JPS62178914A - Automatic focus detecting device - Google Patents

Abstract

PURPOSE:To keep in an extremely well-balanced manner a reduction of a hunting operation and a drop of a focus detection accuracy, by constituting the titled device so that a size of a focusing dead zone becomes large as length of a prescribed reset time which has been updated and set becomes long. CONSTITUTION:The titled device is provided with an arithmetic means for storing an integral value of a signal which has been integrated to a prescribed reset time of the previous time by an integral means, updating and setting a prescribed reset time of this time, based on its signal, making an integral means execute an integral operation in accordance with the prescribed reset time which has been updated and set, detecting a focal state from an obtained integral value, and simultaneously, determining a size of a focusing dead zone which can be varied continuously in accordance with length of the prescribed reset time which has been updated and set. In this state, as length of the prescribed reset time which has been updated and set becomes long, a size of the focusing dead zone is made large. In this way, a reduction of a hunting operation and a drop of a focus detection accuracy can be kept in an extremely well-balanced manner.

Description

Detailed Description of the Invention (Field of Application of the Invention) The present invention relates to an automatic focus detection device used in a video camera, etc., particularly an active type automatic focus detection device that projects light onto a subject and receives the reflected light to detect focus. This invention relates to improvements to automatic focusing devices.

(Background of the Invention) Fig. 6 shows a device of this kind that the applicant of the present application has filed an earlier application for (Japanese Patent Application No. 60-80326, etc.). 1 is a light projecting element, 2 is a driving motor 3, and is moved as a driving source. For example, when the photographing lens 2 moves, the light receiving section 4a
, 4b also begins to move (actually, it moves via a cam etc. in conjunction with the rotation of the drive motor 3).
. 5 is an analog switch that is turned on when a high-level timing signal M1 is input from the microcomputer 6; when the switch is on, the light receiving section 4a is turned on;
, 4b both receive reflected light from the subject, and only the light receiving section 4a receives light when it is off. 7 is a sensor amplifier, 8 is a bypass filter for removing DC components (external light components), and 9 is a sum signal (A + B) input at that time or the gain of signal A, depending on the signal input via the inverter 10. In other words, when the timing signal M1 is a high level signal (in this case, the sum signal (
When the timing signal M1 is a low level signal (in this case, the signal A is input), it is output to the next stage at the same level.

) is a gain control circuit that increases the gain to twice the level and outputs it to the next stage, and 11.12 receives timing signals M2.1 from the microcomputer 6. 13 is an analog switch that turns on and off in accordance with a timing signal M4, which is a pulse signal input from the microcomputer 6; 1;
4 is an integrator circuit, 15 is a constant current source that generates a negative constant current i, which has the opposite polarity to the signals A and B, and generates a sum signal (A+B
) or used when inversely integrating the integral value of signal 2A. 16 is a high-level signal sent to the microcomputer 6, which has an internal counting section that counts pulse signals when the inverse integration of the integral value of the sum signal (A+B) or signal 2A is completely completed (when the integral output reaches zero); A comparator 17 that outputs the signal M4 is a drive circuit that drives the light projecting element 1 in accordance with the timing signal M4.

Next, the operation will be explained with reference to FIGS. 7 and 8.

First, the microcomputer 6 outputs a timing signal M to turn on the analog switch 5 for a preset 10 hours. By turning on the analog switch 5 in this way, the light emitted from the light emitting element 1 and reflected light returning from the subject is received by both the light receiving parts 4a and 4b, and the sum signal (A+B) is output from the sensor amplifier 7. Output. Further, the timing signal M1 is sent to the gain control circuit 1 as a low level signal via the inverter 10.
Therefore, the gain control circuit 12 receives the sum signal (A
+B) at the same level as the next analog switch 1
Output to 1. Also, at this time, as can be seen from FIG. 7(a), the microcomputer 6 sends a timing signal M.
2. Since M4 is output, analog switch 1
1 is analog switch 1 according to timing signal M2.
3 are turned on and off according to the timing signal M4, that is, they are turned on in synchronization with the light emission timing of the light projecting element 1.

Therefore, the sum signal (A+B) output from the gain control circuit 9 is sent to the integration circuit 14 through the analog switch 11.13, and is integrated for 10 hours as shown in FIG. Figure step ■).

