JPH07298120A - Automatic focusing adjustment device - Google Patents

Abstract

PURPOSE:To attain excellent automatic focusing by correcting a focal signal with a luminance difference signal and discriminating the focusing. CONSTITUTION:The signal is subject to sampling and holding, and amplified synchronously with a clock CO from a reference clock generator and the light through a lens 1 is converted into a signal SO by a converter 5. Signals S1, S2 are obtained by a 1H delay device 6 and a 2H delay device 7 and they are given to a chrominance signal generating circuit, added by an adder 8 and only a luminance signal S3 is extracted by a low pass filter LPF 10 and a signal S4 is generated from the signal S0. An automatic focus adjustment signal S5 is obtained from the signals S3,S4. Then the luminance signal and the chrominance signal are separated and the signal S5 is obtained and it is a signal S6 through gamma conversion. The signal is given to a TE-LPF14 whose cut off frequency is high and a TE-LPF15 whose cut off frequency is low, in which frequency components are extracted. A low frequency component is extracted based on each filter characteristic. Excellent automatic focusing is realized by correcting the focal signal with the luminance difference signal and discriminating the focusing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic focusing device using a video signal suitable for a video camera or the like.

[0002]

2. Description of the Related Art Conventionally, as an automatic focusing device used in a video device such as a video camera, a high frequency component is extracted from a video signal obtained from an image pickup device such as a CCD, and the high frequency component is maximized. A so-called hill-climbing method is known in which a focusing lens is driven so as to adjust the focus.

Such an automatic focus adjustment system does not require a transmitter for emitting infrared rays or ultrasonic waves for focus adjustment, a receiver, and a special optical member, and also detects focus based on the sharpness of a subject image. Since it is performed, it has an advantage that the subject can be accurately focused regardless of the distance regardless of the distance.

FIG. 14 shows an example of an automatic focus adjustment system using this type of video signal. In the figure, 1
Is a focusing lens, which is moved in the optical axis direction by the lens driving motor 57 to perform focusing.

The light passing through this lens is imaged on the image pickup surface of the image pickup device 3, photoelectrically converted into an electric signal, and outputted as a video signal. This video signal is sampled and held by a double correlation sampling / AGC circuit (CDS / AGC) 4 and then amplified to a predetermined level, converted by an A / D converter 5 into digital video data, and stored in a camera. It is input to the process circuit and converted into a standard television signal, and a band pass filter (hereinafter BPF) 10
Input to 0.

In the BPF 100, the high frequency component in the video signal is extracted, the gate circuit 101 extracts only the signal corresponding to the portion set in the focus detection area in the screen, and the peak hold circuit 102 extracts an integral multiple of the vertical synchronizing signal. Peak hold is performed at intervals synchronized with.

Hereinafter, the peak-held value will be referred to as an AF evaluation value S20. Based on the AF evaluation value S20, the motor speed determination circuit 104 sets the focusing speed according to the degree of focus. That is, the motor speed is changed by instructing the motor driver 56 to be fast during large blur and slow during small blur.

In the motor direction determination circuit 103, A
The motor drive direction is set in the direction in which the F evaluation value increases.
By performing the above operation continuously, so-called hill climbing control is performed.

However, according to the above-mentioned hill climbing control, if the focus lens is not moving, it is not possible to determine the front focus and the rear focus, and if the focusing lens is once stopped, the lens is stopped. There was a problem that the driving direction was lost and the next time the camera was restarted, it started in the wrong direction and became in a state of large blur, and it took a long time to focus.

Therefore, the image signal is modulated by an actuator such as a piezo element or by vibrating the focus lens by the focus motor at all times back and forth in the optical axis direction. By detecting the phase by detecting, the front focus, rear focus, and focus determination can be performed regardless of whether or not the focus lens is driven.
There is provided a so-called modulation type automatic focus detection device that can always make a determination, and the problem of the direction determination in the hill climbing system as described above has been tentatively solved.

Further, as another direction determining means, there is also a means for making a direction determination by slightly vibrating the focus lens forward and backward in the optical axis direction when the direction determination is required when the focus lens is moving. .

However, in the current autofocus system equipped with an actuator that cannot perform high-speed modulation operation, the modulation cycle becomes long, and as a result, the modulation operation takes a long time, so focus recognition after panning (discrimination of whether or not focus is achieved). ) Takes a long time, and there is a problem that it is not possible to accurately determine the in-focus direction because the start of high-speed mountain climbing is delayed, or the subject changes during the modulation operation.

Further, a signal fluctuation due to the influence of camera shake or the like is mixed in the focus evaluation signal, and a high-speed hill climbing operation is performed even though the focus is in focus, and a good autofocus system cannot be realized.

Further, these problems are such that if an actuator that can be driven at high speed and has high positioning accuracy can be used, the modulation cycle thereof is short, so that more modulation operations can be performed within a unit time and almost always. Since the focus lens can be modulated, the direction of the in-focus point can be detected more accurately, and even if the drive direction is mistaken, the direction can be corrected quickly. Expected to be done.

However, it is difficult to realize a small-sized and lightweight actuator which can be incorporated in a video camera at a high speed and with high accuracy as described above. Even if such an actuator can be mounted, it will be extremely expensive. However, it is impossible to realize it as a consumer device in terms of cost.

Further, in the automatic focus adjusting device for controlling the hill climbing so that the level of the focus evaluation signal is maximized as described above, when trying to perform the hill climbing operation by using the high frequency component in the distance measuring frame, depending on the subject, the hill climbing operation may be performed. During the operation, the image blur causes the contrast to move in and out of the focus detection frame, and the focus evaluation signal is greatly reduced even though the focus is in the focusing direction. Instead, the focus evaluation signal increases, and as a result, hunting is repeated at a position that is not the focus, and the focus is not reached unless the subject is changed.

Further, when there is a perspective competing object in the distance measuring frame, in some cases, as described above, a maximum point is formed at a point that is not in focus, hunting is repeated, and neither subject is in focus. . Therefore, conventionally, a large hill climbing reversal threshold has been set and a value has been set so that the above-mentioned hunting operation does not occur to prevent malfunction.

In this way, however, it is possible to go through the above-mentioned maximum point and reach the true focusing point, but there is a problem that the overshoot amount becomes large when the direction is wrong and the quality becomes extremely poor. .

[0019]

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a focus system capable of high-precision, high-speed and accurate focusing control in a focus system that cannot perform high-speed wobbling, even if high-speed wobbling cannot be performed. It is to be able to realize a special autofocus system.

Another object of the present invention is to make good use of the advantages of both the focus evaluation value having a sensitive direction detection ability and the directional accuracy of the evaluation value which surely increases in the in-focus direction. In order to prevent such a problem, it is possible to stably focus on a target main subject even under any subject or shooting condition, to realize a stable, highly accurate and good malfunction-free autofocus system.

[0021]

In order to solve such a problem, according to the invention described in claim 1 of the present application, a focus signal including a predetermined frequency component in the image pickup signal which changes according to the focus state. And a gate means for extracting a signal corresponding to a focus detection area in the screen from the focus signals output from the filter means, and the focus signal extracted by the gate means. Is provided with a correction unit that corrects the maximum value of the brightness difference component with the maximum value of the brightness difference component in the focus detection area, and a focus adjustment unit that performs a focus adjustment operation based on the output signal of the correction unit. Since the in-focus determination is performed after correction by, it becomes possible to reliably determine whether in-focus or out-of-focus, and erroneously perform high-speed hill climbing operation when in-focus.
If the subject is out of focus for a moment, or if it is out of focus and it is determined to be in focus and then stops, the problem can be solved.

