JPH087323B2 - Focus detection device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus detection device, and more particularly to an autofocus (hereinafter referred to as autofocus) which performs automatic focus detection using a video signal (captured video signal) output from a video camera. , AF) device.

[Prior Art] In recent years, as an AF method that makes full use of the characteristics of video cameras,
The so-called hill-climbing servo system, which detects the screen definition by the high frequency components in the video signal, drives the focusing lens to maximize the definition, and automatically adjusts the optical focus of the TV camera, Attention has been paid. This conventional hill-climbing method (hereinafter referred to as "hill-climbing method") is described in detail, for example, in NHK Technical Research, 1965, Vol. 17, No. 1, Vol. 86, pp. 21-37. The contents will be briefly described with reference to the configuration diagram of the conventional hill climbing automatic focus detection device and the characteristic diagram of FIG. 5 showing the operating characteristics thereof.

First, the light from the subject 1 incident through the focusing lens 2 forms an image on the photoelectric conversion means 3 and is output as an electric signal. An output signal as a video signal output from the photoelectric conversion means 3 is input to a process circuit 5 for a video camera via an amplifier 4, and at the same time, a band pass filter separately arranged in parallel after the amplifier 4 for AF detection. Only the high frequency component in the video signal is extracted by the (BPF) 6 and input to the gate circuit 7 in the next stage.

Further, from the horizontal synchronizing signal HD and the vertical synchronizing signal VD separated by the camera process circuit 5 described above, a window pulse forming circuit (WINDOW) 8 is used to preset a region of one screen, for example, only the central portion of the screen. , A so-called window pulse is formed, the window pulse is input to the gate circuit 7 described above, and the high frequency component signal is extracted and output only in the pulse generation section. The high frequency component signal output from the gate circuit 7 is further processed by a detector (DEF) 9 and an integrator 10.

The relationship between the voltage corresponding to the output of the integrator 10 (hereinafter referred to as the focal voltage) and the focal length of the lens 2 is shown in FIG. Since this focus voltage corresponds to the definition of the captured image, the position of the distance ring (focus ring) for moving and adjusting the position of the focusing lens 2 accurately matches the actual distance between the lens 2 and the subject 1. If it is, that is, at the just focus point (focus point position), the focus voltage becomes maximum,
As it deviates from this maximum position C, it decreases.

Therefore, as can be seen from FIG. 5, if some control means controls the position of the lens distance ring so as to climb the peak of the focus voltage and guides the lens distance ring to the top of the peak where the focus voltage becomes the maximum, Focusing will be possible. The control means is an integrator 10 for each field of the video signal.
It is known that the objective is achieved by holding the output of (1), comparing the value held last time and the value held this time for each field, and driving the lens distance ring in the direction in which it becomes larger. That is, as shown in a hill climbing circuit 16 surrounded by a broken line in FIG. 4, a mono-multivibrator (MM) 12 and a sample pulse forming circuit (SP) 13 are obtained from the vertical synchronizing signal VD separated by the camera process circuit 5 described above. By this, a sample pulse generated at a constant timing is formed, and the sample pulse holds the integrator output input to the sample hold circuit (S / H) 11 for each field. Further, the output of the sample and hold circuit 11 is branched into two, one of them is kept as it is, and the other is delayed by one field by a one field delay circuit (1 field Delay) 14 and input to a comparator 15 in sequence. The difference, that is, the difference between the integrator output of the previous hold and the integrator output of the current hold is calculated by the comparator 15.

If the focus voltage of the output of the integrator 10 is at the level A in the previous field and is at the level B in the next field, the focus voltages A and B are B> A, as shown in FIG.
Therefore, for example, a high-level control signal is output from the comparator 15 to the motor drive control circuit (DRIVER) 17, and the motor (Mo) 18 is driven while maintaining the direction as it is until now. By driving the motor 18, the lens 2 moves in the direction of the focal point. When the focus voltage is detected in the next field, assuming that the level of C in FIG. 5 is C> B, the signal level from the comparator 15 remains unchanged and the motor 18
Yes, they will move in the same direction.

At this time, as shown in FIG. 5, assuming that the maximum point of the focus voltage is at the level of C, if the lens distance ring is moved in the same direction by the motor 18 as a matter of course, the in-focus state of C is reversed. The focus voltage tends to decrease as the focus voltage decreases. As a result, the focus voltage of D level is output in the next field, and since D <C, the comparator 15 outputs a low level signal to the control circuit 17. With this low level signal, the control circuit 17 outputs a control signal for rotating the motor 18 in the reverse direction, and the lens distance ring is sent back in the focusing direction again.

