At that time, as described above, step T
In ₂₀ through T _24, checks the timeout error to check for abnormality of the CCD control unit 4, if there is an abnormality, the process proceeds to abnormality processing routine.

In step T _25, T _26, in step T _22, the synchronization signal SYNC is made to level "H", the data line 1, the data signal transmitted by the data line 2 DATA1, DATA2 port 2
Input from 2e, 22f. At this time, even-numbered bits in each 1-byte data, which are pixel data or AGC data, are sequentially input from the data line 1, and odd-numbered bits are input from the data line 2. The bit information is sequentially stored in the register from the upper bit side as shown in FIG.

Thus, at Step T ₁₃ through T _26, while checking the abnormality of the device, the input of the odd or even 2 bits of each data is completed.

In step T _27, T _28, Step T ₁₃ ~ Step T ₂₆
Is repeated until the contents of the register BITC becomes "0", that is, four times in this embodiment, whereby 1-byte data is input (see FIG. 9).

In step T _29, stores one byte of data input to the memory 24 in the arithmetic unit 6.

In step T _30, T _31, Step T ₁₂ ~ Step T ₂₉
Is repeated until the contents of the register BYTC becomes "0", that is, N + 1 times, thereby completing the input of 1CCD data.

Next, one CCD composed of the obtained N pixel data
The data line 1 and the data line 2 are checked for disconnection from the data, and the peak value of the CCD data used for calculating the reliability of the measured value described later is selected. The disconnection check of the data line is performed by the following method. That is, data line 1
Even bit b _6, b ₄ of the pixel data, b _2, b ₀ is transmitting data signals DATA1 consisting also data line 2 data consisting of odd bits _{b 7, b 5, b 3} , b 1 Outputs the signal DATA2. Each of the data lines 1 and 2 is pulled down to the ground as shown in FIG. 6, and when the data lines 1 and 2 are disconnected, all the outputs become "L". That is, even if one of the obtained N pieces of pixel data has all the even-numbered bits of zero,
It is determined that the data line 1 is disconnected. Similarly, if all the odd bits are zero, it is determined that the data line 2 is disconnected. The above processing will be described with reference to the flowchart of FIG.

In step T _32, sets the number N of the pixel data in the one CCD data.

In step T _33, clears the memory area for storing the peak value "peak" of the CCD data.

In step T _34, T _35, clears the DATA2FLG for determining a flag DATA1FLG for determining the presence or absence of disconnection of the data line 1, the presence or absence of disconnection of the data line 2.

In step _T36 , the pixel data value D (I) is compared with the peak value "peak" stored in the memory.

In step _T37 , if the value D (I) of the pixel data is larger than the peak value "peak", the peak value "peak" stored in the memory is updated; otherwise, step _T37 is skipped.

In step T _38, the pixel data D (I) and the signal "0101010
AND with "1" and set the result to "EVN".

In step T _39, the "EVN" determines whether zero.

In step T _40, will be the "EVN" is output even bit if not zero, is set to 1 in the flag DATA1FLG indicating that the data line 1 is not broken. Further, "EVN" is in the case of zero skips step T _40.

Step T ₄₁ through T _43, the same processing is performed for the odd bits.

In steps _T44 and _T45 , the above steps _T36 to _T43
Is repeated for N data. Therefore,
When the even bits of all pixel data are zero, the flag DATA1FLG remains cleared to "0".

In step T _46, the flag DATA1FLG it is determined whether or not "1", "1" unless the data line 1 proceeds to abnormality processing routine is determined to be disconnected.

In step T _47, similarly flag DATA2FLG it is determined whether or not "1", the process proceeds to "1" unless the abnormality processing routine.

In the abnormality processing routine, the following processing is executed.

In step _T51 , the service request flag is written in the status byte together with the coded data of the cause of the abnormality and the service request flag.

In step _T52 , a service request is made to the host computer 7, and the occurrence of an abnormality is notified to the host computer 7.

In step T _53, the host computer 7 detects the service request reads the status byte of each device performing serial polling, by examining the service request flag, one connected to the interface bus 8 or the plurality of devices Find out which device made the service request. Then, by reading the status byte of the device, the host computer 7, knowing that the device makes a service request, the same know any abnormality has occurred by the contents of the status byte simultaneously, the flow proceeds to step T ₅₄ .

In step _T54 , the apparatus stops the measurement.

As described above, the arithmetic unit 6 receives each of the signals START, SYNC, DATA1 and DATA2 output from the CCD control unit 4,
It monitors hardware failures and abnormalities, and notifies the host computer 7 when the occurrence of a failure or abnormality is detected. Therefore, there is no need for the operator to confirm the occurrence of the failure or abnormality on site, and Operation becomes possible. Further, by checking the service request flag, it is possible to know what kind of failure or abnormality has occurred, and it is possible to reduce work time and work contents.

