JPH04277709A - Range-finder for passive type automatic focusing device - Google Patents

Abstract

PURPOSE:To offer a range-finder for an automatic focusing device which can rapidly measure a distance to an object with high accuracy and whose number of parts is reduced as far as possible in order to surely focus even when plural objects exist. CONSTITUTION:One line sensor is divided to three parts 10b-30b, each part acquisitions the object, the output signal of the line sensor with respect to brightness distribution is digitally converted, a zero cross point for secondary difference is arithmetically calculated and detected, and data concerning the zero cross are stored by complying with the three parts 10b-30b, and the stored data are read out one by one. As to a zero cross behavior with respect to these three parts of the line sensor, one of the three parts is set as a reference and the other two are appropriately moved, so that the identified point of the zero cross behavior of the three parts 10b-30b is detected, and the distance to the object can be measured based on the moved amount of the zero cross behavior of the two parts at the time of identifying.

Description

[Detailed description of the invention]

0001

[Industrial Application Field] This invention is a passive autofocus device that receives light from a subject, measures the distance to the subject, and adjusts the focus of the photographic lens based on the measurement results. The present invention relates to a distance measuring device for measuring the distance to a subject.

0002

[Prior Art] An autofocus device is a device that automatically measures the shooting distance of a camera, etc., and adjusts the shooting lens based on the distance measurement result to focus. Now you can enjoy shooting more easily. Various types of autofocus devices have been developed, but the main one is a distance measurement method using triangulation. Among the triangulation methods, there is a passive type that uses a light-receiving sensor installed in the camera to receive light from the subject and measure the shooting distance.

There are two types of passive distance measuring devices of this type.
Some devices are equipped with several light-receiving sensors. With this distance measuring device consisting of two light-receiving sensors, the measurement result when two objects exist means that the objects can exist in two ways, making it impossible to measure the distance reliably and making it difficult to focus. There is a risk that the image will be shifted.

[0004] Therefore, in order to perform reliable distance measurement and obtain a clear image, the applicant of the present application has already proposed a distance measurement mechanism consisting of three light-receiving element rows (Japanese Patent Application No. 1999-1-1).
No. 77382). Figure 12 shows the principle of distance measurement using this distance measurement mechanism.
This will be explained based on FIG. The distance measuring mechanism consists of a reference light receiving sensor 1, a first light receiving sensor 2, and a second light receiving sensor 3, and these light receiving sensors 1, 2, and 3 have imaging lenses 1a, 2a, and 3a, and light receiving element arrays 1b and 2b, respectively. , 3b, and the object image is transmitted through the imaging lenses 1a, 2a, 3a and formed on the light receiving element arrays 1b, 2b, 3b. Further, FIG. 12 shows a case where one subject P exists. Then, x0 is the amount of displacement of the output signal P0 regarding the luminance distribution of the subject P detected by the reference light receiving element array 1b from the optical axis T0 of the reference light receiving sensor 1, and the subject detected by the first light receiving element array 2b is The amount of displacement of the output signal P1 regarding the brightness distribution of P from the optical axis T1 of the first light receiving sensor 2 is x1, and the second light reception of the output signal P2 regarding the brightness distribution of the subject P detected by the second light receiving element array 3b Let x2 be the amount of displacement of the sensor 3 from the optical axis T2. These displacement amounts x0, x1
, x2 represents a phase difference regarding the brightness distribution of the subject image detected by the light receiving element arrays 1b, 2b, and 3b. Then, the distance between the optical axes T0, T1, and T2 is B,
Imaging lenses 1a, 2a, 3a and light receiving element arrays 1b, 2b,
3b and the light receiving surface is A, and the imaging lenses 1a, 2a, 3
If the distance from a to the subject P is Lp, and the distance from the optical axis T0 to the subject P is X, then from the principle of triangulation,

[Formula 1] X=x0*Lp/A. Also, including the sign of the direction in which the image of the output signal appeared with respect to the optical axis T0,

[Math. 2]-x1={(B-X)/Lp}*A

[Math 3]x
2={(B+X)/Lp}*A. These two formulas,
By substituting formula 1 into each of formula 3, we get

[Formula 4] x1=-{(B/Lp)*A}+x0

[Math 5]
x2=(B/Lp)*A+x0.