As described above, in the conventional hill climbing method shown in FIGS. 4 and 5, when the output of the hill climbing circuit 16 configured by the comparator 15 and the like is positive (high level), that is, the focus voltage changes with time. If it is in the increasing direction, the rotation direction of the motor 18 is kept as it is and the hill climbing is continued. When the output of the hill climbing circuit 16 is negative (low level), that is, if the focus voltage is in the decreasing direction with respect to the passage of time, the motor 18 Reverse the direction of rotation and return to climbing the mountain. Therefore, the fifth created by the focus voltage
Automatic focus detection is performed by climbing a mountain as shown in the figure while referring to the output voltage of the integrating circuit 10, and finally oscillating in small steps at the top of the mountain and reaching a steady state. .

As described above, in the hill climbing method, since the automatic focus detection is performed using the output signal itself from the image pickup means, it is not necessary to separately provide a component for automatic focus detection, and the focus detection can be performed relatively inexpensively and accurately. There is an advantage that you can.

[Problems to be Solved by the Invention] However, in the above-described conventional hill-climbing method, in order to determine whether or not the peak (maximum value) of the focus voltage in the focused state, it is necessary to get off the mountain once. However, since it cannot be determined whether or not it was the top of the mountain, there is a drawback that it reaches a steady state while vibrating in small steps near the top of the mountain. That is, in the focused state, the motor repeats forward and reverse rotations, climbs and descends near the top of the mountain, and is constantly vibrating in small steps. This vibration causes deterioration of the image quality and therefore needs to be removed.

SUMMARY OF THE INVENTION It is an object of the present invention to eliminate the above-mentioned drawbacks, to eliminate a state in which a lens vibrates in small increments on the peak of a focus point, and to provide a focus detection apparatus which performs a focus operation accurately.

[Means for Solving the Problems] In order to achieve the present object, the present invention provides a signal extraction means for sequentially extracting and outputting a signal corresponding to the focus degree of an imaging lens at a predetermined timing, Prediction means for predicting the degree of focus using the plurality of signals output by the signal extraction means, detection means for detecting at least one of lens information such as aperture value and focal length of the imaging lens, and the detection Correction means for correcting the predicted value of the prediction means according to the lens information detected by the means.

[Operation] According to the present invention, the output means sequentially outputs signals (focus voltage) corresponding to the focus degree at a predetermined timing, and the predicting means uses at least three or more signals that are output in succession. Predict the degree of focus. Furthermore, the lens information of the image pickup lens is detected by the detection means, and the prediction value of the prediction means is corrected by the correction means according to the detected lens information. Therefore, accurate automatic focus detection can be performed, and by stopping the driving of the imaging lens using the predicted value of the prediction means, vibration near the in-focus point is eliminated.

Embodiments Embodiments of the present invention will be described in detail below with reference to the drawings.

A. Basic configuration FIG. 1 shows the basic configuration of an embodiment of the present invention. here,
a is an output unit that sequentially outputs a signal corresponding to the focus degree of focus at a predetermined timing, and b is a prediction that predicts the focus degree by using three or more signals that are consecutively extracted by the output unit a. It is a means.

c is detection means for detecting lens information of the imaging lens, d
Is a correction means for correcting the predicted value of the prediction means b according to the lens information detected by the detection means c. Correction means d
Adjusts various constants of, for example, a filter and a gate circuit which constitute the output means a to appropriate values according to the lens information.

B. Circuit Configuration FIG. 2 shows a circuit configuration of an embodiment of the present invention. In the figure, 19 is a 1-field delay circuit as an output means, which inputs the focus voltage B in the previous field delayed by 1 field by the 1-field delay circuit 14 in the previous stage and further delays it by 1 field before the previous field. The focus voltage A of is output. Reference numeral 20 is a peak predicting circuit as a predicting means, and these are calculated from the input focus voltage c of the current field, focus voltage B of the previous field, and focus voltage A of the field two years before.
A predetermined known arithmetic expression using A, B, C as parameters, for example, a quadratic function y = ax ² + bx + c (where x is the number of fields,
y is the focal voltage) Based on the prediction formula of the focal voltage, the focal voltage P at the peak corresponding to the in-focus state in what field (see FIG. 3)
Then, the predicted timing signal after N fields is output as a gate signal, and the predicted top focus voltage P is output as a comparison signal.