According to a second aspect of the present invention, the filter means is composed of a plurality of filters having different pass bands, and the maximum output value is output for each horizontal scanning line of the screen. Since it is configured, it is possible to perform focus detection for each horizontal scanning line and for a plurality of frequency components.

According to the third aspect of the present invention, the focus detection area is composed of a plurality of small areas, and the filter means can extract the focus signal for each of the small areas. It is possible to perform an accurate focus adjustment operation regardless of the shape and position of the.

According to the fourth aspect of the present invention, since the filter means is configured to selectively switch filters of different pass bands between even scan lines and odd scan lines, a plurality of scan lines are scanned. Since the signals are switched every time, a large number of signals having a plurality of frequency components can be obtained in a short period of time.

According to a fifth aspect of the present invention, a filter means for extracting a focus signal containing a predetermined frequency component which changes according to the focus state from the image pickup signal, and the focus output from the filter means. From the signal, gate means for extracting a signal corresponding to the focus detection area in the screen, and the maximum value of the focus signal extracted by the gate means is the maximum value of the luminance difference component in the focus detection area. A correction means for correcting and a focus adjusting means are vibrated in a predetermined cycle, and a direction determining means for determining the focus direction from the change of the focus signal according to the vibration, an output of the direction determining means, and the correction And a control means for controlling the operation of the focus adjusting means on the basis of the output of the means, so that the focus signal is corrected by the brightness difference signal. Since the approximate focusing degree is obtained without modulating the data, etc., and this is used together with the wobbling operation, it is possible to more reliably determine whether it is in-focus or out-of-focus as compared with the case of only the wobbling operation. When you are in focus, you accidentally perform a high-speed mountain climbing motion, and if you blur for a moment, or if it is out of focus, it will be decided that it is in focus and it will stop it can.

According to a sixth aspect of the present invention, the direction determining means determines from the level change of the focus signal when the focusing lens is slightly vibrated back and forth in the optical axis direction.
The configuration is such that it distinguishes between front focus, focus, and rear focus.

According to a seventh aspect of the present invention, in the case where the control means determines that the determination result of the direction determining means and the focus determination from the output of the correction means are both in focus. , I configured to stop the focus lens, so I accidentally perform high-speed hill-climbing operation even when the subject is in focus.
If the image is out of focus for a moment, or if it is out of focus and then stops when it is determined to be in focus, the problem can be solved, and a good autofocus system can be always realized.

Further, according to the invention described in claim 8, in the case where both of the determination result of the direction determining means and the in-focus determination result from the output of the correcting means by the control means are out of focus. In addition, since the focus lens is configured to be driven at high speed, the high-speed hill climbing operation is mistakenly performed when the subject is in focus, and the subject may be out of focus for a moment or may be out of focus. If it stops, some problems can be solved and a good autofocus system can be always realized.

According to the invention described in claim 9, one of the determination result of the direction determining means and the result of the in-focus determination from the output of the correcting means of the control means is in focus, and the other is out of focus. When it is in focus, since it is configured to perform focus detection based on the output of the direction determination means, the high-speed hill climbing operation is mistakenly performed even when the focus is in focus, and a momentary blur occurs. If it is out of focus but stops when it is determined to be in focus, the problem can be solved and an always good autofocus system can be realized.

According to a tenth aspect of the present invention, a filter means for extracting a predetermined frequency component from the image pickup signal outputted from the image pickup means, and a peak for holding the output of the filter means for each horizontal scanning line. Hold means, integrating means for vertically integrating the output of the peak hold means, one or more gate means for taking out the output of the integrating means corresponding to the focus detection area in the screen, and one in the screen Extraction means for extracting the maximum value of the brightness difference component in one or a plurality of focus detection areas, focus adjustment means, the focus signal extracted by the gate means, and the maximum value of the brightness difference component of the extraction means output And a control unit that determines the focus adjustment direction of the focus adjustment unit, the change in the subject is not affected by the subject or environmental conditions. Stomach can be quickly detected,
The target main subject can be stably focused on all subjects and shooting conditions.

According to an eleventh aspect of the present invention, the filter means is composed of a plurality of filters having different pass bands, and the frequency components in the even horizontal scanning lines and the odd horizontal scanning lines are detected. Since it is configured to, even line peak integral evaluation value, odd line peak integral evaluation value is particularly sensitive to direction detection, but on the other hand, a focus evaluation signal that reacts sensitively to a little subject coming in and going out and easily causes hunting, Since the focus adjustment is configured by using both of the features of the direction accuracy of the brightness difference evaluation value that surely increases in the in-focus direction, it is sensitive to the even line peak integrated evaluation value and the odd line peak integrated evaluation value. It is possible to make good use of the advantages of both the direction detection capability and the direction accuracy of the brightness difference evaluation value that surely increases in the focusing direction.

[0032]

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

FIG. 1 is a diagram showing the configuration of an embodiment of the present invention. In the figure, 1 is a focusing lens,
The lens driving motor 57 is moved in the optical axis direction to perform focusing. The light that passed through this lens is the diaphragm 2
The amount of light is controlled to an appropriate value, and an image is formed on the image pickup surface of the image pickup device 3, and photoelectrically converted into an electric signal.

This photoelectrically converted signal is read out in synchronization with the clock C0 output from the reference clock generator 60, sampled and held by the CDS / AGC 4, and at the same time amplified with an optimum gain to obtain A / With D converter 5,
It is converted into a digital signal S0 synchronized with the reference clock C0.

A 1H delay signal S1 is formed by delaying the digital signal S0 by one horizontal period delay unit (hereinafter referred to as 1H delay unit) 6 for one horizontal period. The 1H delay signal S1 is generated by the 1H delay unit 7. By further delaying by one horizontal period, the 2H delay signal S22 is formed.

These S0, S1 and S2 are input to a color signal generation circuit (not shown) of the camera and at the same time, S0
From S2 and S2, only the luminance signal component is extracted by the low pass filter (hereinafter referred to as LPF) 10 which is added by the adder 8 to remove the color signal component, and the coefficient unit 11 sets a predetermined coefficient (0.5).
A multiplied signal S3 is formed. From the signal S0,
Similarly, the LPF produces a signal S4 in which only the luminance signal is extracted.

These signals S3 and S4 are input to a luminance signal generation circuit (not shown) of the camera and, at the same time, are added by the adder 12 to generate an automatic focus adjustment signal S5.

That is, the 1H delay units 6 and 7, the adder 12,
The coefficient unit 11, the LPF 9, and the LPF 10 separate the luminance signal and the chrominance signal from the image pickup signal, supply them to their respective processing circuits, and generate the automatic focus adjustment signal S5 for extracting the focus signal. .

The automatic focusing signal S5 is supplied to the gamma circuit 1
3 and is gamma-converted with a polygonal line gamma curve having characteristics as shown in FIG. 6 to emphasize the low-luminance component and suppress the high-luminance component, thereby making the emission intensity of the cathode-ray tube proportional to the luminance signal level. A signal S6 corrected as above is generated.