Comparing Equation 4 and Equation 5, x1, x
2 are each based on x0,

[Equation 6] It can be seen that there is a deviation of (B/Lp)*A=Xp. Therefore, by finding this Xp,

[Formula 7] Lp=A*B/Xp can be calculated.

FIG. 13 shows the operation for determining the above Xp.
The explanation will be based on. (a) shows output signals related to the brightness distribution of the subject images of the light receiving element arrays 1b, 2b, and 3b that have received light from the two subjects P and Q, and the reference output signal P0.
, compared with Q0, from the state shown in (a) to (b)
If the waveforms of the output signals P1 and P2 are shifted until the output signals P0, P1, and P2 match as shown in FIG. 2, the amount of shift becomes the above Xp. In other words, since the amounts of shift of P1 and P2 are equal at this time, when the output signal of the light receiving element array 2b and the output signal of the light receiving element array 3b are shifted by the same distance and the waveforms of the three signals match, these 3
The waveforms of the two signals become information regarding the same subject P. Next, as shown in (C), if the output signals Q1 and Q2 are shifted until they match the output signal Q0, the amount of shift becomes Xq.

[0007] The above Xp and X obtained as described above
The distance Lp from q to the objects P and Q using the above formula 7
, Lq are required.

[0008]

However, when the conventional distance measuring operation described above is carried out according to the principle, the correlation between the output signal of the reference light receiving element array 1b and the output signal of the first light receiving element array 2b is calculated, Next, the correlation between the output signal of the reference light-receiving element row 1b and the output signal of the second light-receiving element row 3b is calculated, and the waveforms of these reference light-receiving element row 1b, first light-receiving element row 2b, and second light-receiving element row 3b are calculated. , the number of correlation calculations increases and the signal processing time increases. Therefore, the time required for distance measurement increases, and if the subject is dynamic, the subject may be photographed out of focus, resulting in an unclear image.

Therefore, the present invention has three light receiving sensors, can perform signal processing in a short time, can obtain as clear an image as possible, and has a distance measuring system with a minimum number of parts. The purpose is to provide equipment.

0010

[Means for Solving the Problems] For the above purpose, a distance measuring device for a passive autofocus device according to the present invention divides one line sensor into three parts and displays a subject image in each part. a light-receiving sensor that captures the brightness distribution of a subject, which is composed of three imaging lenses to form an image; a second-order difference calculation circuit that digitally converts the output signal of the line sensor and calculates a second-order difference between the digital values; a zero-crossing detection circuit that detects a zero-crossing point of the output signal of the second-order difference calculation circuit; a zero-crossing memory circuit that separately stores the zero-crossing behavior signal obtained by the zero-crossing behavior signal for each of the three parts of the line sensor; a coincidence detection circuit that compares the zero-crossing behavior signals stored in each of the zero-crossing memory circuits and detects a coincidence; The zero-crossing behavior signals obtained from the other two parts are sequentially slid relative to the zero-crossing behavior signal obtained from the above, and the matching of these zero-crossing behavior signals is detected by the coincidence detection circuit, and the distance from the sliding amount to the subject is detected. It is characterized by calculating the distance between.

[0011]

[Operation] An output voltage corresponding to the brightness distribution of the subject is obtained by the array of light receiving elements constituting the light receiving sensor, and the quadratic difference distribution of this output voltage behaves around the zero level. The zero-crossing points of this behavior are the same for the three parts of the line sensor, with appropriate deviations from a predetermined reference part, for the luminance distribution regarding the same part of the object.

[0012] This amount of deviation can be obtained as a sliding amount by sliding and detecting the signal waveform of zero-crossing behavior in the coincidence detection circuit.

[0013] Then, the distance to the subject can be calculated from this slide amount by triangulation.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The distance measuring device for an autofocus device according to the present invention will be specifically explained below based on the illustrated embodiments.

The light-receiving sensors 10, 20, and 30 are constructed by combining one line sensor consisting of a row of light-receiving elements in which an appropriate number of pixels are arranged in parallel, and three imaging lenses, as shown in FIG. There are three imaging lenses 10 on the front of the camera.
a, 20a, and 30a are disposed, and light emitted from the subject passes through these imaging lenses 10a, 20a, and 30a, and forms an image on a corresponding portion of the line sensor 8 disposed at the rear. Therefore, the line sensor 8 is divided into three parts, each of which is a central part 10b of the line sensor, a right part 20b of the line sensor, and a left part 30b of the line sensor. In addition, these light receiving sensors 10, 20,
30 are the center sensor 10, the right sensor 20, and the left sensor 30, respectively, and the optical axes 20c and 30c of the right sensor 20 and the left sensor 30 are located at symmetrical positions with respect to the optical axis 10c of the center sensor 10. It is in.