As shown in FIG. 3, the focus voltage in the current field (i) is C, the focus voltage in the previous field (i-1) is B, and the focus voltage in the previous field (i-2) is A. Reference numeral 21 is a first subtraction circuit, which subtracts the focus voltage B from the focus voltage C and outputs a signal corresponding to CB. A second subtraction circuit 22 subtracts the focus voltage A from the focus voltage B described above and outputs a signal corresponding to BA.

Reference numeral 23 denotes a first comparator, which is a focus voltage C input to the non-inverting input terminal and a predetermined threshold V _{ref I} set at the inverting input terminal.
Is compared to determine whether C> V _{ref I} , and when a positive determination is made, a high level signal is output. Reference numeral 24 denotes a second comparator, which is an output (CB) of the first subtraction circuit 21 input to the inverting input terminal and a predetermined threshold V set at the non-inverting input terminal.
by comparing the _{ref II} determines whether C-B <V _{ref II,}
When the determination is affirmative, a high level signal is output. A third comparator 25 is an output (CB) of the first subtraction circuit 21 input to the inverting input terminal and an output (BA) of the second subtraction circuit 22 input to the non-inverting input terminal. Compare with and CB
<B-A is determined, and if a positive determination is made, a high-level signal is output.

Reference numeral 26 is an AND circuit (AND), which is the above-mentioned three comparators 2
The output of 3, 24 and 25 is logically ANDed, and when all its outputs are high level, the gate is opened and the high level signal is output. The output signal of the AND circuit 26 is supplied as a control signal for the top prediction circuit 20 and a motor speed control signal. Reference numeral 27 denotes a fourth comparator, which is a sample and hold circuit 11 for inputting to the non-inverting input terminal, the predicted maximum focus voltage P from the peak predicting circuit 20 and the inverting input terminal.
From the current focus voltage C in the current field from
Whether or not it is C is determined, and when the determination is affirmative and the gate pulse (timing signal) from the top prediction circuit 20 is ON, the motor stop signal is output. The comparator 27 itself may be configured to have a range of comparison judgment.

Reference numeral 28 is a motor control circuit (M. control) for driving and controlling the lens moving motor 18 for moving the distance ring of the focusing lens 2, and when the output of the mountain climbing circuit 16 is at a high level, the motor 18 is normally rotated. , Its output reverses the motor 18 when its output is low. The control circuit 28
Makes the motor 18 slow when the output of the AND circuit 26 is high level, makes the motor 18 high speed when the output is low level, and stops the motor 18 by the motor stop signal of the fourth comparator 27. The hill-climbing circuit 16 has substantially the same configuration as that of the conventional example shown in FIG. 4, operates according to the output of the first comparator, compares the focus voltages C and B, and when C> B, the motor It outputs a high level signal for normal rotation, and outputs a low level signal for motor reverse rotation when C <B.

Further, 29 is a lens information sensor as a detecting means, which detects information about the lens state of the focusing lens 2, for example, the F number F _NO , the absolute address of the distance ring, or the focal length. Reference numeral 30 is a control circuit as a correction means, and according to the detection output of the lens information sensor 29,
It controls the frequency characteristic of the BPF 6, the magnitude of the window pulse input to the gate circuit 7, and the correction of all or part of the predicted value of the peak prediction circuit 20. For example, since the information on the screen changes according to the focal length, the control circuit 30 causes the gate circuit 7 to narrow the window when the wide scope is used and widen the window when the telescope is used. Further, when the F number F _NO is large, the depth of field becomes deep, and the slope of the peak of the focus voltage characteristic becomes shallow,
The control circuit 30 selects the filter value of BPF6 so that the depth of field becomes deep.

That is, since the amount of information on the screen changes as shown in FIG. 6 by zooming, the size of the gate of the gate circuit 7 is changed by the control circuit 30 in order to keep this amount as constant as possible. . For example, from the left side of this figure (large focal length) to the right side of wide (small focal length)
The control circuit 30 changes the size of the gate in the distance measurement area with respect to the entire screen as the display changes.

In addition, since the video signal has a steep peak (a peak of the focus voltage characteristic) on the high frequency side, the filter 6 is switched to the high frequency side under the condition that the peak is gentle according to the lens information at that time. When there is a mountain at, switch to the low frequency side. For example, when the focal length of the lens 2 is short or the aperture value is small, the peaks are gentle, and on the contrary, the peaks are sufficiently large. When a high pass filter as shown in FIG. 7 is used as the band pass filter 6, when the FET (field effect transistor) is turned on by the signal from the control circuit 30, the pass band is switched to the high frequency side.