The gamma-converted signal S6 is input to a TE-LPF 14 which is an LPF having a high cutoff frequency and an FE-LPF 15 which is an LPF having a low cutoff frequency, and respective predetermined frequency components are extracted.

TE-LPF14 and FE-LPF15
The respective filter characteristic values of are the microcomputer computer interface (hereinafter referred to as the microcomputer interface) 53.
It is determined and supplied by a system control micro computer (hereinafter referred to as a microcomputer) 55 via the.

The low frequency components are extracted based on the filter characteristic values set in this way, and the TE-LPF14 output signal S7 and the FE-LPF15 output signal S8 are formed.

The switch 16 is a signal Lin for identifying whether the horizontal scanning line is an even-numbered line or an odd-numbered line.
The switching is controlled by the eE / O, and the signal S7 and the signal S8 are selectively switched to a high pass filter (hereinafter referred to as HPF) 17
To enter.

That is, the even-numbered line outputs the signal S7 to HPF1.
7, the odd line passes signal S8 to HPF 17.

In the HPF 17, only the high frequency components in the signals S7 and S8 are extracted based on the filter characteristic values of the odd line / even line determined by the microcomputer 55 via the microcomputer interface 53, and the absolute value circuit A positive signal S9 is formed by making the absolute value at 18.

The absolute value signal S9 is input to the peak hold circuits 25, 26, 27 and the line peak hold circuit 31 for detecting the peak value of each horizontal line.

The frame generation circuit 54 generates gate signals L frame, C frame, and R frame for focus adjustment at positions on the screen as shown in FIG.

FIG. 5 illustrates the setting position of the focus adjustment frame on the screen. As is apparent from the movement, nine rectangular frames are set on the screen, but the number, size, and position of these can be appropriately changed according to the design. Therefore, the horizontal scanning line is moved alternately from the left side to the right side and from the top side to the bottom side of the screen alternately with the even number line and the odd number line to detect the information in each frame.

The peak hold circuit 25 includes a frame generation circuit 5
A LineE / 0 signal, which is a signal for identifying whether the 4-output L frame and the horizontal line are even-numbered or odd-numbered, is input, and as shown in FIG. 5, the top left LR1 and LR2 that are the heads of the focus adjustment L-frames are input. , LR3, the L frame peak hold circuit 25 is initialized, that is, the contents are reset.
The signal S9 in each frame of either the even line or the odd line designated through the microcomputer interface 53 from the microcomputer 55 is peak-held, and IR1, IR2, IR
At each position of 3, the peak hold value in the frame is transferred to the buffer 28 to generate the TE / FE peak evaluation value.

Similarly, the C-frame peak hold circuit 26 receives the C-frame output from the frame generation circuit 54 and the LineE / 0 signal, and CR1 at the upper left, which is the head of the C frame for focus adjustment shown in FIG. At each of CR2 and CR3, the peak hold circuit 26 is initialized, that is, the contents are reset.
The signal S9 in each frame of either the even line or the odd line designated through the microcomputer interface 53 from the microcomputer is peak-held, and the peak-hold value in the frame is transferred to the buffer 29 at each position of IR1, IR2 and IR3. Then, a TE / FE peak evaluation value is generated.

Similarly, the R frame peak hold circuit 27 is supplied with the R frame and the LineE / 0 signal output from the frame generation circuit 54, and the upper left RR1 which is the head of the focus adjusting R frame shown in FIG. , RR2, RR3 at respective timings, the peak hold circuit 27 is initialized, that is, the contents are reset, and the signal S9 in each frame of either the even line or the odd line designated from the microcomputer through the microcomputer interface 53 is transmitted. Peak hold, IR
The peak hold value in the frame is transferred to the buffer 30 at the timing of each position of 1, IR2, IR3 to generate the TE / FE peak evaluation value.

To the line peak hold circuit 31, the signal S9 and the L frame, C frame, and R frame of the output of the frame generation circuit 54 are input, initialized or reset at the horizontal start point in each frame, and each frame is initialized. The peak value of the signal S9 in one line is held. Integrating circuits 32, 33, 34, 35,
The line peak hold circuit 31 output and the LineE / 0 signal, which is a signal for discriminating whether the horizontal line is an even number or an odd number, are input to 36 and 37, and at the same time, the integration circuits 32 and 35 include a frame generation circuit. The L frame output corresponding to the left side of the screen output from 54, the C frame output corresponding to the central part of the screen output from the frame generating circuit 54 to the integrating circuits 33 and 36, and the frame generating circuit to the integrating circuits 34 and 37. The R frame outputs corresponding to the right side of the screen output from 54 are input.

The integrator circuit 32 initializes or resets the integrator circuit 32 at each of the upper left LR1, LR2 and LR3, which is the head of the focus adjustment L frame, and immediately before the end of the even lines in each frame. The output of the line peak hold circuit is added to the internal register, and the peak hold value is transferred to the buffer 38 at each of IR1, IR2 and IR3 to generate the line peak integral evaluation value.

The integrator circuit 33 initializes or resets the integrator circuit 33 at each of the upper left CR1, CR2, and CR3, which is the head of the focus adjustment C frame, and immediately before the end of the even lines in each frame. The output of the line peak hold circuit is added to the internal register, and the peak hold value is transferred to the buffer 39 at each position of IR1, IR2 and IR3 to generate the line peak integral evaluation value.

The integrator circuit 34 initializes or resets the integrator circuit 34 at each of the upper left RR1, RR2, and RR3, which is the head of the focus adjustment R frame, and immediately before the end of the even lines in each frame. The output of the line peak hold circuit is added to the internal register, and the peak hold value is transferred to the buffer 40 at each position of IR1, IR2 and IR3 to generate the line peak integral evaluation value.

The integrating circuits 35, 36 and 37 add the data of the odd lines respectively instead of the adding of the data of the even lines by the integrating circuits 32, 33 and 34 respectively, and the result is given to the buffers 41, 42 and 43, respectively. To transfer.

Further, the signal S7 is the peak hold circuit 1
9, 20, 21 and the line maximum value hold circuit 44 and the line minimum value hold circuit 45. The L frame output from the frame generation circuit 54 is input to the peak hold circuit 19, and the peak hold circuit 19 is initialized at each of the upper left LR1, LR2, and LR3, which is the head of the L frame, and the Peak hold signal S7, IR1, I
At each location of R2 and IR3, the peak hold result is transferred to the buffer 22, and the Y peak evaluation value which is the peak value of the brightness Y is generated.

Similarly, the C-frame output from the frame generation circuit 54 is input to the peak hold circuit 20, and the peak-hold circuit 20 is initialized at each of the upper left CR1, CR2 and CR3, which is the head of the C frame. Then, the signal S7 in each frame is peak-held, and at each place of IR1, IR2, IR3,
The peak hold result is transferred to the buffer 23, and the Y peak evaluation value is generated.

