JPH0614697B2 - Imaging device - Google Patents

Description

TECHNICAL FIELD The present invention relates to a video camera which automatically adjusts focus and exposure by using a picked-up image signal.

(B) Conventional technology In an autofocus device for a video camera, the method of using the video signal itself from the image sensor to evaluate the focus control state has essentially no parallax and has a depth of field. It has many advantages such as being able to focus accurately with a shallow subject or a distant subject. Moreover, a special sensor for autofocus is unnecessary, and the mechanism is extremely simple. JP-A-61-105978 (H04N
5/232) discloses an example of the autofocus device as described above.

In the above technique, the high frequency component level of the picked-up image signal is A / D converted within the range of the focus area set at the center of the screen, and the conversion data is integrated for each field by the integration circuit, The digital data of is held as the focus evaluation value, and compared with the evaluation value one field before,
The focus motor is always driven and controlled in the direction that maximizes the focus evaluation value.

On the other hand, the exposure is controlled via the servo system so that the screen illuminance becomes constant using the average value or peak value of the video signal from the image sensor. If the detection and control are performed uniformly over the entire area, when a bright subject such as the sky exists in the screen, the amount of light in the subject other than the sky becomes insufficient, which makes the screen difficult to see. As a method to solve this problem, a so-called center-weighted metering method that extracts only the signal in the center of the screen using a wide blanking pulse or this method is further improved to use a part of the screen as a metering area than other parts. A weighted auto-iris circuit has been proposed (Japanese Patent Laid-Open No. 58-11489).

The present applicant has further developed the above-mentioned conventional example, inputs the image pickup video signal at the center of the screen to the HPF and LPF, and outputs the filter output alternately at regular intervals by a digital integrator circuit or the like. To a focus evaluation value and an exposure evaluation value, respectively, and executes focus (AF) and exposure (AE) control based on these evaluation values to extract the imaged video signal at the center of the screen. Or
Japanese Patent Application No. 62-223380 proposes a technique in which a digital integrator circuit can be commonly used for both AF and AE.

(C) Problems to be Solved by the Invention According to the prior application technique, a common circuit such as a gate circuit and a digital integrator circuit is used in a time-division manner, so that it is impossible to obtain the focus and exposure evaluation values at the same time. Since the AF and AE controls are also time-shared, discontinuous changes in evaluation values, delays in reactions, etc. are likely to occur.

(D) Means for Solving the Problem The present invention calculates a missing period of each evaluation value generated by using a common circuit in a time-division manner based on the evaluation value obtained before this missing state occurs. It is characterized in that it is interpolated according to the predicted evaluation value.

(E) Action Since the present invention is configured as described above, the discontinuous change in the evaluation value and the delay in the reaction associated therewith are eliminated.

(F) Embodiment One embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a circuit block diagram of this embodiment. (1) is a video camera unit, a focus motor (4) that drives a focus ring (3) that supports the focus lens (2) and moves back and forth in the optical axis direction, an aperture mechanism (6) that controls exposure,
An iris motor (7) that drives the diaphragm mechanism (6) and a solid-state image sensor (CCD) that converts subject light into an imaged video signal.
The image pickup circuit (8) having the is arranged.

The luminance signal in the imaged video signal obtained by the image pickup circuit (8) includes a first high pass filter (HPF) (9) having different cutoff frequencies, a second HPF (10), a low pass filter (LPF) (11) and It is sent to the sync separation circuit (12).

The vertical sync signal (VD) and the horizontal sync signal (HD) separated from the luminance signal by the sync separation circuit (12) are supplied to the switching control circuit (13) to set the sampling area. The switching control circuit (13) controls the vertical / horizontal synchronization signal (V
D), (HD) and a selection signal (S ₂ ) so that a rectangular sampling area (A ₁ ) can be set in the center of the screen based on a fixed oscillator output that serves as a clock for driving the CCD.
Is output to the selection circuit (15) in the subsequent stage, and the first HPF
(9) The output of the second HPF (10) is switched every one field, and a switching signal (S ₁ ) for selecting the output of the LPF (11) is output once every four fields to the switching circuit (14).

