JP2014021151A - Focus adjustment device and focus adjustment method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a focus adjustment device and focus adjustment method which shorten focus adjustment time by appropriately setting a decrease threshold value even after the start of focus and ensure high stability.SOLUTION: In the beginning of a focus adjustment operation, focus adjustment values are calculated from the high frequency component of an image signal obtained by an imaging element 2 and the component of a frequency lower than the high frequency component. The focus evaluation values calculated in the beginning of the focus adjustment operation are stored as reference evaluation values (S13). The focus evaluation values are calculated (S25) from the high frequency component and low frequency component each time an image signal is obtained by the imaging element 2 after the start of the focus adjustment operation. According to the magnitude relation between each reference evaluation value and the corresponding focus evaluation value, the threshold value for detecting the peak of the focus evaluation value is calculated (S29). Using the calculated threshold value for detecting the peak, the peak of the focus evaluation value is determined.

Description

  The present invention relates to a so-called hill-climbing servo system focus adjustment apparatus and focus adjustment method.

  In the conventional automatic focus adjustment device using the hill-climbing servo system, the focus lens is actually driven to detect the maximum focus evaluation value (also called contrast value). It is necessary to detect a certain amount of decrease. However, since the focus evaluation value characteristic changes depending on the subject and the shooting environment, if the amount of decrease from the maximum focus evaluation value, that is, the reduction threshold value is uniquely determined, the focus lens is determined before the maximum focus evaluation value is determined. The maximum overrun amount increases, and it takes time to detect the focus.

  In order to solve this problem, for example, in the autofocus device disclosed in Patent Document 1, the current focus evaluation value and the previous focus evaluation value, and the previous focus evaluation value and the previous focus evaluation value. Thus, the difference between the evaluation values is calculated, and the decrease threshold value from the maximum value is switched between the case where both of the difference values are larger than the predetermined threshold value and the case where one of them is equal to or less than the threshold value. By this processing, the focus lens overrun amount from the maximum evaluation value is reduced.

Japanese Patent Laid-Open No. 08-125913

  However, in the method disclosed in Patent Document 1 described above, the decrease threshold is determined from the evaluation value in the vicinity of the maximum value, and therefore a change immediately after the start of focus is not included. The evaluation value difference is implemented by one type of high-pass filter. For this reason, there is a case where the threshold selection filter cannot detect these disturbances immediately after the focus is started, that is, even when disturbances such as blurring due to vibration occur when the release button is pressed. Depending on the selection of the threshold value, the focusing accuracy may be deteriorated.

  For example, FIGS. 6 and 7 show changes in the focus evaluation value when the focus lens is driven from the infinite direction toward the closest direction. In either case, since the maximum focus evaluation value exists in the current lens driving direction, the focus direction is determined to be the “forward direction”. Here, under an ideal environment, as shown in FIG. 6, the focus evaluation value LLR of the image signal processed by the low-pass filter and the focus evaluation value HLR of the image signal processed by the high-pass filter It continuously increases from the focus evaluation value (reference evaluation values HVR1, LVR1) at the position (LA1) to the maximum value (HVR5, LVR5).

  However, in the example shown in FIG. 7, the image signal processed by the low-pass filter is decreased from the reference evaluation value (LV2) to the next evaluation value (LV1) due to the influence of camera shake, and thereafter. There is a place where the evaluation value decreases to the maximum value (evaluation value LV2 at the lens position LB4). Furthermore, in the image signal processed by the high-pass filter, the reference evaluation value (HV1) increases from the next evaluation value (HV2), but the evaluation value decreases to the maximum value due to camera shake. (Evaluation value HV3 at lens position LB4). Here, although the evaluation value (HV3) of the high-pass filter is decreased at the lens position LB4 in FIG. 7, the reduction threshold for the high-pass filter at this time is the same as the reduction threshold (ThA) in FIG. The lens position LB3 is erroneously determined as the peak position.