Similarly, the R frame output from the frame generation circuit 54 is input to the peak hold circuit 21, and the peak hold circuit 21 is initialized at each of the upper left RR1, RR2, and RR3, which is the beginning of the R frame. Signal S in each frame
7 is peak-held, and the peak-hold result is transferred to the buffer 24 at each of IR1, IR2, and IR3, and Y
Generate a peak rating. Line maximum value hold circuit 4
The L frame, the C frame, and the R frame output from the frame generation circuit 54 are input to the 4 and the line minimum value hold circuit 45, initialized at the horizontal start point in each frame, and the signal S7 in each frame is output. The maximum value and the minimum value of each line are held.

The maximum value and the minimum value held by these are input to the subtraction 46, the subtraction (maximum value Max-minimum value Min) is performed, and the signal S10 is calculated,
It is input to the peak hold circuits 47, 48 and 49. The L frame output from the frame generation circuit 54 is input to the peak hold circuit 47, and the top left LR1, LR2, and L of the L frame are input.
At each location of R3, the peak hold circuit 47 is initialized, the signal S10 in each frame is peak-held, and IR
The peak hold result is transferred to the buffer 50 at each of the positions 1, IR2 and IR3 to generate the Max-Min evaluation value.

Similarly, the C frame output from the frame generation circuit 54 is input to the peak hold circuit 48, and the peak hold circuit 48 is initialized at each of the upper left CR1, CR2, and CR3, which is the head of the C frame. , The signal S10 in each frame is peak-held, and at each position of IR1, IR2, IR3,
The peak hold result is transferred to the buffer 51, and Max-
Generate a Min value.

Similarly, the R frame output from the frame generation circuit 54 is input to the peak hold circuit 49, and the peak hold circuit 49 is initialized at each of the upper left RR1, RR2, and RR3, which is the head of the R frame. Signal S1 in each frame
0 is peak-held, the peak-hold result is transferred to buffer 52 at each of IR1, IR2, and IR3, and M
Generate an ax-Min evaluation value.

At the timing of each location of IR1, IR2 and IR3, buffers 22, 23, 24, 28, 29 and 3 are provided.
0,38,39,40,41,42,43,50,5
At the same time as transferring the data to 1, 52, the frame generation circuit 54
Sends an interrupt signal to the microcomputer 55.

Upon receipt of this interrupt signal, the microcomputer 55 sends a buffer 2 via the microcomputer interface 53.
2,23,24,28,29,30,38,39,4
Each data in 0, 41, 42, 43, 50, 51, 52 is read until the processing of the lower frame is completed and the next peak hold data is transferred to the buffer.

The microcomputer drives the motor 57 through the motor driver 56 based on the read data and the output of the aperture encoder 58 to move the focusing lens 1 in the optical axis direction for focusing. Also in FIG.
The area surrounded by the dotted line is realized with a one-chip IC.

FIG. 2 is a diagram for explaining the configuration of the TE-LPF. The signal S6 is latched by the register 70 at the timing of the output C0 of the reference clock generator 60.
The adder 71 adds the input signal and the output signal of the register 70 and outputs the added signal to the switch 79 and the register 72. Similarly, the registers 72, 74 and 76 also latch the input signal at the timing of C0.

The adder 73 adds the input signal and the output signal of the register 72 and outputs the addition result to the switch 79 and the register 74, and the adder 75 adds the input signal and the output signal of the register 72 to the switch. 79 and register 7
6, and the adder 77 adds the input signal and the output signal of the register 76 and outputs the result to the switch 79.

The register 78 stores the filter characteristic value from the microcomputer 55 through the microcomputer interface 53. The switch 79 selects a signal based on the output of the register 78 and outputs it as a signal S7.

FIG. 3 is a diagram for explaining the configuration of the FE-LPF. In the FE-LPF 15, the signal S6 output from the gamma circuit 13 is multiplied by K in the coefficient unit 80 and input to the adder 81. The outputs of the coefficient multipliers 80 and 83 are input to the adder 81 and are added to the register 82.
Is output to. The coefficient "K" indicates a filter characteristic value "K" which will be described later.

The register 82 latches the above addition result supplied from the adder 81 according to the output C0 of the reference clock generator 60, and outputs it to the coefficient unit 83 and the coefficient unit 84. The coefficient unit 83 multiplies the addition result output from the register 82 by (1−K) and supplies the result to the adder 81.

The coefficient units 84 and 87, the adder 85, and the register 86 perform the same operations as the coefficient units 80 and 83, the adder 81, and the register 82, respectively, except that the input signal changes from S6 to the output of the register 82.

The register 88 stores the filter characteristic value K supplied from the microcomputer 55 through the microcomputer interface 53, and each coefficient unit 80, 83, 84, 8 is stored.
7 is supplied.

FIG. 4 is a diagram for explaining the structure of the HPF. The signal S9 is input to the register 90 and sequentially transferred to the registers 91, 92, 93, 94 in accordance with the output C0 of the reference clock generator 60. Register 9
Only one of the outputs 1, 92, 93 and 94 is selected by the switch 95 and input to the subtractor 96.

The subtractor 96 receives the signal S9 and the switch 95.
The signal S10 is output by calculating the difference between the outputs of the.

The register 98 stores the filter characteristic values of even-numbered lines from the microcomputer 55 via the microcomputer interface 53. Further, the register 99 stores the filter characteristic value of the odd line.

The switch 97 is a line E / 0 signal for identifying whether the current horizontal line is an even-numbered line or an odd-numbered line.
The output of the register 98 is transmitted as a selection signal of the switch 95 when the line is even, and the output of the register 99 is transmitted when the line is odd.

That is, two types of HPFs having different frequency bands can be realized by the even lines and the odd lines, and the information can be obtained in one field period, and the AF speed can be increased. . Further, in the present embodiment, the even line filter characteristic is set to have a higher pass frequency than the odd line filter characteristic.

FIG. 5 is a diagram for showing an image of a frame on the screen and for explaining the overall timing. The outer frame is an effective video screen output from the image sensor 3. The inner horizontally divided frame is a focus adjustment gate frame, which is an L frame on the left column, a C frame on the center column, and an R frame on the right column.
The frame is output from the frame generation circuit 54.

Further, in order to divide these frames into three parts in the vertical direction, the reset signal is divided into three parts for each of the L, C and R frames in one screen.
Output twice, LR1, LR2, LR3, CR1, CR2
CR3, RR1, RR2, RR3 are generated, the integration circuit,
Reset the peak hold circuit, etc. Further, the data transfer signals IR1, IR2, IR3 are generated and the respective integrated values and peak hold values are transferred to the respective buffers.

Scanning with even fields is shown by a solid line, and scanning with odd fields is shown by a dotted line. In both the even field and the odd field, the even line selects the output of the TE-LPF 14, and the odd line selects the output of the FE-LPF 15.

That is, considering the scanning of one entire screen, in scanning within the focus detection area (distance measuring frame), LR1, CR1,
RR1 resets the three areas at the left end of the upper part of the L frame, C frame, and R frame, and at the time of completing the scanning of these three areas, at IR1 when the even and odd lines in each of these three areas are reset. The peak value will be output.

Similarly, the peak value is output for each of the middle three areas starting with LR2, CR2 and RR2, and then for the lower three areas starting with LR3, CR3 and RR3.

By the above operation, the peak value of the focus signal (high frequency component) in the nine distance measuring frames can be obtained.

Therefore, the focus state can be detected over a wide range within the screen in a short time, and the distance measuring frame can be selected according to the photographing state.