The switching circuit (14) receives the switching signal (S ₁ ), selects the first HPF (9) output and the second HPF (10) output for each field, and outputs them to the selection circuit (15) in the subsequent stage. The LPF (11) output is selected only once in the field and output to the selection circuit (15).

The selection circuit (15) is a switching circuit based on the selection signal (S ₂ ).
The output selected in (14) is selectively output to the integrating circuit (16) according to the sampling area (A ₁ ). That is,
Each filter output related to the first sampling area (A ₁ ) is output to the integrating circuit (16).

The integrating circuit (16) is composed of an A / D converter (22), an adder (23) and a memory circuit (24), and the A / D converter (22) is a selection circuit (1
Each filter output passing through 5) is sequentially A / D converted and output to the adder (23). The adder (23) is the A / D of the previous stage
A digital integrator is constructed with the memory circuit (24) in the latter stage of the converter (22), and the output of the memory circuit (24) and the A / D converter.
(22) Add the outputs and add the result again to the memory circuit (2
Supply to 4). The memory circuit (24) is reset for each field, and the output of the adder (23), that is, the sampling area (A
_One field of ₁ ) will be retained.

This integrated value is further stored in the memory circuit (25) at the subsequent stage.

The cutoff frequency of each of the first HPF (9), the second HPF (10) and the LPF (11) is 600 KHz or more, 200 KHz
Above, it is set to 2.4MHz or less, actually 60
0 KHz to 2.4 MHz, 200 KHz to 2.4 MHz, 0
It can be set with a BPF having a pass band of 2.4 MHz. At this time, 2.4 MHz is an extremely high frequency that has little relation to the luminance signal, and therefore the LPF (11) may be omitted. Therefore, the high band or low band component of the luminance signal that has passed through one of the first and second HPF (9) (10) and LPF (11) is digitally integrated for one field, and the memory circuit (25) Will be stored in. Of the integrated values stored in the memory circuit (25), L
The integrated value of the low frequency component when PF (11) is selected is used as the exposure evaluation value for exposure control, and the first HPF (9) or the second
The integrated value of the high frequency component when the HPF (10) is selected is processed by a microcomputer (26) in the subsequent stage as a focus evaluation value for focus control.

These evaluation values are processed by software by the microcomputer (26), and based on the processing result, a command is issued to the focus motor control circuit (27) to drive the focus motor (4) to drive the focus lens (2). To move the iris motor (7) to operate the diaphragm mechanism (6) by issuing an instruction to the iris motor control circuit (28) to execute the autofocus operation so that the focus evaluation value becomes maximum. Then, automatic exposure adjustment can be performed so that the exposure evaluation value becomes a predetermined value.

Next, referring to the flow chart in Fig. 2, microcomputer (26)
A main routine of the auto focus operation and the auto iris operation of the focus motor control circuit (27) and the iris motor control circuit (28) will be described.

When the video camera enters the operating state, the microcomputer (26) goes to the second
The main routine shown in the figure is executed for each field.

First, in STEP (30), the integrated value for one field in the sampling area in the current field is read from the memory circuit (25) into the microcomputer (26).

Next, the counter (AECNT) provided for performing the autofocus operation and the auto-iris operation in time division is decremented, that is, subtracted by 1 (STEP (32)), and it is determined whether the counter value is 0 (STEP). (33)), if the counter value is not 0, AF for autofocus operation
The routine is executed first, and only when the counter value is 0, the AE routine for the auto iris operation is executed first. When the value of the counter (AECNT) is not 0, STEP
The integrated value read into the microcomputer (26) at (30) is used to calculate a focus evaluation value in a focus evaluation value calculation routine (31). The calculation of this focus evaluation value will be described.

First, the integrated value DA in the sampling area (A ₁ ) held in the memory circuit (25) in the current field
TA is the H of either the first HPF (9) or the second HPF (10)
It is determined in STEP (51) whether or not the first HPF (9) has been extracted in STEP (52) (53).
If it is due to the second HPF (10), DATA is substituted into the memory A, and if it is due to the second HPF (10), it is substituted into the memory B.