  The present invention has been made in view of such circumstances, and even immediately after the start of focus, by appropriately setting a decrease threshold, the focus adjustment node time is shortened and a highly stable focus adjustment device and An object is to provide a focus adjustment method.

  In order to achieve the above object, a focus adjustment apparatus according to a first aspect of the present invention includes an imaging unit that generates an image signal by forming an optical image of an optical system on an imaging element, and repeatedly while moving the optical system. In a focus adjustment device that performs imaging to acquire an image signal and performs a focus adjustment operation based on the image signal, a high-pass filter unit that passes a high-frequency component of the image signal, and an image signal that is higher than the high-pass filter unit A low-pass filter that passes a lower frequency component, a focus evaluation value calculator that calculates a focus evaluation value based on the output of the low-pass filter unit or the high-pass filter unit, and a position at which a focus adjustment operation is started An increase / decrease detection unit for detecting an increase / decrease direction and an increase / decrease rate between a focus evaluation value calculated based on the image signal acquired in step 1 and a focus evaluation value calculated based on the image signal acquired thereafter; A continuity detection unit that detects continuity of a plurality of focus evaluation values calculated based on a plurality of image signals acquired from a position at which an operation is started in a predetermined period, and an increase / decrease direction detected by the increase / decrease detection unit A control unit that sets a threshold for determining the peak of the focus evaluation value based on the increase / decrease rate and the continuity detected by the continuity detection unit and performs a focus adjustment operation based on the peak of the focus evaluation value Are provided.

  The focus adjustment apparatus according to a second invention is the focus adjustment apparatus according to the first invention, wherein the increase / decrease detection unit detects the increase / decrease direction and the increase / decrease rate of the focus evaluation value based on the output of the low-pass filter unit. The sex detection unit detects the continuity of the focus evaluation value based on the output of the low-pass filter unit, and the control unit detects the increase / decrease direction and the rate of increase / decrease detected by the increase / decrease detection unit, and the continuity detection A threshold value for detecting a peak of the focus evaluation value based on the output of the high-pass filter unit is set based on the continuity detected by the unit, and a focus adjustment operation based on the peak of the focus evaluation value is performed.

  The focus adjustment apparatus according to a third invention is the focus adjustment apparatus according to the first invention, wherein the increase / decrease detection unit detects the increase / decrease direction and the increase / decrease rate of the focus evaluation value based on the output of the high-pass filter unit. The sex detection unit detects the continuity of the focus evaluation value based on the output of the high-pass filter unit, and the control unit detects the increase / decrease direction and the rate of increase / decrease detected by the increase / decrease detection unit, and the continuity detection Based on the continuity detected by the unit, a threshold for determining the peak of the focus evaluation value based on the output of the high-pass filter unit is set, and the focus adjustment operation based on the peak of the focus evaluation value is performed.

  In the focus adjustment apparatus according to a fourth aspect of the present invention, in the first to third aspects of the invention, the control unit sets a reduction amount from the maximum value of the focus evaluation value, and the focus evaluation value is set from the maximum value. The maximum value is determined as a peak when the amount of decrease exceeds the amount of decrease.

  According to a fifth aspect of the present invention, there is provided a focus adjustment method for calculating a focus evaluation value from a high frequency component of an image signal acquired by an image sensor and a lower frequency component than the high frequency component at the start of a focus adjustment operation, The focus evaluation value calculated at the start of the focus adjustment operation is stored as a reference evaluation value, and each time an image signal is acquired by the image sensor after the start of the focus adjustment operation, the high frequency component and the low frequency component respectively. A focus evaluation value is calculated, a threshold value for peak detection of the focus evaluation value is calculated according to the magnitude relationship and continuity between the reference evaluation value and the focus evaluation value, and the threshold value for peak detection is used. The peak of the focus evaluation value is determined.

  According to the present invention, it is possible to provide a focus adjustment device and a focus adjustment method that can shorten the focus adjustment node time and have high stability by setting the decrease threshold appropriately even immediately after the start of focus.