As described above, according to the automatic focus adjustment circuit of the present invention, the evaluation value TEpe obtained by peak-holding the peak value of each of even-numbered lines and odd-numbered lines in the nine distance measuring frames is held.
ak hold, FEpeak hold, evaluation values obtained by integrating peak values of even lines and odd lines in each distance measuring frame for each frame TE Line peak sum, FE Line peak sum, peak of each line for each distance measuring frame Difference between maximum value and minimum value of hold value Max-Min evaluation value, luminance Y for each ranging frame
The peak hold value Ypeak hold of is generated. These AF evaluation signals are output to the external system control microcomputer 55 via the microcomputer interface 53, and A
Used for F control.

In addition, the above-mentioned circuit configuration is integrated by a one-chip IC, and an interrupt request is issued to the system control microcomputer 55 when processing based on the output of the automatic focus adjustment IC is performed. The signal INR is provided.

That is, since AF control is performed by interruption, parallel processing with other processing becomes possible as a whole system, and an efficient system can be realized.

FIG. 6 is a diagram showing the input / output characteristics of the gamma circuit 13 of FIG. The horizontal axis represents the input signal S5 and the vertical axis represents the output signal S6.

Next, TE / FE peak evaluation value, Max-M
An in-value, TE line peak integral evaluation value, FE line peak integral evaluation value, and Y peak evaluation value will be used to explain how the microcomputer performs a good automatic focus adjustment operation.

The TE / FE peak evaluation value is an evaluation value showing the degree of focusing on even lines and odd lines, respectively, and since it is a peak hold value which is not integrated and is close to real time, it is relatively independent of the subject and is affected by camera shake and the like. It is most suitable for the focus degree determination and the restart determination that require an instant determination.

Further, the TE line peak integral evaluation value and the FE line peak integral evaluation value also represent the degree of focus, but since they are stable evaluation values with little noise due to the integration effect, they are not easily affected by instantaneous noise, and the focus is slightly moved. It is most suitable for direction determination because the signal changes sensitively.

Further, both the peak evaluation value and the line peak integration evaluation value are high frequency components extracted by TE (even line), so they are optimal for focusing control and object discrimination near the in-focus point, and conversely FE. (Odd line) is most suitable for large blur far from focus.

Further, the luminance Y peak evaluation value and Max-Mi
Since the n evaluation value does not much depend on the focus degree but depends on the subject, it is optimal for grasping the situation of the subject in order to surely perform the focus degree determination, the restart determination, and the direction determination. That is, the Y-peak evaluation value is used to determine whether the subject is a high-luminance object or a low-illuminance object, the Max-Min evaluation value is used to determine whether the contrast is large, the TE / FE peak evaluation value, the TE line peak integration evaluation value, and the FE line peak. Optimal control is performed by predicting and correcting the size of the peak of the integrated evaluation value.

Next, the algorithm of the automatic focus adjustment operation in the present invention will be described with reference to FIGS.

First, in order to facilitate understanding, an algorithm of the automatic focusing operation before applying the present invention will be described. When the processing is started (101), first, a wobbling operation is performed in which the focusing lens is slightly vibrated back and forth in the optical axis direction, and the phase of the change in the focus evaluation value at that time is detected to determine whether it is in focus or out of focus. A determination is made (102).

When it is determined that the focal point is in focus, the focusing lens is stopped and a restart determination routine is performed to monitor whether or not the lens is restarted (108). The focusing lens is stopped until restarted. (109).

If the result of the processing of 103 is blurring, high-speed mountain climbing is performed in the direction of the determination result by the wobbling operation (104).

Then, it is determined whether or not the focal point, that is, the vertex of the focus evaluation signal is exceeded (105). If the focal point is not exceeded, high-speed hill climbing is continued. If the focal point is exceeded, the focusing lens is reversed and the vertex is reversed. Perform processing to return to (1
06, 107).

However, the subject may change due to panning or the like during the operation of returning to the apex, so if the focusing lens reaches the apex, the next place is the apex, that is, the focal point. In order to determine whether or not there is a wobbling operation, the process returns to step 102.

If it is determined that the object is in focus in this wobbling operation, restart monitoring is performed, and if it is determined that the object is not in focus, a high speed hill climbing operation is performed in the direction of the determination result. By repeating such an operation, the focusing lens operates so as to constantly maintain the focus.

However, as described above, the normal video camera cannot perform high-speed wobbling, which causes the following problems.

That is, in a focus system in which the wobbling cannot be performed at high speed, the wobbling cycle becomes long, and as a result, a large amount of unnecessary signal components are mixed due to camera shake during the wobbling and external vibration, and whether the peak is accurate. There is a problem that it cannot be determined.

For example, a high-speed hill climbing operation is erroneously performed despite being in focus, or a high focus hill climbing operation is not performed because it is determined that the subject is in focus but is out of focus, and a good focus system is constructed. Can not.

If a camera shake or the like is detected so that the determination by the wobbling operation can be performed accurately, the wobbling operation is redone due to a slight camera shake.
As a result, the wobbling operation is performed many times,
If it takes time to determine the focus, problems will occur.

These problems can be solved considerably if there is an actuator capable of wobbling at high speed and with extremely high precision. In other words, when performing minute vibrations for direction determination, if an actuator that can be driven at high speed and has high positioning accuracy can be used, the modulation cycle is short, so more modulation operations can be performed within a unit time. , It becomes possible to modulate the focusing lens almost always, it is possible to detect the direction of the in-focus point more accurately, and it is expected that even if the drive direction is wrong, the direction can be corrected quickly. This is because that.

However, it is difficult to realize a small-sized and lightweight actuator which can be incorporated in a video camera at a high speed and with high accuracy as described above. Even if such an actuator can be mounted, it will be extremely expensive. However, it is impossible to realize it as a consumer device in terms of cost.

The present invention realizes an autofocus system capable of performing high-precision, high-speed and accurate focusing control even if high-speed wobbling cannot be performed in a focus system that cannot perform such high-speed wobbling. Is what you can do.

That is, the TE / FE peak evaluation value is set to Ma
By normalizing with the x-Min evaluation value, it is possible to cover the wide dynamic range of the evaluation signal at the time of focusing and to roughly understand the in-focus state of various subjects.

Further, even in an autofocus system that cannot perform high-speed wobbling operation, by using the wobbling judgment result and the above-mentioned normalized value together, it is more accurate than in the case where the focus judgment is made only by the wobbling judgment result. The focus state of can be grasped, and a good autofocus system can be realized.

Here, the TE / FE peak evaluation value is Max-
It will be described that the focusing degree can be roughly determined without performing the wobbling operation by normalizing with the Min evaluation value.

FIG. 9 shows the luminance signal of a certain horizontal scanning line when the subject is in focus. The waveform of (1) in the figure is a luminance signal when in focus, and the waveform of (2) is obtained by delaying the waveform of (1) by several clocks of the synchronization clock. When the waveform of (2) is subtracted from the waveform of (1), the waveform of (3) is obtained, and a pulse of height A is output. That is, this is the TE / FE peak evaluation value.

Since the Max-Min evaluation value is the maximum value-minimum value of the luminance signal, it becomes the level A of the waveform of (1).

That is, as shown in the figure, it is found that the TE / FE peak evaluation value is 1 when the TE / FE peak evaluation value is divided by the Max-Min evaluation value for a subject having a contrast with an infinite inclination.