Next, in STEP (54), the focus evaluation value of the sampling area (A ₁ ) in the previous field is transferred to the memory Y.

In STEP (55), according to the data in the memories A and B,
The focus evaluation value X of the sampling area (A ₁ ) of the current field is calculated. Here sampling area (A ₁ )
The focus evaluation value X of the sum of the values of the memory A and the value of B, that is, the latest integrated value when the first HPF (9) is used in the sampling area (A ₁ ) and the second HPF (10) are used. It is the moving sum obtained by adding the latest integrated value of time. For example, data is taken in as a1, b1, a2, ... As DATA for each field (however, a1, a2, a3 ,.
(Integrated value of sampling area (A ₁ ) by F (9) output, b1, b2, b3, ... Assume integrated value of sampling area (A ₁ ) by second HPF (10) output)
And the focus evaluation value X is a1 + b1, b1 for each field.
+ A2, a2 + b2, and so on. Therefore, the focus evaluation value for the sampling area is the sum of the integrated value of one of the HPF outputs in the current field and the integrated value of the other HPF output in the previous field,
The integrated value of the odd field and the even field is included in one focus evaluation value, and as a result, the variation of the evaluation value for each field due to interlacing or the like and the influence of noise can be alleviated and the result becomes stable.

The focus evaluation value calculated as described above is stored in the memory circuit (29).
Even if it is updated for each field, it is stored as the focus evaluation value of the previous time, two times before, and three times before.

An autofocus (AF) routine (35) for controlling the rotation of the focus motor (4) is performed based on the focus evaluation value calculated as described above. This AF routine is executed based on the flowchart of FIG. This AF
The five routines forming the core of the routine, that is, the evaluation value stability confirmation routine, the direction determination routine, the hill climbing routine, the vertex return routine, and the evaluation value variation monitoring routine will be briefly described in order.

The evaluation value stability confirmation routine (45) uses the operation mode code (M
ODE) is “0” when the power is turned on or when the subject changes and the autofocus operation is restarted. Focus evaluation values in the current field and the previous field with the focus motor (4) stopped. (X) and (Y) are compared, and it is determined whether or not the difference is within the allowable range. If the difference is within the allowable range for a predetermined period (for example, between 5 fields), it is recognized that the subject is stable. , The operation mode code (MODE) is set to "1", and the direction determination routine is executed from the next field. Therefore, while the subject is moving and is not stable, this evaluation value stability confirmation routine is continued and substantially no focusing operation is performed.

Next, the stability of the subject is confirmed, and the operation mode code (MO
When DE) is set to "1", in the direction determination routine (46) that is recognized and executed in STEP (44), first, a command to activate the focus motor (4) is issued to focus the focus motor (4). After moving the lens (2) in the specified direction (for example, the ∞ point direction) to initialize it, compare the focus evaluation values (X) (Y) of the current field and the previous field over a predetermined period to obtain the focus evaluation value. If it can be confirmed that there is an increasing tendency by determining whether (X) is increasing or decreasing, the rotation direction of the focus motor (4), that is, the moving direction of the focus lens (2) is set to the default direction. The direction is discriminated by issuing a command so that the direction continues, and conversely, if it is confirmed that there is a decreasing tendency, a command is issued to immediately reverse the rotation of the focus motor (4) and the direction is discriminated. When this direction determination is completed, the operation mode code (MODE) is changed to “2” and the hill climbing routine (4
7) is executed.

In this mountain climbing routine (47), the focus evaluation values of the current field and the previous field are kept in the state where the focus motor (4) is continuously rotated in the direction determined by the direction determination routine described above.
(X) and (Y) are compared, and the focus evaluation value (X) of the current field is the focus evaluation value of the previous field while climbing the slope of the mountain that has a peak at the in-focus position as shown by the arrow in FIG. While the value is larger than the value (Y), a command is issued to connect the rotation direction of the focus motor (4). As shown in Fig. 6, the focus evaluation value exceeds the peak and exceeds the reference value (Δx) in the current field. Focus evaluation value at
When it is confirmed that (X) has dropped, the operation mode code (MODE) is changed to "3" and the vertex return routine (48) is executed from the next field. By the way, in this mountain climbing routine (47), the rotation amount of the focus motor (4) which is a stepping motor is set to 10 steps per field.