In this way, the sum signal (A+B) is generated by the 10-hour integration circuit 14.
When the integration of is performed, the microcomputer 6 starts outputting the timing signal M3, and this time the analog switch 1
Turn on 2.

When the analog switch 12 is turned on, inverse integration is started by the constant negative current i flowing from the constant current source 15 (see FIG. 7), and this inverse integration continues until a high level signal is output from the comparator 16. It will be held twice. During this time (time required for inverse integration)
! The counting section counted by t counts the number of pulses generated by the pulse generating section also arranged inside. If the number of pulses at this time is PO, then P is the sum signal (A + B)
An A/D conversion signal (A/D conversion value) corresponding to the integral value of is obtained (step 2).

As described above, the number of pulses p is determined by pre-distance measurement.

is obtained, based on the pulse number P0, the integrating circuit 1
Step 4 sets an integration time suitable for integrating the sum signal (A+B). In other words, as mentioned above, when the distance to the subject and its reflectance are different, the sum signal (A +
When B) is integrated, the integral value fluctuates greatly and may exceed the dynamic range that can be measured depending on the subject conditions at that time. , that is, the sum signal (A
If the integral value of +B) is high, the integral time is shortened, and conversely, if it is low, the integral time is lengthened. Therefore, the following calculation is performed in the microcomputer 6, and the next sum signal (A
+ B) Determine the integration time. However, k is a constant and Tn is the next integration time.

Tn = Tn-% k / PI-1 After calculating the integral time using the previous equation (this time it is calculated from T, =To・ni/Po) (step ■), the microcomputer 6 calculates this T, time timing. Signals M1, M2
, Ma. As a result, the T1 time sum signal (A
+ B) is integrated (step ■), and then the timing signal M3 is output, so inverse integration is performed at a constant current i, and the integrated value of the sum signal (A+B) is A/D converted to the number of pulses PA. ↑8 is obtained (step ■). The number of pulses P^ya6 at this time is stored in the microcomputer 6. After the above operations are completed, the microcomputer 6 again outputs the timing signal M2.

M4 is output (see FIG. 7(a)). In this case, since the timing signal M1 is not output, the analog switch 5 is off, and the gain control circuit 9 doubles the gain of the input signal. mode is being switched. Therefore, in this case, the signal 2A, which is twice the signal A that is received and outputted by the light receiving section 4a, is integrated (step ■), and then the timing signal M3 is outputted as described above, so at a constant current i. Inverse integration is performed and signal 2
The number of pulses P2^ is obtained by A/D converting the integral ti of A (step ■).

Based on the pulse numbers PzA and PAte obtained as described above, the microcomputer 6 performs calculations as shown in FIG. 9, for example, to determine in-focus or out-of-focus (front focus, rear focus). At the same time, for out-of-focus, the pulse width of the automatic focus control signal N (see Fig. 6) to the drive motor 3 is modulated according to the pulse number ratio P2^/P4+5 (see Fig. 6).
PWM). Thereby, the rotational direction and rotational speed of the drive motor 3 are controlled. In this example, as shown in FIG. 9, Pz^/PAAIl is, for example, ro, 8~
If it is within 1.2 J, it is determined that the focus is in focus, and the automatic focus control signal N is not output to the drive motor 3 (no voltage is applied). 2"
This range is actually a dead zone, and ideally rl, O
The true in-focus state is when J, that is, 2A=A+B, that is, A=B, but the configuration is such that it has a range that can be considered as in-focus. Also, 5 of the pulse numbers PA stored in the microcomputer 6 is the integration time of the sum signal (A+B) during the next distance measurement (12 as shown in FIG.
The same distance measurement operation is performed (steps ■ to ■).The number of pulses P'A↑8 obtained in the same 12 hours is used to set the next integration time (13 hours).
used to set.