However, since the luminance signal does not rise ideally in this way, the waveform height of (3) does not actually become equal to the waveform height of (1).

That is, TE / FE peak evaluation value / Max
-The Min evaluation value becomes a constant value of 1 or less.

However, as described above, this normalized value represents the degree of focusing. TE / F
Line and M when the peak is held at the E peak evaluation value
The line when peak-holding with the ax-Min value is not always the same line, so the above theory does not always hold.

However, the sharpness is high and Max-Min is high.
Since the TE / FE peak evaluation value increases as the subject having a large evaluation value comes into focus, the line holding the Max-Min evaluation value at the peak and the line holding the TE / FE peak evaluation value often coincide with each other. . As described above, the normalized value can grasp the in-focus state to some extent without performing a modulation operation such as a wobbling, and therefore, when used in combination with the wobbling operation, the subject determination becomes more accurate.

Further, if the focus determination is performed only by the wobbling as in the past, an incorrect signal is included in the focus evaluation signal due to camera shake or the like, and it is determined that the focus is out of focus in spite of being in focus, and the high-speed mountain climbing operation is performed. However, if this normalized value indicates in-focus, it is prohibited to shift to high-speed hill climbing operation, and the normalized value is determined to be in-focus even if the wobbling result is determined to be vague. In this case, the high-speed mountain climbing operation is not performed.

By operating in this way, the stability at the time of focusing is increased. If the wobbling result is determined to be in-focus despite being blurred, the focus motor will stop and cause blurring unless the normalized value is used, but this normalized value is unclear. If it is determined that the focus motor is stopped, the wobbling is performed again to prevent the blur from stopping. That is, the focusing lens is stopped only when both results are determined to be in focus.

As described above, when the determination result of the wow ring does not match the determination result of the normalized value, it is prohibited to change the control of the focusing lens, so that it is more accurately focused or unfocused. Can be determined, and a more stable autofocus system can be realized.

FIG. 8 shows a flow chart of the control of the autofocus system when a normalized value is used, which is a feature of this embodiment.

When the processing is started (201), first, it is judged whether or not the position at which the focusing lens is currently at the in-focus point by using the wobbling operation of the focusing lens (202).

Then, the above-mentioned normalized value is calculated (20
3) If the determination result of the wobbling is determined to be in-focus, then it is determined whether the normalized value represents the in-focus state (210).

That is, only when it is determined that both the wow ring determination result and the normalized value are in focus, the restart determination is performed (211) and it is monitored whether or not the object deviates from the in-focus point. Then, the subject is changed, and the focusing lens is held in the stopped state until the restart is performed (21
2).

When the wobbling determination result and the normalized value are not equal to each other, that is, in the process of 204, it is determined that the result of the wobbling is in focus, but in the process of 210, the determination by the normalized value is in focus. If it is not, it may not be in perfect focus, but it may still be in focus, so instead of immediately moving to high speed mountain climbing, it moves to the process of 202 and the normalization value and wobbling judgment result Both wobbling are repeated until it is determined that they are in focus.

If it is determined that the object is not in focus as a result of the wobbling in the processing of 204, the in-focus determination is made by the normalized value (205), but if it is also determined that the object is not in focus, That is, when neither the wobbling determination result nor the normalized value is determined to be in-focus, the hill climbing operation is performed in the direction of the wobbling determination result (20
6) It is controlled to reach the top of the focus evaluation value signal.

That is, it is judged whether or not the in-focus point, that is, the apex of the in-focus evaluation signal is exceeded (207). If it is not exceeded, the high-speed mountain climbing is continued, and if it is exceeded, the focusing lens is reversed and the apex is reached. The processing for returning to (208, 209) is performed.

If it is determined in the processing of 205 that the in-focus state is obtained from the normalized value, the in-focus state is not complete, but the in-focus state may still be present. Therefore, immediately go to high-speed mountain climbing. Is not performed, the process proceeds to the process of 202, and wobbling is repeated until it is determined that both the normalized value and the wobbling determination result are in focus.

Here, the wobbling operation will be described with reference to FIG. In the figure, two driving states are shown, one with and without a flicker such as a fluorescent lamp.

What is common to both operations here is that the focusing lens is driven in a certain direction, the driving direction is not changed when the signal is increased, and the driving direction is changed when the signal level is decreased. It is to invert. That is, the focusing lens is always driven in the direction in which the focus signal increases.

By being controlled in this way, the lens is continuously driven in the direction of the in-focus point during the wobbling operation. Therefore, although the speed is slow, the in-focus point can be finally reached only by repeating the wobbling operation.

As described above, by using the focus determination based on the normalized value and the focus determination based on the wobbling together, a more accurate focus determination can be performed as compared with the case where the focus determination is performed only on the wobbling. It becomes possible to realize a more stable autofocus system.

Further, since this normalized value represents the degree of focusing, the focus speed should be slow in the area where the value is determined to be in-focus, and the focus speed should be increased in the area where it is determined to be unfocused. By controlling it, it can be used for speed control of autofocus.

<< Second Embodiment >> Next, the second embodiment of the present invention will be described.
An example will be described. According to the above-described first embodiment, the focus detection operation based on the direction determination operation of the focusing lens is performed with high accuracy and reliability, and as a result, the speed of focus adjustment is increased and the accuracy is improved. is there.

According to the automatic focus adjusting device for controlling the hill climbing such that the high frequency component in the video signal is used as the focus evaluation signal and the level of the focus evaluation signal is maximized, the high frequency in the distance measurement frame is increased. When attempting to perform a hill climbing motion using a component, the subject may focus in the middle of the hill climbing motion due to image blurring, causing contrast to appear in and out of the focus detection frame, even though the subject is oriented in the focusing direction. The evaluation signal greatly decreases, or the focus evaluation signal increases even though it is in the direction of large blurring, and as a result, hunting is repeated at a position that is not the in-focus point. If you do not reach the focus, problems will occur.

When there is a perspective competing subject in the distance measuring frame, in some cases, as described above, a maximum point is formed at a point that is not the in-focus point, and hunting is repeated. Problems occur. Then, conventionally, a large threshold for hill climbing reversal is set and a value is set so that the above-mentioned hunting operation does not occur to prevent malfunction.

However, in this way, it is possible to go beyond the above-mentioned maximum point and reach the true focal point,
There is a problem that the overshooting amount becomes large when the direction is wrong and the quality becomes extremely poor.

The second embodiment shown below is to solve the above-mentioned problems, and is characterized in that a certain frequency component is extracted from the image pickup signal output from the image pickup means. Filter means, peak hold means for peak-holding the output of the filter means only for each focus detection area in the screen for each line, and integrating means for vertically integrating the output of the peak hold means in the focus detection area in the screen in the vertical direction. An autofocus system that includes an extraction unit that extracts a brightness difference component of a focus detection area in the screen, and that performs automatic focus adjustment based on the output of the extraction unit and the output of the integration unit. It is designed to stably focus on the intended main subject under the conditions.

FIG. 11 shows an algorithm of the automatic focus adjustment operation before applying the present embodiment in order to facilitate understanding of the present embodiment.

In the figure, when the processing is started (3
01), first, by performing the wobbling operation (3
02), it is determined whether the focus is on or out of focus (303).