As shown in FIG. 1, a drive signal from a sensor driver 11 is input to the line sensor 8, and the line sensor 8 starts capturing light from an object based on the drive signal. Further, the sensor driver 11 is connected to the control circuit 40 through a drive control signal line 40a, and is controlled by a drive control signal output from the control circuit 40.

On the other hand, the output terminal of the line sensor 8 has
As shown in FIG. 1, the A/D converter 9 is connected to
The output signal of the line sensor 8 is digitally converted. A second-order difference calculation circuit 12 is connected to the output side of the A/D converter 9, and the second-order difference calculation circuit 12 converts the luminance distribution signal of the object obtained by the line sensor 8 into two.
Compute the order difference. As shown in FIG. 3, this secondary difference calculation circuit 12 sequentially stores the output signal AD of the A/D converter 9 in storage circuits 12a and 12b using D flip-flops in synchronization with the clock pulse φ2.

[Formula 8] DIFF(n) =AD(n-2)-2*AD(n-1
)+AD(n) to obtain the second-order difference. It is also determined whether the value of DIFF at this time has a positive or negative sign (SIGN).

The output signal of the second-order difference calculation circuit 12 is
As shown in FIG. 1, the signal is input to a zero-cross detection circuit 13, which detects the zero-cross point of the second-order difference obtained by the second-order difference calculation circuit 12. As shown in FIG. 4, the sign output SIGN of the second-order difference calculation circuit 12 is input to the D flip-flop 13a of this zero-cross detection circuit 13, and the sign output SIGN and the inverted Q terminal of the D flip-flop 13a are ANDed. The output signal of the code output SIGN, which is inputted to the circuit 13b and passed through the inverter 13c, and the Q terminal of the flip-flop 13a are inputted to the AND circuit 13d.

As shown in the time chart of FIG. 5, the output signal Vin of the line sensor 8 is input to the A/D converter 9 and digitally converted to obtain the output signal AD. When this output signal AD is input to the secondary difference calculation circuit 12, a signal DIFF and a code signal SIGN are obtained in synchronization with the clock pulse φ2. This code signal SIGN
When P-ZERO is input to the zero-cross detection circuit 13, the AND circuit 13b outputs a pulse P-ZERO that rises when the code signal SIGN changes from L to H as a zero-cross signal in synchronization with the clock pulse φ2. AN
The D circuit 13d outputs a pulse N-ZERO that rises when the code signal SIGN changes from H to L as a zero-cross signal. That is, pulse P-ZER
O rises when the secondary difference signal DIFF changes from positive to negative and crosses zero, and pulse N-ZE
RO becomes a signal that rises when the secondary difference signal DIFF changes from negative to positive and crosses zero.

The signal waveform of the zero-crossing behavior obtained by the zero-crossing detection circuit 13 is divided into a portion corresponding to the center portion 10b of the line sensor, a portion corresponding to the right portion 20b of the line sensor, and a portion corresponding to the line sensor 30b. and a zero-cross memory circuit 14, respectively.
24 and 34 and are stored. At this time, the zero crossing behavior is stored in correspondence with the addresses output from the address calculation circuits 25 and 35 in correspondence with the pixel positions in the right part 20b and left part 30b of the line sensor 8, and in the center part 10b of the line sensor 8, Count signal of counter 50 (
COUNTER1). That is, the count signal (COUNTER1) of the first counter 50 is input to the address calculation circuits 25 and 35, and the count signal (COUNTER1) is input to the central memory circuit 14, and the central memory circuit 14 is sequentially incremented. So,

[Equation 9] ADDRESS=COUNTER1 In the right memory circuit 24,

[Formula 10] ADDRESS=COUNTER1 In the left side memory circuit 34,

[Formula 11] According to ADDRESS=COUNTER1, each pixel is stored at a respective address.