In the above-mentioned device, the circuit scale can be kept small because the distance measurement information is obtained based on the integral values of the sum signal (A + B) and signal 2A, which are output in a time-division manner. Although there are advantages such as being able to more effectively cancel the offset voltage, there may be disadvantages such as the photographic lens 2 trembling or hunting near the focus. This is because the light that enters the light-receiving element 4 includes, in addition to the light emitted from the light-emitting element 1 that is reflected back from the subject surface, external light such as fluorescent lamps and sunlight. This is because the external light noise components of the above are superimposed on the signal to be detected.As a means for removing the external light noise components as described above, a bypass filter 8 and a synchronous detection circuit are used as shown in FIG. switch 13
Countermeasures have been taken, such as installing an AC component or installing an integrating circuit 14 to remove the alternating current component, but it is difficult to completely remove the alternating current component. In such cases, the influence of noise components cannot be ignored. That is, FIG. 10 is the same as FIG. 7(b) above.
This figure shows the integral output when obtaining the number of pulses PA+s and PEA shown in .
From defense MIN to P(AI)-Mabe, pulse number P2^
It is assumed that there may be variations from p2A-Hty to PEA-MAX. At this time, P described in Figure 9
When the calculation 2A/P^↑6 is performed, it is conceivable that the fluctuation range due to noise during focusing may take a value from 0°7 to 1.3 depending on the case.

If the movement of the distance #mC photographing lens 2) at this time is extreme, the P2/I/P Yaya 6 may alternate between 0.7 and 1.3, causing a hunting operation, or if not However, there will be an inconvenience that the range ring will start moving again after it stops.

(Object of the Invention) An object of the present invention is to provide an automatic focus detection neck device that solves the above-mentioned problems and can maintain a good balance between reducing hunting operation and reducing focus detection accuracy.

(Features of the Invention) In order to achieve the above object, the present invention stores an integral value of a signal that was previously integrated over a constant integration time by an integrating means,
The current fixed integration time is updated based on the signal, the integrating means is caused to perform an integration operation according to the updated fixed integration time, and the focus state is detected from the obtained integral value. A calculation means is provided for determining the size of a focusing dead zone that can be continuously changed according to the length of a certain integration time, and thereby the updating,
A feature of the present invention is that the size of the focusing dead zone increases as the length of the determined constant integration time increases.

(Embodiments of the Invention) Hereinafter, the present invention will be described in detail based on illustrated embodiments.

FIG. 1 is a block diagram showing one embodiment of the present invention. The same parts as in FIG. 6 are denoted by the same reference numerals, and explanations of their operations will be omitted.

FIG. 2 is a graph showing an example of control of the drive motor 3 by the microcomputer 101 shown in FIG.
*)/100) is soresolet. Note that the dotted line represents an example of control of the drive motor 3 shown in FIG. 9 above. The straight line a is the calculation formula η in the case of 0≦δ≦0.8.
=1-1.256, the straight line C is the calculation formula η==0 in the area of 0.8≦δ≦1.2, and the straight line C is 1.2≦δ≦2.
The calculation formulas η=−1, 5+1.256 in the region of 0° are illustrated.

FIG. 3 is a graph showing an example of calculating the dead zone increase coefficient σ in the microcomputer 101, and the horizontal axis is Ty
l (msec) and the vertical axis is the dead zone increase coefficient σ, respectively. The straight line d is 0≦Tn≦28 (msec
), the dead zone increase coefficient σ (=0) in the region of
are graphs of the dead zone increase coefficient σ (=-0,3192+0.0114Tn) in the region of 28 (msec)≦Tn.

Here, FIG. 2 shows the relationship between δ and η when the dead zone increase coefficient σ is “0”.

Figure 4 shows when crk, o (2g (tasec) < T
This is a graph showing an example of the relationship between δ and η (when n). The straight line f is based on the calculation formula η:1-σ-1,256 in the area of 0≦δ≦(1-σ)/1.25, and the straight line g is based on (1
-σ)/1.25≦δ≦(1,5+σ)/1.25.
Arithmetic formula η=-1 in the area of 1.25≦δ≦2.0,
5-δ+1.256, respectively.

As can be seen from this figure, the dead zone increase coefficient σ is the sum signal (
It has the function of changing the size of the dead zone according to the length of the Tn time which is the integration time of A+B) and the signal 2A.