When it is determined that the focal point is in focus, the focusing lens is stopped and a restart determination routine is performed to monitor whether or not it has been restarted (310), and the focusing lens is stopped until restarted. Hold the state (31
1).

As a result of the determination in 303, if the subject is out of focus and is out of focus, high speed hill climbing is performed in the direction of the result determined by the wobbling operation (394).

Then, it is judged whether or not the in-focus point, that is, the apex of the in-focus evaluation signal has been exceeded (305), and if it does not exceed the apex, a shift is made to a monitoring routine to check whether the in-focus evaluation signal has decreased. (308) If the focus evaluation signal has decreased by a predetermined amount, the driving direction is reversed (30
9) The hill climbing operation is continued again, and if the focus evaluation signal has not decreased by a predetermined amount, the hill climbing operation is continued with the driving direction unchanged.

In the process of 305, if the vertex is exceeded, the operation of returning to the vertex is performed (306). Then, it is determined whether or not the apex is reached (307), and when it is determined that the apex is reached, the process returns to the process of 302 and wobbling is performed.

In this operation, the subject may change due to panning or the like during the operation of returning the focusing lens to the apex. Therefore, if the focusing lens reaches the apex, the operation is next. However, the wobbling operation is performed to determine whether or not the vertex is really the focal point.

If it is determined in this wobbling operation that the object is in focus, restart monitoring is performed (310). If it is determined that the object is not in focus, a high-speed hill climbing operation is performed in the direction of the determination result (3).
04). By repeating such an operation, the focusing lens operates so as to constantly maintain the focus.

However, even if the operation is performed as described above, depending on the subject, the contrast may move into and out of the distance measuring frame due to the image blur during the hill climbing operation, and the object may be oriented in the focusing direction. Instead, the focus evaluation signal decreases, or the focus evaluation signal increases even if the focus evaluation signal is facing the large blur direction, and as a result, hunting is repeated at a position that is not in focus, and the subject is not changed. As long as the focus is not reached, problems will occur.

When there is a perspective competing subject in the distance measuring frame, in some cases, as described above, a maximum point is formed at a point that is not in focus, and hunting is repeated until both subjects are in focus. Problems occur. In the past, a large threshold for hill climbing inversion was set and a value was set so that the above-mentioned hunting operation did not occur to prevent malfunction.

However, in this way, it is possible to go through the above-mentioned maximum point and reach the true focusing point, but there is a problem that the amount of overshoot becomes large when the direction is wrong and the quality becomes extremely poor. .

Therefore, in this embodiment, the increase and decrease of the TE line peak integral evaluation value and the FE line peak integral evaluation value and the increase and decrease of the luminance difference signal are monitored together to prevent the above-mentioned problems during hill climbing, It is designed to perform good automatic focus adjustment.

That is, the TE line peak integral evaluation value, F
If hunting is repeated only by monitoring the increase or decrease in the E-line peak integral evaluation value, or if the reversal threshold is increased to prevent hunting, if you make a mistake in the direction, it will be slightly blurred before focusing and good autofocus. The waste system could not be realized. In this embodiment, in addition to these signals, the increase / decrease of the Max-Min evaluation value is monitored to prevent the above-mentioned hunting.

FIG. 12 shows the luminance signal of the subject. In this figure, the Max-Min evaluation value when there is a large blur in a focus detection area such as C is B, and as the focusing lens approaches focus, the slope of the luminance signal increases and the Max-Min evaluation value becomes It becomes like A.

As the focus approaches, Max-Mi
It can be seen that the n evaluation value gradually increases and is saturated near the in-focus point. Especially in a region that is not near the focus, Max-
The rate of change of the Min evaluation value is large, and it is optimal for the determination of the focusing direction in the hill climbing operation.

FIG. 13 shows a flow chart showing the operation of this embodiment. In the figure, when the process is started (4
01), first, by performing the wobbling operation (4
02), it is determined whether it is in-focus or out of focus (403).

When it is determined that the focal point is in focus, the focusing lens is stopped and the restart determination routine is performed to monitor whether or not the lens has been restarted (408), and the focusing lens is stopped until restarted. Hold the state (40
9).

If it is determined in step 403 that the image is out of focus, high-speed hill climbing is performed in the direction of the determination result by the wobbling operation (404).

Then, it is judged whether or not the focal point is exceeded (405). If it does not exceed the in-focus point, the focus evaluation signal shifts to a monitoring routine for monitoring increase / decrease (41
0) If the focus evaluation signal has decreased by a predetermined amount, it is monitored whether the brightness difference signal has decreased by a predetermined amount (411).

Here, the drive direction is inverted only when the brightness difference signal has decreased by a predetermined amount, that is, when both the focus evaluation signal and the brightness difference signal have decreased by a predetermined amount (412), and in other cases. The hill climbing operation is continued while keeping the driving direction (404).

When it is determined in step 405 whether or not the vertex is exceeded, if it is determined that the vertex is exceeded, the operation of returning to the vertex is performed (406) and it is determined whether or not the vertex is reached ( 406) If it is confirmed that the vertex has been reached, the process proceeds to 402 and the wobbling operation is performed.

This is because the subject may change due to panning or the like during the operation of returning to the apex, so if the focusing lens reaches the apex, the next place is the apex, that is, the focal point. This means performing a wobbling operation for determining whether or not

If it is determined in this wobbling operation that the subject is in focus (403), restart monitoring is performed (408), and if it is determined that it is not in focus, a high speed hill climbing operation is performed in the direction of the determination result (404). . By repeating such an operation, the focusing lens operates so as to constantly maintain the focus.

By operating in this way, the TE line peak integral evaluation value and the FE line peak integral evaluation value can be reliably captured in the focusing direction without deteriorating the direction detecting ability.

Conventionally, TE line peak integral evaluation value, FE
The focus evaluation signal such as the line peak integral evaluation value is particularly sensitive to the direction detection, but also sensitively reacts to the comings and goings of a small subject, and easily causes the hunting as described above. However, as in the present invention, the advantages of both the sensitive direction detection ability of the TE line peak integrated evaluation value and the FE line peak integrated evaluation value and the directional accuracy of the Max-Min evaluation value that surely increases in the in-focus direction are well achieved. By using it, the above-mentioned problems can be prevented and a good autofocus system can be realized.

[0164]

As described above, according to the first aspect of the present invention, since the focus signal is corrected by the brightness difference signal before the focus determination is made, the focus signal is reliably detected. It becomes possible to determine whether it is in focus or out of focus, and when it is in focus, it mistakenly performs a high speed mountain climbing motion, blurs for a moment, or it is out of focus It is possible to solve such problems and always realize a good autofocus system.

According to the invention described in claim 2 of the present invention, in addition to the invention of claim 1, focus detection can be performed for each horizontal scanning line and for a plurality of frequency components. Regardless of the state, it is possible to always realize stable and highly accurate focus detection operation.

According to the invention of claim 3 of the present invention, in addition to the invention of claim 1, since the focus detection area is constituted by a plurality of blocks, the shape and position of the object are not affected. Therefore, it is possible to perform a precise focus adjustment operation without malfunction.