The data input and the hardware check are completed as described above.

The calculation of the original distance to the measured object 2 in step S7 in FIG. ₇ will be described in detail with reference to the flowchart in FIG.

In step T _61, picks the pixel data string from consecutive pixel rows on CCD3 corresponding to two image formed on the CCD3, referred to it as a reference part. The reference portion includes the reference portion in a state where the reference portion is shifted by a certain number of pixel data, and includes more pixel data than the reference portion. Based on each of the pixel data from the reference unit and the reference unit, a difference process is performed on each of the pixels as follows.

Now, assume that the reference part of the CCD data is D'l (i '), the reference part is D'r (j'), and the number is N'l, N'r. Further, the reference portion D′ l (i ′) and the reference portion D′ r
(J ′) and the data obtained by the difference processing (hereinafter referred to as difference data) are Dl (i) and Dr, respectively.
(J) and the numbers are Nl and Nr, Dl (i), Dr
(J) is Dl (i) = D'l (i ')-D'l (i' + k) Dr (j) = D'r (j ')-D'r (j' + k) where k : Constant, 1 ≦ i ≦ Nl = N′l−k, 1 ≦ j ≦ N′rk. The image interval is calculated using the difference data thus obtained.

In step T _62, calculates a contrast value indicating the contrast of the image on the CCD 3. The contrast value is calculated by the following equation.

In step _T63 , a correlation operation is performed to obtain each correlation value. This correlation operation is an operation for calculating an image interval between two images formed on the CCD 3, and is performed as follows. That is, as shown in FIG. 11, the reference portion is sequentially shifted from the same position as the reference portion with respect to the reference portion,
At each shift position, a correlation value between the pixel data on the reference portion and the pixel data on the reference portion is determined. The correlation value is calculated by the following equation.

However, 1 ≦ I ≦ Nr-Nl + 1 I: the number of shift pixels step T _64, among the correlation values HF (I), to calculate the minimum correlation value HFN.

In step T _65, obtains a value IM of the number of shift pixels I to give the minimum correlation value HFN, and image interval IM with the IM.
In this way, the image interval IM with the accuracy of one pixel pitch of CCD3
Can be requested.

In step T _66, by performing an interpolation operation with respect to IM value obtained in step T _65, further calculates a high image distance XM accuracy. This is because HFN = HF (IM) and the correlation values HFM = HF (IM-1) and HFO = HF (IM) before and after I (that is, IM) at that time.
Calculated by the following equation using the three values of +1) (see FIG. 12).

When HFM≤HFO, XM = IM + 1/2 (HFM-HFO) / (HFM-HEN) When HFM <HFO, XM = IM + 1/2 (HFM-HFO) / (HFO-HFN) By multiplying the XM by the pitch of the CCD3 pixels, the actual CCD image interval can be obtained. The image interval and the distance to the measured object 2 obtained as described above are associated with each other in a relationship unique to the optical system 1 (the
FIG. 13 (c)).

In step T _67, the actual image distance obtained in step T _66, as follows determine the distance to the object to be measured 2. That is, when converting from the image interval to the distance, first, the amount of defocus (the amount of defocus) is once calculated from the image interval.
Is calculated, and the distance is calculated from the defocus amount. As shown in FIG. 13A, the relationship between the image interval and the defocus amount is such that the linearity is degraded as the image interval increases, so that the curve is divided into zones and approximated by a straight line. The relationship between the image interval and the defocus amount is stored in the memory 24 as a table, and the defocus amount is obtained from the image interval using this table, and further, the distance value is obtained. The original distance value calculation is completed as described above.

We describe a method of calculating the reliability in step S ₈ of FIG. 7. The measurement result of the original distance value in this embodiment is:
The reliability depends on the following factors. That is, the brightness of the measured object, the contrast of the measured object, YM /
Con value (described later) and original distance value. Each of the above factors is evaluated by a flag that assigns a value of 0 to n according to the degree (the smaller the value of each flag, the lower the measurement accuracy). In addition, the total reliability of one CCD data is calculated based on the combination of the flags for each factor described above.
The flag for each factor and the overall reliability are transmitted to the host computer 7, and when the overall reliability falls below a certain value, the host computer 7 checks the flag for each factor to perform processing according to the situation. Operation can be performed, and unattended operation becomes possible. The meaning of the flag for each of the above factors and an example of calculating the total reliability are shown below.