The address calculation circuits 25 and 35 also receive a count signal (COUNTER2) of the second counter 60.
), and the second counter 60 and the first counter 50 are counted up and reset based on the output signal of the control circuit 40. This second counter 60 is connected to the zero cross memory circuit 2, as will be described later.
When reading data from 4 and 34, the address is incremented. Further, the address calculation circuit 25,
35 receives address processing information from the control circuit 40, and based on the address processing information, the address calculation circuit 25
, 35 output predetermined write signals and read signals to the zero-cross memory circuits 24, 34.

[0022]The above zero-cross memory circuit 14,
A coincidence detection circuit 70 is connected to the output sides of 24 and 34, and the output side of the coincidence detection circuit 70 is connected to the control circuit 40.

The count signal of the first counter 50 is input to the address port 81 of the data memory circuit 80, and the count signal of the second counter 60 is input to the distance data port 82 of the data memory circuit 80. Furthermore, the count signals of the first counter 50 and the second counter 60 are both input to the control circuit 40. Further, a data memory signal is output from the control circuit 40 to the data memory circuit 80, and address data and distance data are stored in the data memory circuit 80 based on the signal.

Next, based on FIGS. 6 to 9, a procedure for writing and reading luminance information of an object into and from the memory will be explained.

When distance measurement is started, charge is accumulated in the line sensor 8 (step 601), the second counter 60 is reset (step 602), and a readout pixel number counter (not shown) in the control circuit is reset ( Step 603).

Since the line sensor 8 is composed of one line sensor, it is first determined whether reading of the first pixel in the portion related to the left part 30b of the line sensor has started based on the value of the readout pixel number counter (step 60
4), 1 until reading of the first part starts.
Each pixel is output (step 605). Once reading of the first part is started, the first counter 50
is reset (step 606). Then, data corresponding to one pixel on the left side of the line sensor 30b is read out (step 607), and the read data is written into the left zero cross memory circuit 34 (step 60).
8). Note that zero cross detection is performed between step 607 and step 608. Then step 609
The process proceeds to step 610, and it is determined based on the value of the first counter 50 whether reading has been completed for all pixels, and if the reading has not been completed, the process proceeds to step 610, where the first counter 50
After counting up, return to step 607 and perform 1
Pixel reading and writing to the left zero cross memory circuit 34 are performed (step 608). When data is written to the left zero cross memory circuit 34,
Based on the count signal of the first counter 50, an address is designated by the address calculation circuit 35 and stored. At this time, the address to be memorized is designated according to Equation 11 above.

When the reading of all pixels in the left part 30b of the line sensor is completed and the answer is YES in step 609, the process proceeds to step 703 (FIG. 7), where the central part 1 of the line sensor is read out.
Whether reading of the first pixel of the portion related to 0b has started is determined based on the value of the read pixel number counter, and one pixel at a time is outputted until reading of the first portion is started (step 704). Once the reading of the first part is started, the first counter 50 is reset (step 705), and as in steps 607 to 610, the line sensor central part 10b is read out pixel by pixel while zero cross detection is performed. reading (step 706) and central zero cross memory 14
(step 707), and judgment of completion of reading for all pixels of the line sensor central portion 10b based on the value of the first counter 50 (step 708), while counting up the first counter 50 (step 709).
)Repeated. Note that the address to be memorized is specified according to Equation 9.

When the readout of all pixels in the line sensor central portion 10b is completed, the answer is YES in step 708.
If it is determined that
While outputting one pixel at a time (step 804), the start of reading out the first pixel in the part related to the right part 20b of the line sensor is determined by the value of the readout pixel number counter, and if the reading out of the first part has started, the first counter 5
0 is reset (step 805). Then, in the same way as the procedure for the line sensor left part 30b and the line sensor center part 10b, for the line sensor right part 20b, zero cross detection is performed while reading out each pixel (step 806) and writing to the right zero cross memory 24. (Step 807), determination of completion of reading for all pixels of the right side of the line sensor 20b based on the value of the first counter 50 (Step 808), while counting up the first counter 50 (Step 809)
Repeated. When data is written to the right zero cross memory circuit 24, it is stored at an address specified by the address calculation circuit 25 according to Equation 10 based on the count signal of the first counter 50.