According to the invention of claim 4 of the present invention, in addition to the invention of claim 2, since a plurality of filters are switched for each scanning line, signals of a plurality of frequency components can be supplied in a short period of time. In addition, a large number can be obtained, and more precise focus adjustment can be performed.

According to the fifth and sixth aspects of the present invention, the approximate focus can be achieved without modulating the focus motor or the like with the correction value obtained by correcting the focus signal with the brightness difference signal. Since it is used in combination with the wobbling operation, it is possible to more reliably determine whether the focus is in focus or out of focus as compared to the case of only the wobbling operation. A high-speed hill climbing operation can be performed to solve a problem when the camera suddenly blurs for a moment, or if it is out of focus but stops when it is judged to be in focus, it can always solve a problem and realize a good autofocus system. be able to.

According to the seventh to ninth aspects of the present invention, when both the focus signal correction value and the wobbling determination result are in focus, the focusing lens is stopped and the focus signal correction value When both of the wow ring judgment results are out of focus, the focusing lens is searched at high speed, and when one is out of focus, focus detection is performed only by the wobbling without performing high speed search. Therefore, if you are in focus but you accidentally perform a high-speed mountain climbing motion, you will be out of focus for a moment, or if it is out of focus, it will be decided that it is in focus and it will stop. It is possible to always realize a good autofocus system.

According to the tenth aspect of the present invention, a signal obtained by vertically integrating a focus signal peak-held for each horizontal scanning line in the focus detection area,
Since automatic focus adjustment is performed based on the brightness difference component of the focus detection area, changes in the subject can be detected quickly without being affected by the situation of the subject or environment, and any subject or shooting conditions can be detected. Can focus on the desired main subject in a stable manner.

According to the eleventh aspect of the present invention, although it is particularly sensitive to the direction detection of the even line peak integral evaluation value, the odd line peak integral evaluation value, etc., it is also suitable for the entry and exit of a small subject. Since it is configured to perform focus adjustment by using both the focus evaluation signal that reacts sensitively and is likely to cause hunting, and the feature of the direction accuracy of the brightness difference evaluation value that surely increases in the focus direction, The advantages of both the sensitive direction detection ability of the even line peak integrated evaluation value and the odd line peak integrated evaluation value and the directional accuracy of the brightness difference evaluation value that surely increases in the focusing direction can be used successfully. It is possible to prevent such problems, realize a stable, highly accurate, good malfunction-free autofocus system.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an automatic focus adjustment device of the present invention.

FIG. 2 is a diagram showing a configuration of a TE-LPF filter used in the present invention.

FIG. 3 is a diagram showing a configuration of an FE-LPF filter used in the present invention.

FIG. 4 is a diagram showing a configuration of an HPF filter used in the present invention.

FIG. 5 is a diagram for explaining the setting operation of the ranging frame within the screen and the sampling timing of each signal according to the embodiment of the present invention.

FIG. 6 is a diagram showing a gamma characteristic of the embodiment of the present invention.

FIG. 7 is a flow chart showing the processing of the apparatus that is the premise of the present invention.

FIG. 8 is a flow chart showing the operation of the first embodiment of the present invention.

FIG. 9 is a diagram for explaining the operation of the first exemplary embodiment of the present invention.

FIG. 10 is a diagram for explaining the operation of the first exemplary embodiment of the present invention.

FIG. 11 is a flow chart showing the processing of the device that is the premise of the present invention.

FIG. 12 is a diagram for explaining the operation of the second exemplary embodiment of the present invention.

FIG. 13 is a flow chart showing the operation of the second embodiment of the present invention.

FIG. 14 is a block diagram showing a configuration of a conventional automatic focus adjustment device.

[Explanation of symbols]

 1 Focusing lens 3 CCD 6 1H delay device 7 1H delay device 14 TE-LPF (for even line) 15 FE-LPF (for odd line) 19-21 Peak hold circuit 25-27 Peak hold circuit 32-37 Integration circuit 47 to 49 Peak hold circuit 53 Microcomputer interface 55 Microcomputer 56 Focus motor driver 57 Focus motor

Claims (11)

[Claims] 1. A filter means for taking out a focus signal including a predetermined frequency component which changes according to a focus state from an image pickup signal, and a focus signal output from the filter means.
Gate means for extracting a signal corresponding to the focus detection area in the screen, and correction means for correcting the maximum value of the focus signal extracted by the gate means with the maximum value of the brightness difference component in the focus detection area. And an automatic focus adjusting device for performing a focus adjusting operation based on the output signal of the correcting means. 2. The filter means according to claim 1, wherein the filter means is composed of a plurality of filters having different pass bands, and is configured to output the maximum value of the output for each horizontal scanning line of the screen. An automatic focus adjustment device characterized in that 3. The automatic focus adjustment according to claim 2, wherein the focus detection area comprises a plurality of small areas, and the filter means can extract a focus signal for each of the small areas. apparatus. 4. The automatic focusing apparatus according to claim 2, wherein the filter means is configured to selectively switch filters of different pass bands between even scan lines and odd scan lines. 5. A filter means for extracting a focus signal containing a predetermined frequency component that changes according to the focus state from the image pickup signal, and the focus signal output from the filter means,
Gate means for extracting a signal corresponding to the focus detection area in the screen, and correction means for correcting the maximum value of the focus signal extracted by the gate means with the maximum value of the brightness difference component in the focus detection area. The focus adjustment means is vibrated at a predetermined cycle, and the direction determination means for determining the focus direction from the change of the focus signal according to the vibration, the output of the direction determination means, and the output of the correction means. Based on the control means for controlling the operation of the focus adjusting means,
An automatic focusing device characterized by being equipped with. 6. The direction determining means according to claim 5, wherein the front focus state, the in-focus state, and the rear focus state are determined from the level change of the focus signal when the focusing lens is slightly vibrated back and forth in the optical axis direction. An autofocus device that is configured to. 7. The focus control device according to claim 5, wherein the control unit controls the focus lens when both the determination result of the direction determination unit and the focus determination result from the output of the correction unit are in focus. An autofocusing device that is configured to stop. 8. The focus lens according to claim 5, wherein the control means determines that both the determination result of the direction determining means and the result of focus determination based on the output of the correction means are out of focus. An automatic focusing device, characterized in that it is configured to drive at high speed. 9. The control unit according to claim 5, wherein one of the determination result of the direction determination unit and the result of the in-focus determination based on the output of the correction unit is in-focus and the other is out-of-focus. The automatic focus adjusting device is configured to perform focus detection based on the output of the direction determining means. 10. A filter means for extracting a predetermined frequency component from an image pickup signal output from the image pickup means, a peak hold means for holding a peak of the output of the filter means for each horizontal scanning line, and an output of the peak hold means. In the vertical direction, one or more gate means for extracting the output of the integrating means corresponding to the focus detection area in the screen, and the brightness in one or more focus detection areas in the screen Extracting means for extracting the maximum value of the difference component; focus adjusting means; and the focus adjusting direction of the focus adjusting means based on the focus signal extracted by the gate means and the maximum value of the brightness difference component of the output of the extracting means. An automatic focus adjusting device comprising: a control unit for determining. 11. The filter according to claim 11, wherein the filter means is composed of filters having different pass bands, and is configured to detect the frequency components in even horizontal scanning lines and odd horizontal scanning lines, respectively. Automatic focus adjustment device.