1 is a block diagram mainly showing an electrical configuration of a camera according to an embodiment of the present invention. It is a block diagram regarding the reduction | decrease threshold value calculation of the camera concerning one Embodiment of this invention. 5 is a flowchart showing an auto focus (AF) operation of the camera according to the embodiment of the present invention. It is a flowchart which shows the operation | movement of the reduction | decrease threshold value calculation of the camera concerning one Embodiment of this invention. It is a flowchart which shows the operation | movement of the reduction | decrease threshold value calculation of the camera concerning one Embodiment of this invention. 6 is a graph showing a change in focus evaluation value under an ideal environment in a conventional camera and a camera according to an embodiment of the present invention. 6 is a graph showing changes in focus evaluation values when camera shake occurs in a conventional camera and a camera according to an embodiment of the present invention.

  A preferred embodiment will be described below using a camera to which the present invention is applied according to the drawings. The camera according to a preferred embodiment of the present invention is a digital camera, and is an imaging device having an imaging unit that forms an optical image by an optical system on an imaging device and generates an image signal based on the optical image, Imaging is repeated while moving the optical system to acquire an image signal, and focus adjustment is performed based on the image signal. That is, it has an imaging unit (corresponding to an imaging element 2 and the like described later), and the imaging unit converts the subject image into an image signal. Further, while moving the optical system from infinity to the near side or from the close to the infinity side, an image signal is acquired from the imaging unit, and based on this image signal, focus adjustment of the optical system is performed by a hill-climbing AF (contrast AF) method.

  Further, the decrease threshold for peak determination in hill-climbing AF includes the focus evaluation value maximum value determination direction, and the reference focus evaluation values (S13 in FIG. 3) of the low-pass filter and the high-pass filter (see 20, 21 in FIG. 2). Reference), the increase / decrease direction and increase / decrease rate of the focus evaluation value (see S17 in FIG. 3) of the lens position of the next frame (see 22, 24 in FIG. 2, S41, S51 in FIG. 4), and the maximum focus evaluation value determination The direction is set based on the continuity of the focus evaluation values for a certain period (23 and 25 in FIG. 2, S61 and S71 in FIG. 5) from the reference focus evaluation values of the low pass filter and the high pass filter.

  FIG. 1 is a block diagram mainly showing an electrical configuration of a camera according to an embodiment of the present invention. This camera includes an optical system 1, an image sensor 2, AGC / CDS3, A / D4, a signal processing unit 5, a lens control unit 6, an imaging control unit 7, an AE processing unit 8, an image processing unit 9, a CPU 10, and a focus evaluation value. It comprises a generation unit 11, a bus 12, and the like.

  The optical system 1 is composed of a plurality of optical lenses for forming a subject image, and is a single focus lens or a zoom lens. The optical system 1 is connected to the lens control unit 6, and the lens control unit 6 drives the focus lens of the optical system 1 in the optical axis direction in accordance with a control signal from the CPU 10 input via the bus 12.

  The image sensor 2 is disposed on the optical axis of the optical system 1 at a position where a subject image is formed. The image sensor 2 captures a subject image formed by the optical system 1. In this imaging device 2, photodiodes constituting each pixel are two-dimensionally arranged in a matrix, and each photodiode generates a photoelectric conversion current corresponding to the amount of received light. Charges are accumulated by a capacitor connected to.

  The image pickup device 2 is connected to the image pickup control unit 7, and the image pickup control unit 7 performs charge accumulation control of the image pickup device 2, image signal readout control, and the like in accordance with a control signal from the CPU 10 input via the bus 12. I do. The imaging control unit 7 repeatedly outputs an image signal from the imaging element 2 and outputs the output image signal (also referred to as image data) to the AGC / CDS 3.

  The AGC / CDS3 is a circuit that performs preprocessing of an image signal. The image signal amplification processing (auto gain control) and correlated double sampling for removing amplifier noise and reset noise (Correlated Double sampling). Sampling) and the like.