Furthermore, the angle of view of the screen changes during zooming, specifically, the longer the focal length, the smaller the angle of view, and the less likely the edge of the subject will enter the screen, the lower the output of BPF6, and vice versa. The longer the focal length, the larger the angle of view, and the more likely the edge of the subject will enter the screen.
BPF6 output increases. Therefore, when the focal length becomes longer due to zooming, the prediction value of the apex prediction circuit 20 is lowered by the signal of the control circuit 30, and when the focal length becomes shorter, the apex prediction circuit 20 of the apex prediction circuit 20 is made by the signal of the control circuit 30. Correct to increase the predicted value.

Since the other components are the same as those in the example of FIG. 4, detailed description thereof will be omitted.

C. Operation of Embodiment Next, the operation of the embodiment of the present invention shown in FIG. 2 will be described with reference to FIG.

As described above, the information on the focusing lens 2 is detected by the lens information sensor 29, and the detected lens information is supplied to the control circuit 30. The control circuit 30 receives the focus voltage extracting filter (BP
F) The control signal for changing the predetermined values of 6 and the gate circuit 7 and for correcting the predicted value in the top prediction circuit 20 is output.

The light of the subject 1 is imaged on the photoelectric conversion means 4 by the lens 2, and the video signal converted into an electric signal is the BPF 6, the gate circuit 7, the detector 9 and the integrator as in the conventional example described above with reference to FIG. Sample and hold circuit (S / H) 11 via device 10
Enter in and hold for each field. The output of the sample hold circuit 11 is the first field delay circuit 1 in the subsequent stage.
4, the first comparator 23, the first subtraction circuit 21, the fourth comparator 27, the prediction circuit 20, and the hill climbing circuit 16, respectively.

The focus voltage B of the previous field delayed by one field by the first field delay circuit 14 is two subtraction circuits 21,2.
2 and the peak predicting circuit 20 and also to the second field delay circuit 19. The focus voltage A of the previous field, which is further delayed by one field by the second field delay circuit 19, is supplied to the second subtraction circuit 22 and the peak prediction circuit 20.

The subtraction output (CB) of the first subtraction circuit 21 is the second comparator 24.
Is supplied to the inverting input terminal of the third comparator 25, and the output of the second subtraction circuit 22 is supplied to the non-inverting input terminal of the third comparator 25. The current field focus voltage C output from the sample hold circuit 11 is compared with a predetermined threshold value V _{ref I} in the first comparator 23, and when C> V _{ref I}
A high level signal is output from 23 to the hill climbing circuit 16 and the AND circuit 26. The output (CB) of the first subtraction circuit 21 is compared with a predetermined threshold value V _{ref II} in the second comparator 24, and _CB
When <V _{ref II} , the comparator 24 outputs a high level signal to the AND circuit 26. Furthermore, the output (CB) of the first subtraction circuit 21 and the output (BA) of the second subtraction circuit 22 are the third
The comparator 25 compares them, and when CB <B-A, the comparator 25 outputs a high level signal to the AND circuit 26.

When all the judgment outputs of the above-mentioned comparators 23, 24 and 25 are at the high level, that is, when all the following conditional expressions that indicate that they are approaching sufficiently close to the maximum focus voltage P are satisfied, the AND circuit 26 It becomes a high level, and the top prediction circuit 20 is activated.

C> V _{ref I} C−B <V _{ref II} C−B <B−A In other words, as shown in FIG. 3, the focus voltage C of the current field is sufficiently high (C> V _{ref I} ) and the previous field is When the difference between the focus voltage B of the current field and the focus voltage B of the current field is less than a predetermined level (CB−V <V _{ref II} ), that is, when the slope of the peak is not so steep and is sufficiently close to the peak P, If the slope of the mountain determined in this field is smaller than the slope of the previous mountain (C-
B <B-A), it is approaching the summit of the mountain and is sufficiently close to the summit of the mountain. Therefore, by operating the summit prediction circuit 26, the focus voltage (estimated focus voltage) P at the summit of the summit and It is possible to easily predict how many fields will elapse before the peak can be predicted by a simple quadratic function or the like. For example, y = ax ² + bx + c is used as this quadratic function, where x is the number of fields and y is the focus voltage.

In response to the output of the AND circuit 26 described above, the top prediction circuit 20 operates, and from the input focus voltages A, B and C, these
Based on a quadratic function with A, B, and C as parameters, the maximum focus voltage P at the peak corresponding to the in-focus state and the number of fields after which the maximum focus voltage P will be reached are predicted, and those predicted values are also predicted. Is corrected by the lens information supplied from the control circuit 30. For example, when the zoom lens is moving, the focal length and the like change from moment to moment, so the previous period predicted value P after a few fields may not match and may go up or down.
This predicted value is corrected in the top prediction circuit 20 based on the lens information.