Next, the operation will be explained using the flowchart in FIG. 5. After integrating the T1 time sum signal (A+B) and signal 2A and obtaining the number of pulses P2A and pA,
■), the microcomputer 101 first uses P2^/FA
+ beat calculation is performed to find δ (step ■) 1. Next, T is the integration time of the sum signal (A+B) and signal 2A.
, it is determined whether or not the time is in the relationship O≦T,≦28 (mgec) (Step 2). As a result, 0≦T1
If it is determined that the relationship is ≦28 (mseC), the dead zone increase coefficient σ is set as rQJ, and η is calculated based on δ obtained from the above formula. For example, when the relationship is O≦δ≦0.8, η is calculated by calculating η=1-1.258 (step (fD)), and the drive motor 3 is controlled in accordance with the η value, that is, in accordance with FIG. 2 (step O). The size of the dead zone in this case is 0.8≦δ≦1.2, as shown by the straight line in FIG.

In addition, if it is determined that there is a relationship of 28 (msec)≦T1, the dead zone increase coefficient σ is calculated from the digital calculation formula σ=-0,3192+0.0114T1, and δ and σ found from the above formulas are calculated. Calculate η based on, for example, O≦δ≦(1-σ) /1, when in the region of 25, calculate η by calculating η=1-σ-1.25δ (
Step @), controlling the drive motor 3 according to the η value, for example, according to FIG. 4 (Step ◎),
In addition, in Fig. 4, it is assumed that T and time are approximately 50 m5ec, so the dead zone increase coefficient σ is "0.25
'', and the size of the dead zone in such a case changes to 0.6≦δ≦1.4, as shown by the straight line g.

According to this embodiment, as the integration time (Tn time) of the sum signal (A+B) and signal 2A becomes longer, the width of the dead zone becomes larger. Since the size of the dead zone is determined by continuously changing the size, it is possible to achieve a perfect balance between reducing hunting operation and reducing focus detection accuracy when the reflectance of the subject is low or the subject is far away. It can be kept well.

(Correspondence between the invention and the embodiments) In this embodiment, the light projecting element l is the light projecting means of the present invention,
The light receiving element 4 serves as a light receiving means, the integrating circuit 14 serves as an integrating means,
The microcomputer 101 corresponds to calculation means.

(Effects of the Invention) As explained above, according to the present invention, the integral value of the signal that was previously integrated over a constant integration time by the integrating means is stored,
The current fixed integration time is updated based on the signal, the integrating means is caused to perform an integration operation according to the updated fixed integration time, and the focus state is detected from the obtained integral value. Calculating means is provided for determining the size of a focusing dead zone that can be continuously changed according to the length of the constant integration time, so that as the length of the updated constant integration time becomes longer, By increasing the size of the focusing dead zone, it is possible to maintain a very good balance between reducing hunting operation and reducing focus detection accuracy.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment that has not yet been developed, FIG. 2 is a diagram similarly explaining an example of controlling the drive motor, FIG. 3 is a diagram similarly explaining an example of calculating the dead zone increase coefficient, and FIG. 4 is a diagram explaining an example of controlling the drive motor, FIG. 5 is a flowchart, FIG. 9 is a flowchart, and FIG.
The figure is a diagram for explaining an example of control of the drive motor by the microcomputer in FIG. 6, and FIG. 10 is a diagram for explaining the integral output when affected by external light or the like. l... Light projecting element, 4... Light receiving element, 7
......Sensor amplifier, 14...Integrator circuit, 101...Microcomputer, A, B.
... Signal, σ ... Dead band increase coefficient, Tn
...Integration time. Patent Applicant Canon Co., Ltd. Agent Minoru Nakamura Figure 3 Figure 9 Figure 1o

Claims (1)

[Claims] 1. A light projecting means that projects light toward an object, and a light projector that receives reflected light from the object and generates two types of signals that relatively change depending on the distance of the object. In an automatic focus detection device comprising a light receiving means for outputting light and an integrating means for integrating a signal from the light receiving means, an integral value of a signal that was previously integrated over a predetermined integration time by the integrating means is stored, and The current constant integration time is updated based on the updated constant integration time, the integration means is caused to perform an integration operation according to the updated constant integration time, and the focus state is detected from the obtained integral value.At the same time, the updated constant integration time is An automatic focus detection circuit characterized by being provided with arithmetic means for determining the size of a focusing dead zone that can be continuously changed according to the length of time.