  The image signal preprocessed by the AGC / CDS 3 is output to the A / D 4. A / D 4 is an analog-digital converter that converts an image signal into digital data and outputs the digital data to the signal processing unit 5. The signal processing unit 5 performs digital processing of the image signal converted into digital data, and outputs it to the bus 12.

  The AE processing unit 8 inputs the image signal output to the bus 12 and calculates subject luminance information corresponding to the subject luminance. The luminance information is calculated in accordance with instructions from the CPU 10 such as average photometry, center-weighted photometry, spot photometry, and face spot photometry. The calculated luminance information is output to the CPU 10 via the bus 12. The CPU 10 calculates exposure control values such as a shutter speed, an aperture value, and ISO sensitivity based on the luminance information, etc., performs exposure control for live view display, and when the release button is fully pressed, Control of the shutter and aperture (not shown) is performed.

  The image processing unit 9 inputs the image signal processed by the signal processing unit 5 via the bus 12, performs image processing, and outputs the image signal subjected to the image processing to the bus 12. As image processing, noise reduction processing, white balance correction, synchronization processing, color conversion, gradation conversion (gamma conversion), edge enhancement, and YC conversion are performed. Also, compression processing of YC converted image data and decompression processing of compressed image data are performed.

  The focus evaluation value generation unit 11 inputs the image signal processed by the signal processing unit 5 via the bus 12, generates focus evaluation values of the low frequency component and the high frequency component of the image signal, and determines the focus in the CPU 10. To the unit 10a. That is, the focus evaluation value generation unit 11 includes a high-pass filter unit that passes a high-frequency component of the image signal, and a low-pass filter unit that passes a lower-frequency component of the image signal than the high-pass filter unit. The focus evaluation value is calculated based on the output of the high pass filter unit, and the focus evaluation value calculating unit calculates the focus evaluation value based on the output of the high pass filter unit.

  A CPU (Central Processing Unit) 10 connected to the bus 12 controls the entire camera according to a program stored in a non-volatile memory (not shown) such as a flash memory.

  Further, the CPU 10 has a focus determination unit 10a. The focus determination unit 10a inputs the low-frequency and high-frequency focus evaluation values calculated by the focus evaluation value generation unit 11, and maximizes the focus evaluation value. Thus, a lens drive command for driving the focus lens of the optical system 1 is output to the lens control unit 6 to move the optical system 1 in the optical axis direction. The focus determination unit 10a determines that the focus is in focus when the focus evaluation value decreases from the maximum value by a decrease threshold or more. In addition, although the focus determination part 10a may be comprised with a hardware, in this embodiment, it is implement | achieved by software.

  Further, the focus determination unit 10a calculates a focus evaluation value calculated based on the image signal acquired at the position where the focus adjustment operation is started and a focus evaluation value calculated based on the image signal acquired thereafter. It functions as an increase / decrease detection unit that detects an increase / decrease direction and an increase / decrease rate (see the low-frequency evaluation value increase / decrease detection unit 22 and the high-frequency evaluation value increase / decrease detection unit 24 in FIG. 2).

  Also, it functions as a continuity detection unit that detects the continuity of a plurality of focus evaluation values calculated based on a plurality of image signals acquired in a predetermined period from the position at which the focus adjustment operation is started (FIG. 2). Low frequency evaluation value continuity detection unit 23 and high frequency evaluation value continuity detection unit 25).

  Further, the focus determination unit 10a sets a threshold value for determining the peak of the focus evaluation value based on the increase / decrease direction and the increase / decrease rate detected by the increase / decrease detection unit and the continuity detected by the continuity detection unit. It also functions as a control unit that performs the focus adjustment operation based on the peak of the focus evaluation value that is set (see the threshold value calculation unit 32, the peak determination unit 33 in FIG. 2, and S29 to S35 in FIG. 3).