The peak predicting circuit 20 outputs the timing signal after N fields, which is predicted to reach the peak in N fields later, to the fourth comparator 27 as a gate signal and outputs the predicted maximum focus voltage P to the non-inverting input terminal of the comparator 27. Output to. The fourth comparator 27 opens the gate with the gate signal and outputs a motor stop control signal to the motor control circuit 28 under the condition of P ≧ C.

In the hill climbing circuit 16, the current field focus voltage C and the previous field focus voltage B are C> B as shown in FIG.
The motor 18 is rotated in the forward direction when
A control signal for reversing is output to the motor control circuit 28.
The motor control circuit 28 rotates the motor 18 in the forward direction or the reverse direction according to the output signal of the hill climbing circuit 16 and, if the output of the AND circuit 26 is at a low level, rotates the motor 18 at a high speed and outputs the output of the AND circuit 26. If is a high level, the rotation of the motor 18 is switched to a low speed rotation. Therefore, the motor control circuit 28 rotates the motor 18 at a high speed to bring the motor 18 closer to the in-focus position when the top prediction circuit 20 is not operating, that is, when only the hill climbing circuit 16 is operating. The rotation is switched to stop the motor 18 according to the motor stop control signal output from the comparison circuit 27 when the in-focus state is reached. Therefore, according to this embodiment, the lens 2 does not vibrate near the in-focus position, and the lens 2 can be stopped accurately at the in-focus position.

It should be noted that the present invention is not limited to outputting the signal corresponding to the focus degree from the output of the image pickup means as described above. For example, when a line sensor indicating the focus degree is used separately from the image pickup means. Can also be applied to.

[Effects of the Invention] As described above, according to the present invention, the in-focus point is predicted by taking lens information into account by three or more signals that are adjacent to the signal corresponding to the in-focus degree, and the predicted value thereof is calculated. Since the movement / stop of the taking lens is controlled according to the above, small vibrations around the in-focus position (top of the mountain) are eliminated, and the effect of accurate focus detection can be obtained.

As a result, even if the aperture value and the focal length party change, the depth of field changes, and the characteristics of the focus signal change, the in-focus predicted value is corrected according to the information of these imaging lenses. As a result, it is possible to always perform accurate and highly reliable focus detection regardless of the shooting state.

Further, according to the present invention, more accurate automatic focus detection can be performed by changing, for example, various constants of the focus voltage detection filter and the gate circuit according to the detected lens information.

[Brief description of drawings]

FIG. 1 is a block diagram showing the basic structure of the embodiment of the present invention, FIG. 2 is a circuit diagram showing the circuit structure of the embodiment of the present invention, and FIG. 3 is a focal length for explaining the operation of the embodiment of the present invention. FIG. 4 is a characteristic diagram showing the relationship with the focus voltage, FIG. 4 is a block diagram showing the circuit configuration of the conventional example, and FIG. 5 is a characteristic diagram showing the operation of the conventional example. FIG. 6 is an explanatory diagram used for explaining the operation of the control circuit 30 of the embodiment of the present invention, and FIG. 7 is a circuit diagram showing an example of the configuration of the filter 6. 3 ... Photoelectric conversion means, 6 ... Bandpass filter, 7 ... Gate circuit, 8 ... Window pulse forming circuit, 9 ... Detector, 10 ... Integrator, 11 ... Sample and hold circuit, 14, 19 ...... 1 field delay circuit, 16 ...... hill climbing circuit, 20 ...... top prediction circuit, 21,22 …… subtraction circuit, 23,24,25,27 …… comparator, 26 …… and circuit, 28 …… motor Control circuit, 29 …… Lens information sensor, 30 …… Control circuit.

Claims (3)

[Claims] 1. A signal extraction means for sequentially extracting and outputting a signal corresponding to a focus degree of an image pickup lens at a predetermined timing, and a focus degree using a plurality of the signals output by the signal extraction means. Predicting means for predicting, and a detecting means for detecting at least one of lens information such as aperture value and focal length of the imaging lens, and predicting value of the predicting means according to the lens information detected by the detecting means. A focus detection device comprising: a correction unit that corrects 2. The apparatus according to claim 1, wherein the correction means adjusts various constants of a filter and a gate circuit which constitute the signal extraction means according to the lens information. Detection device. 3. The apparatus according to claim 1 or 2, further comprising control means for activating the predicting means when the signal extracted by the signal extracting means is above a predetermined level. A focus detection device characterized by the above.