When the luminance of the measured object 2 is reduced, the output from the pixel corresponding to the dark part of the image on the CCD 3 is affected, and the accuracy of the correlation calculation is deteriorated. Therefore, the measured object 2
., N1 in order from the lowest, a luminance flag is provided. Measured object 2
Is determined based on the peak values of the AGC data and the CCD output data. AGC data is stored in the AGC circuit 17 as described above.
Represents how much gain was applied at low luminance, and the larger the value, the lower the luminance of the measured object 2 is determined. Further, when the gain reaches the maximum value, when the luminance of the measured object 2 further decreases,
Steps T ₃₆ and T ₃₇ in FIG. 8B of the CCD output data
The peak value obtained in the step (1) decreases. Therefore, the luminance of the measured object 2 is determined by the AGC data and the CCD output data peak value.

Contrast of the object to be measured 2 is checked by the contrast value Con obtained in step T ₆₂ of the original distance flowchart of the calculation of FIG. 10. If the contrast value Con is low, the correlation value HF obtained by the correlation calculation step T ₆₃
(I) becomes lower as a whole, and HFN, HFO, HFM in FIG.
Is small, and the accuracy of the interpolation calculation is reduced. Therefore, a contrast flag is provided which is obtained by determining the contrast value HF (I) of the measured object 2 in n stages of 0, 1,... N2 in ascending order.

The YM / Con value will be described. Where the YM value is 12
As shown in the figure, the minimum correlation value YM obtained by interpolation = HF
(XM) is a value normalized by the contrast value. If this value is large, the slope of the interpolation straight line is gentle as shown in Fig. 14, and small fluctuations in the values of HFN, HFM, and HFO are It greatly affects the accuracy of interpolation calculation. Therefore, YM / Con
A YM / Con flag obtained by determining values in n stages of 0, 1,... N3 in order from the highest value is provided.

Regarding the original distance value, as is clear from FIG. 13 (c), the smaller the image interval is, the larger the variation of the original distance value with respect to the image interval is, and the lower the accuracy is. Therefore, a distance value flag is provided which is obtained by determining in n stages of 0, 1,... N4 in descending order of the original distance value according to the original distance value. When the distance exceeds the measurable range, the distance value flag is set to 0.

The overall reliability of the original distance value calculated from the combination of the flags for each factor calculated as described above is obtained as a value of 0 to n. FIG. 15 shows an example in which the flag for each factor is determined in three stages of 0, 1, and 2 and the overall reliability is determined in three stages of 0, 1, and 2.

Described averaging process in step S ₉ of FIG. 7. Since the relationship between the image interval and the distance is non-linear in the averaging process as described above, the averaging process is first performed on the image interval, and then the average image interval is converted into a distance value to obtain an average distance value. The image interval averaging process is performed by stocking the obtained image interval values from the same measured object 2 for m times, discarding the oldest data every time a new image interval value is obtained, and adding a new image interval. Of data is averaged. FIG. 16 shows a flow of calculating the average distance value.

In step T _71, in order to eliminate the large value of the variation in the m data d (1) ~d (m) , the simple average or other from the m data d (1) ~d (m) select the reference value d ₀ by means.

In step _T72 , the number m of data is set.

In step T _73, sets the margin Δd in the reference value d _0, in step T _73, T _74, the input data d and (i) d ₀ ± Δ
Compare d.

In step _T75 , if d (i) falls within the range of d ₀ ± Δd, ₁ is set to the variation flag f ₁ (i).

If d (i) does not fall within the range of d ₀ ± Δd at step T ₇₆ , f ₁ (i) = 0.

In steps _T77 and _T78 , i = 0, ie, m
Repeatedly performing the step T ₇₃ ~T ₇₈ for the pieces of data.

In step T _79, overall reliability of the original distance values determined in the manner described above for each input data d (i) (0~n
Is the reliability value f ₂ (i), and f ₁ (i), f ₂ (i) and d (i) Is calculated.

In step T _80, further Ask for.

In step _T81 , these are substituted into the following equation to calculate an average distance value.

In this way, the influence of data having low reliability or data having large variation with respect to other data is reduced by the weighted averaging process in which data having high reliability and small variation with respect to other data are weighted. An average distance value with high accuracy is obtained.

In step _T82 , the reliability of the average distance value obtained by the averaging process is obtained. That it is, calculated in step T ₈₀ The larger the value, the higher the reliability of the data and the smaller the variation with respect to other data.
This value is used as the reliability of the average distance.