If the reading of all pixels of the line sensor 8 is completed and the determination in step 808 is YES, the process proceeds to step 901 (FIG. 9) where the first counter 5
Reset to 0. And zero cross memory circuit 14
, 24, and 34 (step 9)
02), the coincidence detection circuit 70 determines whether or not the data of the center zero-cross memory circuit 14, the right zero-cross memory circuit 24, and the left zero-cross memory circuit 34 match (step 903). If the data match, the process proceeds to step 904, where the value of the count signal (COUNTER1) of the first counter 50 at that time is used as address data, and the value of the counter signal (COUNTER2) of the second counter 60 is used as the distance. Each is written to the data memory circuit 80 as data. If the determination in step 903 is NO, the process proceeds to step 905, where a first check is made as to whether reading of memory data (reference data) corresponding to all valid pixels of the central portion 10b of the line sensor 8 has been completed. It is determined based on the value of the counter 50, and if the process has not been completed, the process proceeds to step 906 to increment the first counter 50, and then returns to step 902 and executes steps up to step 905.

When the reading of the reference data is completed, the process proceeds from step 905 to step 907, in which the data in the right and left zero cross memory circuits 24 and 34 is shifted by a specified amount with respect to the data in the center zero cross memory circuit 14. The second counter 60 checks whether steps 901 to 905 have been executed (shift read).
The judgment is made based on the value of (step 907). If the shift reading is not completed, the second counter 60 is counted up and the process returns to step 901.
Step 905 is repeated from 02. Then, if the shift read is completed, step 9
Proceed to 09.

Reading of memory data from step 901 to step 908 is performed by the first counter 5.
0 and the address arithmetic circuits 25 and 35, the central zero cross memory circuit 14 outputs,

[Formula 12] ADDRESS=COUNTER1 From the right zero cross memory circuit 24,

[Formula 13] ADDRESS=COUNTER1+COUNTER2
From the left zero cross memory circuit 34,

[Formula 14] ADDRESS=COUNTER1+S-COUN
Read according to TER2. Note that S in Equation 14 is a constant. The relationship between the write address and read address at this time will be explained with reference to FIGS. 10 and 11.

FIG. 10A shows a case where the count signal of the second counter 60 is 0 (COUNTER2=0), and at this time, the first counter 50 is incremented from 0 to (W-1) (step 906). , line sensor 8
The data stored in the addresses corresponding to the pixels of each section 10b, 20b, and 30b are compared to detect a match between the data. Therefore, when COUNTER2=0, the addresses of the pixels in the center part 10b of the line sensor and the right part 20b of the line sensor are incremented from 0 to (W-1), and the addresses of the pixels in the left part 30b of the line sensor are incremented from S to (S+W- It is incremented to 1). Next, the second counter 60 is incremented (step 90).
8), the counter signal of the second counter 60 is set to 1 (COU
NTER2=1) while incrementing the first counter 50 from 0 to (W-1) (step 9
06), each part 10b, 20b, 30 of the line sensor 8
The data stored in the address corresponding to each pixel of b are compared to detect a match between the data. Therefore,
When COUNTR2=1, line sensor central part 10
The address is from 0 to (W-1) for line sensor b, from 1 to W for line sensor right part 20b, and from (S-1) to (S+W-) for line sensor left part 30b.
2). That is, the memory data of the right zero-cross memory circuit 24 and the left zero-cross memory circuit 34 are shifted by one pixel with respect to the memory data of the center zero-cross memory circuit 14, and coincidence detection is performed.

Then, while the second counter 60 is incremented (step 908), the coincidence detection is repeated until the count signal of the second counter 60 reaches COUNTER2=S. Note that FIG. 11(a) shows COUNTE
When R=S−1, FIG. 11(b) shows COUNT
This shows the case when R2=S.

That is, zero cross memory circuits 14, 2
The value of the second counter 60 when the memory data Nos. 4 and 34 match with respect to the zero-crossing point corresponds to the deviation amount Xp in Equation 6 above. This amount of deviation is then stored in the data memory circuit 80 as distance data in step 904.

If it is determined in step 907 that the predetermined shift readout has been completed, the process proceeds to step 909;
The distance data written in the data memory circuit 80 in step 904 is output to a photographing lens driving device (not shown), and the photographing lens moves to a predetermined position and focuses on the subject.