In the above-mentioned mountain climbing routine, every time the focus evaluation value (X) of the current field becomes larger than the focus evaluation value (Y) of the previous field, the focus evaluation value (X) at that time is set as the maximum evaluation value (X _M ). Hold and at the same time this maximum evaluation value (X _M )
The lens position that becomes is stored as the focus position. The focus position is stored every time the maximum evaluation value (X _M ) is updated, and the rotation step amount of the focus motor (4) is always counted. For example, the infinity direction is added and the near point direction is subtracted. Position memory (ME) which is UP / DOWN counter
Is used. Therefore, when the drop amount reaches the reference value (Δx) two fields after reaching the top of the mountain,
20 steps are stored in the position memory (ME).

The vertex return routine (48) is a command to issue a command to reverse the focus motor (4) by the amount of overshoot from the focus position of the focus lens (4), and execute the operation to return to the top of the mountain. Yes, specifically, by reversing the focus motor (4) and stopping the focus motor (4) when the position memory (ME) becomes zero,
The focus lens (2) has returned to the in-focus position. At this point, the operation mode code (MODE) is "4"
The focus operation from the evaluation value stability confirmation routine to the vertex return routine is completed. Thereafter, the evaluation value fluctuation monitoring operation is executed in the evaluation value fluctuation monitoring routine from the next field, and if it is confirmed that the focus evaluation value (X) has changed from the maximum evaluation value (X _M ) to the allowable range or more, the subject is detected. Assuming that the camera has moved, the operation mode code (MODE) is changed to "0", the evaluation value stability confirmation routine (45) is performed from the next field, and the focusing operation is restarted.

In the AF routine described above, while the operation mode code (MODE) does not change, one of the five routines described above is continuously updated based on the new focus evaluation value every time the focus evaluation value is updated. To be executed. Further, in the present embodiment, the focus evaluation value is calculated by adding the integral values of the high frequency component levels having different bands in successive fields, but the integral value of one field of one of the high frequency component levels is calculated. Needless to say, may be defined as the focus evaluation value.

In STEP (41), by executing any of the above five routines, the drive command information (Daf) for rotating the focus motor (4) in the infinity direction or the near point direction or maintaining the stopped state Let me decide. That is, while executing any of the five routines, the focus motor
If it is decided to issue a command to rotate (4) in the direction of ∞, Daf> 0 is set. If it is determined to issue a command to rotate in the direction of near point, Daf <0 is set, and a command to maintain the stopped state is given. When it is decided to emit the message, Daf = 0.

On the other hand, when the value of the counter (AECNT) becomes "0", that is, when four fields have elapsed from the start of the focusing operation, the exposure evaluation value (Z) is calculated by the exposure evaluation value calculation routine (34). Later, the auto iris (AE) routine (38) is executed based on this exposure evaluation value, and the iris motor
Control of (7) is performed.

First, in the exposure evaluation value calculation routine (34), the integrated value DATA for one field of the luminance signal that has passed through the LPF (11) in the sampling area (A ₁ ) is divided by the area (SM ₁ ) of the area (A ₁ ). The integrated value per unit area obtained by the calculation is calculated as the exposure evaluation value (Z). This exposure evaluation value is stored in the memory circuit (29) similarly to the focus evaluation value. Next, the AE routine (38) shown in the flowchart of FIG.
In step ( ₈ ), it is judged whether the exposure evaluation value (Z) is equal to, larger than, or smaller than the target value (Z ₀ ).
0) (81), if both are equal, STE
Drive command information (D) for the iris motor (7) at P (82)
Substituting "0" into ae) and when Z> Z ₀ , STEP
At (83), "-1" is substituted for the drive command information (Dae), and when Z <Z ₀ , at STEP (84) "1" is substituted for the drive command information (Dae).