In the above embodiment, a weighted averaging process is performed based on the calculated original distance value and its reliability to obtain a measurement value closer to the true value, but further, to obtain a stable measurement value An embodiment having a calibration function is shown. As shown in FIG. 17, the main lens 1 'true distance from measures the distance to the object of l _0, the original distance value l based on the obtained image interval XM' when give (in dashed lines By adding an offset value -ΔXM of the relationship between the image interval and the distance indicated by the solid line to this XM and correcting the measured value to ₁₀ , accurate measurement can be performed. This offset value can be input from the outside by a DIP (dual in-line package) switch or the like, and the calibration can be automatically performed by providing a calibration routine in the main routine of the distance value calculation. .

In the above embodiment, the distance measurement is performed by one set of the optical system 1. However, as shown in FIG.
A wide range of distance measurement can be performed using a plurality of main lenses 1 'having different focal lengths and different distances from the main lens 1' to the planned focal plane 9. In this embodiment, a table representing the relationship between the image interval and the distance according to the plurality of main lenses 1 'is stored in the memory 24 by the number of the main lenses 1'. By replacing the table for image spacing-distance conversion with the table corresponding to the main lens 1 'using an external switch or the like when the' is replaced, the distance can be measured over a wide range.

In the above embodiment, the optical system 1 is used to optically measure the distance. However, the present invention is not limited to this, and the length of a straight line connecting two points of a triangle and two angles on both sides thereof are measured to obtain a distance to another point. There is no problem if a distance measuring method or the like that uses a method is used.

<Effects of the Invention> As is clear from the above, the automatic distance measuring device of the present invention includes the distance measuring means, the reliability calculating means, and the connecting means, and measures the distance to the object to be measured. By obtaining the reliability of the distance and connecting the connection means to another device, a signal corresponding to the measured distance and a signal representing the reliability of the distance are output to the outside of the device. Therefore, the other device can know the distance value to the measured object and the reliability together.

Further, if the signal corresponding to the measured distance is obtained by calculating a weighted average value based on the reliability for a plurality of distances obtained by the distance measuring means, the measured distance having a low reliability is obtained. The influence can be reduced and a more accurate measurement value can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of an automatic distance measuring device of the present invention, FIG. 2 is a block diagram of an optical system in FIG. 1,
FIG. 3 is an explanatory view of the principle of distance measurement in FIG.
FIG. 5 is a block diagram of the CCD control unit in FIG. 1, and FIG.
FIG. 6 is a block diagram of an arithmetic unit in FIG. 1, FIG. 7 is a flowchart of an average distance calculation, and FIGS. 8 (a) and (b) are data input and hardware. FIG. 9 is a flowchart of a wear check, FIG. 9 is an explanatory diagram of a state of storing pixel data in a register, FIG. 10 is a flowchart of an original distance value calculation, FIG. 11 is an explanatory diagram of an image interval calculation method, and FIG. Fig. 13 shows the image spacing,
Diagram showing correspondence between defocus amount and distance, Fig. 14 shows YM / Con
FIG. 15 is a diagram showing an example of determining the overall reliability, and FIG.
FIG. 17 is a flowchart for calculating an average distance, FIG. 17 is an explanatory diagram of measurement value correction, FIG. 18 is a diagram showing a relationship between image distances and distances when a plurality of lenses are used, and FIG. It is a figure showing a measuring device. DESCRIPTION OF SYMBOLS 1 ... Optical system, 1 '... Main lens, 2 ... Object to be measured, 3 ... CCD linear image sensor, 4 ... CCD control part, 6
Calculation unit, 7 Host computer, 8 Interface bus, 12a, 12b Re-imaging lens, 16 CCD drive circuit, 17
AGC circuit, 18 A / D converter, 20 I / O circuit, 22
... I / O port, 23 ... CPU, 24 ... memory, 25 ... interface circuit, 26 ... timer circuit.

Claims (4)

(57) [Claims] An optical system; a photoelectric conversion element for receiving an object to be measured via the optical system; a distance measuring means for obtaining a distance to the object to be measured based on an output of the photoelectric conversion element; A reliability calculating means for obtaining the reliability of the measured distance based on the output of the device, and a signal corresponding to the measured distance and a signal representing the reliability of the distance, which are connected to another device. An automatic distance measuring device comprising a connecting means for outputting to the outside. 2. The automatic distance calculation method according to claim 1, wherein said reliability calculation means obtains the reliability of said measured distance based on said measured distance itself. measuring device. 3. The signal according to claim 1, wherein the signal corresponding to the measured distance is calculated based on the measured distance and the reliability of the measured distance. 3. The automatic distance measuring device according to claim 1 or 2. 4. A signal corresponding to the measured distance is calculated based on the plurality of distances obtained by the distance measuring means and the reliability obtained by the reliability calculating means corresponding to the plurality of distances. 4. The automatic distance measuring apparatus according to claim 1, wherein the signal is a signal representing a weighted average value of the plurality of distances.