[0036]

[Effects of the Invention] As explained above, according to the distance measuring device for a passive autofocus device according to the present invention, the luminance of the subject is captured by the line sensor divided into three parts, and the second-order difference is calculated from the data. and store the zero-crossing data with the zero-crossing point of the second-order difference as a feature point,
Using the zero-crossing data obtained from one of the three parts as a reference, the zero-crossing data obtained from the other two parts are sequentially shifted one pixel at a time, and comparing whether or not the three data match with respect to the zero-crossing point. Since the distance to the subject is calculated using the shift amount when they match as distance data, the calculation processing speed is faster than when distance data is obtained by correlation calculation. Therefore, it is possible to reliably capture a dynamic subject and perform quick focusing. Moreover, since the output signal of the line sensor is A/D converted and the above-mentioned second-order difference is calculated by digital processing, the calculation process is simpler and the processing speed is faster than when analog processing is performed. Can be done.

Moreover, with one line sensor, there are three
Since the circuit is divided into two parts, there is one set each of the second-order difference calculation circuit and the zero-cross detection circuit. For this reason, in a distance measuring device that uses three line sensors,
The number of parts can be reduced compared to a structure in which three sets of secondary difference calculation circuits and zero-cross detection circuits are provided for each line sensor.

In addition, since zero-cross data of second-order differences are compared, distance data can be obtained with high precision because it does not depend on the pattern of object brightness distribution on the line sensor.

[Brief explanation of the drawing]

FIG. 1 is a circuit block diagram of a distance measuring device for a passive autofocus device.

FIG. 2 is a side view showing a schematic structure of a light receiving sensor.

[Figure 3] A/D conversion of line sensor output, Part 2
FIG. 2 is a circuit block diagram of a second-order difference calculation circuit that digitally processes and calculates a second-order difference.

FIG. 4 is a circuit diagram of a zero-cross detection circuit that detects a zero-cross point from a second-order difference signal obtained by a second-order difference calculation circuit.

FIG. 5 is a time chart of output signals of a line sensor, an A/D converter, a second-order difference calculation circuit, and a zero-cross detection circuit.

FIG. 6 is a flowchart showing a procedure for writing data obtained from a line sensor into a zero-cross memory circuit, and relates to the left side of the line sensor.

FIG. 7 is a flowchart showing a procedure for writing data obtained from a line sensor into a zero-cross memory circuit, and relates to the central portion of the line sensor.

FIG. 8 is a flowchart showing a procedure for writing data obtained from a line sensor into a zero-cross memory circuit, and relates to the right side of the line sensor.

FIG. 9 is a flowchart showing a procedure for reading predetermined data from a zero-cross memory circuit in order to detect coincidence of data stored in the zero-cross memory circuit.

FIG. 10 is a diagram illustrating an operating procedure when reading and comparing data stored in a zero-cross memory circuit.

FIG. 11 is a diagram illustrating an operating procedure when reading and comparing data stored in a zero-cross memory circuit.

FIG. 12 is an optical path diagram showing the principle of distance measurement.

FIG. 13 is a diagram for explaining a measurement procedure based on the distance measurement principle, and is a signal diagram regarding the brightness distribution of a subject image detected by a light receiving element array.

[Explanation of symbols]

8 Line sensor 9 A/D converter 10 Light receiving sensor 20 Light receiving sensor 30 Light receiving sensor 10a Imaging lens 20a Imaging lens 30a Imaging lens 10b Line sensor center section 20b Line sensor right section 30b Line sensor left section 11 Sensor driver 12 Secondary Difference calculation circuit 13 Zero cross detection circuit 14, 24, 34 Zero cross memory circuit 25, 35
Address calculation circuit 40 Control circuit 50 First counter 60 Second counter 70 Coincidence detection circuit 80 Data memory circuit

Claims (1)

[Claims] 1. A light-receiving sensor that captures the brightness distribution of a subject, comprising three imaging lenses that divide one line sensor into three parts and form a subject image on each part; Converts the output signal to digital,
A second-order difference calculation circuit that calculates the second-order difference of the digital value, a zero-cross detection circuit that detects the zero-cross point of the output signal of the second-order difference calculation circuit, and a zero-cross behavior signal obtained by the zero-cross detection circuit, It consists of a zero-crossing memory circuit that separately stores the three parts of the line sensor, and a coincidence detection circuit that compares the zero-crossing behavior signals stored in each of the zero-crossing memory circuits and detects their coincidence. One of the three parts of the sensor is used as a reference, and the zero-crossing behavior signals obtained from the other two parts are sequentially slid against the zero-crossing behavior signal obtained from the reference part. A distance measuring device for a passive autofocus device, characterized in that a coincidence of signals is detected by the coincidence detection circuit, and a distance to a subject is calculated from the slide amount.