  Next, the decrease threshold value calculation control in the focus determination unit 10a will be described with reference to the block diagram shown in FIG. The low-pass filter evaluation value memory 20 is a memory for temporarily storing a focus evaluation value calculated based on an image signal from the low-pass filter unit of the focus evaluation value generation unit 11. The high pass filter evaluation value memory 21 is a memory for temporarily storing a focus evaluation value calculated based on an image signal from the high pass filter unit of the focus evaluation value generation unit 11. As described above, the image signal is repeatedly output from the image sensor 2. The focus evaluation value generation unit 11 calculates a focus evaluation value every time an image signal for one frame is output, and the focus evaluation value is stored in the evaluation value memories 20 and 21 each time the focus evaluation value is calculated. Stored.

  The low-frequency evaluation value increase / decrease detection unit 22 detects the increase / decrease direction and increase / decrease rate of the reference evaluation value and the focus evaluation value at the lens position of the next frame among the evaluation values stored in the low-pass filter evaluation value memory 20. The high frequency evaluation value increase / decrease detection unit 24 detects the increase / decrease direction and rate of increase / decrease of the reference evaluation value and the focus evaluation value of the lens position of the next frame among the evaluation values stored in the high frequency filter evaluation value memory 21. . The increasing / decreasing direction is increasing or decreasing, and the increasing / decreasing rate is a ratio between the reference evaluation value and the focus evaluation value of the lens position of the next frame.

  Here, the reference evaluation value is an evaluation value at the time when the AF operation is started by operating the release button or the like. In the example shown in FIG. 6, the evaluation values HVR1 and LVR1 at the lens position LA1 are shown in FIG. In the example, the evaluation values HV1 and LV2 at the lens position LB1 are referred to. Further, the focus evaluation value at the lens position of the next frame is after the reference evaluation value has been calculated. In the example shown in FIG. 6, the focus evaluation values HVR2 and LVR2 at the lens position LA2, and in the example shown in FIG. The focus evaluation values HV2 and LV1 at the position LB2.

  The low-frequency evaluation value continuity detection unit 23 detects the reference evaluation value and the focus evaluation value continuity for a predetermined period among the evaluation values stored in the low-pass filter evaluation value memory 20. The high-frequency evaluation value continuity detection unit 25 detects the reference evaluation value and the focus evaluation value continuity for a predetermined period among the evaluation values stored in the high-frequency filter evaluation value memory 21. As described above, the evaluation value memories 20 and 21 store evaluation values after the calculation of the reference evaluation value. Therefore, the continuity detection units 23 and 25 compare the evaluation value of the latest frame with the evaluation value of the frame immediately before the latest frame, and if the increase is continuous within a certain period, there is continuity. It is determined, and if it decreases in the middle, it is determined that there is no continuity.

  The threshold coefficient A calculation unit 26 calculates the threshold coefficient A based on the increase / decrease direction and the increase / decrease rate detected by the low-frequency evaluation value increase / decrease detection unit 22 and outputs the threshold coefficient A to the threshold calculation unit 32. The threshold coefficient B calculation unit 28 calculates the threshold coefficient B based on the increase / decrease direction and the increase / decrease rate detected by the high-frequency evaluation value increase / decrease detection unit 24, and outputs the threshold coefficient B to the threshold calculation unit 32 (S43, S47 in FIG. 4). , S53, S57). The threshold coefficients A and B may be calculated based on mathematical formulas, but in the present embodiment, the threshold coefficients A and B are stored in a table corresponding to the increase / decrease rate, and the corresponding coefficients A and B are stored. Is read.

  The threshold coefficient C calculation unit 27 calculates a threshold coefficient C based on the evaluation value continuity detected by the low frequency evaluation value continuity detection unit 23 and outputs the threshold coefficient C to the threshold calculation unit 32. The threshold coefficient D calculation unit 29 calculates the threshold coefficient D based on the evaluation value continuity detected by the high frequency evaluation value continuity detection unit 25, and outputs the threshold coefficient D to the threshold value calculation unit 32 (S63, S67 in FIG. 5). (See S73, S77). The threshold coefficients C and D may be calculated based on mathematical formulas. In the present embodiment, the threshold coefficients C and D are stored in a table corresponding to the evaluation value continuity, and the corresponding coefficient C is calculated. , D are read out.