STEP according to the count value of the counter (AECNT)
(33) divided into cases, focus evaluation value calculation routine (31),
When the drive command information (Daf) to the focus motor (4) is determined in the AF routine (35), the switching circuit (14) selects the first or second HPF (9) (10) output and Since the integrated value in the field is used as the focus evaluation value, the integrated value due to the LPF (11) output in the current field cannot be taken out, and the exposure evaluation value is temporarily lost during this period. Therefore, based on the exposure evaluation value obtained last time, the exposure evaluation value prediction routine (1
In step 30), a value predicted as the current field is calculated and interpolated, and the AE routine (13
Execute 8). This AE routine (138) is A
It is executed according to the same flow chart as the E routine (38). In this way, the AF routine (35) and the AE routine (13
8) is executed, drive command information (Daf) to the focus motor control circuit (27), and the iris motor control circuit (28)
When the drive command information (Dae) to the motor is determined, both motor control circuits (27) and (28) control the drive of both motors in the motor drive routine (131).

This motor drive routine (131) is shown in the flowchart of FIG. 4, and when the drive command information (Daf) of the focus motor (4) is positive in STEP (141) (142), the focus motor (4) is set to ∞ point. Direction, in the negative direction, in the near point direction, in the case of "0", it is maintained in the stopped state. Also ST
When the drive command information (Dae) of the iris motor (7) is "0" in EP (143) (144), the iris motor
Keep (7) in a stopped state, when positive, the iris motor (7)
The iris motor (7)
Is driven in the closing direction.

When the drive control of the focus motor (4) and the iris motor (7) is executed by the motor drive routine (131) in this way, the content of the counter (AECNT) is decremented by 1 and the counter value becomes "0". Is determined (STEP. (3
6)), if 0, the switching signal (S ₁ ) is issued from the microcomputer (26) to the switching control circuit (13) to select the LPF (11), and the LPF (11) is selected. Done (S
TEP (37)). The integrated value of the next field is used as the exposure evaluation value.

On the other hand, the count value of the counter (AECNT) is “0”, and the exposure evaluation value calculation routine (34) and the AE routine (3
8) to execute the drive command information (D
When ae) is determined, the focus evaluation value prediction routine (132) interpolates a temporary loss of the focus evaluation value caused by selecting the LPF (11) and using the integral value as the exposure evaluation value. In this focus evaluation value prediction routine (132), a value predicted in the current field is calculated from the focus evaluation values obtained the previous time and the time before the second, as will be described later, and this is used as interpolation data. The AF routine (135) is executed using the interpolation data as the focus evaluation value of the current field. The AF routine (135) is executed according to the same flowchart as the AF routine (35). In this way, when the AE routine (38) and the AF routine (135) are executed and the drive command information (Dae) and (Daf) are determined, the focus motor control circuit (27) is executed by the motor drive routine (133) as described above. ) Controls the focus motor (4), and the iris motor control circuit (28) controls the iris motor (7). The motor drive routine (133) is shown in the same flow chart as the motor drive routine (131).

When the motor drive routine (132) is completed, the counter (AECNT) is returned to the initial state (STEP (3
9)), select the filter as the first HPF (9) (STEP
(40)), wait for the sum of the evaluation value of the next field. Here, the initial state of the counter (AECNT) is a state in which the initial value "4" is set in order to calculate the exposure evaluation value based on the luminance signal that has passed through the LPF (11) every four fields.

Next, the interpolation operation of the focus evaluation value and the exposure evaluation value by the focus evaluation prediction routine (132) and the exposure evaluation value prediction routine (130) will be described. First, in the focusing operation using the focus evaluation value, the change state of the evaluation value is important. Therefore, regarding the interpolation of the focus evaluation value, the focus evaluation value (Xn- ₁ ) previously captured (the focus evaluation value calculation routine described above is used.
(31), shown as (Y) in AF routine (35)),
The focus evaluation value two times before (Xn− ₂ ) and the focus evaluation value three times before (Xn− ₃ ) are read from the memory circuit (29) and are calculated by substituting them into the following equation.

X = Xn- ₁ + (Xn- ₁ -Xn- ₂ ) + (Xn- ₁ -2Xn- ₂ + Xn- ₃ ) / 2 The above equation is derived by Taylor expansion. That is, tailoring By ignoring terms of the third order or higher from the equation, the unit quantity (always one field interval) is set to (x−x ₀ ) and 1 is set as the previous focus evaluation value (Xn− ₁ ) as f (x ₀ ). , F '(x
₀ ) as the focus evaluation value (Xn− ₁ ) (Xn− ₁ ) (X
substituting slope _(Xn- 1 -Xn- ₂₎ between the _n-2),
Further, as f ″ (x ₀ ), the focus evaluation value (X
(n- ₁ ) (Xn- ₂ ) and the difference between the gradients of the focus evaluation values (Xn- ₂ ) (Xn- ₃ ) _two and three times before (Xn- _1- Xn-). It is approximated by substituting ₂ )-(Xn- _2- Xn- ₃ ).

Further, in the exposure control operation using the exposure evaluation value, the evaluation value itself at that time is important, so only the movement of the iris motor (7) is considered as a factor of fluctuation. Here, in the normal continuous drive of the iris motor (7), the exposure evaluation value is taken in the cycle, that is, the exposure is 1 EV during 4 fields.
It is done at a changing rate. That is, if the iris motor (7) continuously rotates in the aperture opening direction, the exposure will increase by 1 EV during four fields. Here, EV is a unit of brightness, and if EV is increased by 1 EV, brightness doubles. Therefore, as the formula for predicting the exposure evaluation value, (However, Is 2 to the m-th power). In the above formula (Zn
_-1 ) is the exposure evaluation value fetched last time, read from the memory circuit (29), and (Dae) is the drive control command determined in the previous AE routine, and is either 0 or ± 1. , And the increase / decrease of exposure is decided in the formula.
1/4 is converted into 1 field by changing 1 EV in 4 fields.

In the above-mentioned evaluation value interpolation operation, in a complicated conditional expression,
Interpolation data was calculated by substituting the evaluation values up to the previous time,
It is also possible to realize the interpolation of the evaluation value by simply adding the increase / decrease amount of the previous time and the value of two times before to the previous evaluation value or holding the previous evaluation value.

(G) Effect of the Invention As described above, according to the present invention, the evaluation value that is the basis of the focusing operation and the exposure evaluation value that is the basis of the exposure control operation are missing in the evaluation value caused by time division detection. Can be predicted and interpolated to eliminate the discontinuous change in the evaluation value and the delay in the reaction in the focusing operation and the exposure control operation that accompany this.

[Brief description of drawings]

The drawings all relate to one embodiment of the present invention. FIG. 1 is a circuit block diagram, FIG. 2 is a flowchart of a main routine, FIG. 3 is a flowchart of a focus evaluation value calculation routine, and FIG.
Figure is a flow chart of the motor drive routine, and Figure 5 is A
FIG. 6 is a flowchart of the F routine, FIG. 6 is a diagram for explaining changes in the lens position and the focus evaluation value during the focusing operation, and FIG. 7 is a flowchart of the AE routine. (26) …… Microcomputer (evaluation value detection means), (27) …… Focus motor control circuit (focus control means), (28) …… Iris motor control circuit (exposure control means).

Claims (1)

[Claims] 1. A focus evaluation value is detected by accumulating sampling values of high frequency component levels of a luminance signal obtained from an image pickup means having an image pickup element over a fixed period, and the focus evaluation value is detected periodically. Interrupting, and instead an evaluation value detecting means for detecting the exposure evaluation value by integrating the sampling value of the brightness level of the brightness signal over a fixed period, and the focus lens and the focus evaluation value so that the focus evaluation value becomes maximum. Focus control means for controlling the distance to the image sensor, and exposure control means for performing exposure control so that the exposure evaluation value becomes a target value, the other occurring during the detection period of either one of the both evaluation values The image pickup device, wherein the missing period of the evaluation value is interpolated by a value calculated based on a value obtained before the missing period of the evaluation value on the missing side.