  The maximum value determination unit 30 determines the maximum value among the plurality of focus evaluation values stored in the high frequency evaluation value memory 21. The reference threshold value table 31 stores a reference decrease threshold value as a table according to the focus evaluation value. The reference threshold value table 31 outputs a reference decrease threshold value corresponding to the maximum focus evaluation value determined by the maximum value determination unit 30 to the threshold value calculation unit 32.

  The threshold value calculation unit 32 is based on the threshold coefficient A to D calculated by the threshold coefficient A calculation unit 26, the threshold coefficient C calculation unit 27, the threshold coefficient B calculation unit 28, and the threshold coefficient D calculation unit 29, and the reference threshold value table 31. Based on the output reference decrease threshold, a final decrease threshold is calculated and output to the peak determination unit 33.

  The peak determination unit 33 determines whether or not the maximum value among the focus evaluation values output from the high-pass filter evaluation value memory 21 has decreased more than the final decrease threshold calculated by the threshold calculation unit 32. When it decreases to a value equal to or greater than the final decrease threshold, the lens position corresponding to the maximum value at that time becomes the focal point.

  As described above, in the focus determination unit 10a in the present embodiment, the low-frequency evaluation value increase / decrease detection unit 22 detects the increase / decrease direction and the increase / decrease rate of the focus evaluation value based on the output of the low-frequency filter unit, and the low-frequency evaluation value The continuity detector 23 detects the continuity of the focus evaluation value based on the output of the low-pass filter unit. The threshold calculation unit 32 and the peak determination unit 33 are based on the increase / decrease direction and the increase / decrease rate detected by the low-frequency evaluation value increase / decrease detection unit 22 and the continuity detected by the low-frequency evaluation value continuity detection unit 23. Thus, the threshold value for detecting the peak of the focus evaluation value based on the output of the high-pass filter unit is set, and functions as a control unit that performs a focus adjustment operation based on the peak of the focus evaluation value.

  Moreover, in the focus determination part 10a in this embodiment, the high-pass evaluation value increase / decrease detection part 24 detects the increase / decrease direction and increase / decrease rate of a focus evaluation value based on the output of a high-pass filter part, and continues high pass evaluation value The sex detection unit 25 detects the continuity of the focus evaluation value based on the output of the high-pass filter unit. The threshold calculation unit 32 and the peak determination unit 33 are based on the increase / decrease direction and the increase / decrease rate detected by the high frequency evaluation value increase / decrease detection unit 24 and the continuity detected by the high frequency evaluation value continuity detection unit 25. Thus, the threshold value for detecting the peak of the focus evaluation value based on the output of the high-pass filter unit is set, and functions as a control unit that performs a focus adjustment operation based on the peak of the focus evaluation value.

  Further, the threshold value calculation unit 32 and the peak determination unit 33 set a reduction amount from the maximum value of the focus evaluation value, and determine that the maximum value is a peak when the focus evaluation value decreases from the maximum value beyond the reduction amount. Functions as a control unit. That is, when the focus evaluation value stored in the high-pass filter evaluation value memory 21 is smaller than the final decrease threshold calculated by the threshold calculation unit 32, the peak determination unit 33 determines that the focus evaluation value at that time is a peak. .

  Next, the AF operation by the hill-climbing servo system in this embodiment will be described with reference to the flowcharts shown in FIGS. These flowcharts are executed by the CPU 10 according to a program stored in a memory (not shown).

  When the flow of AF operation shown in FIG. 3 is entered, first, pre-processing is performed (S11). Here, based on the luminance information from the AE processing unit 8 and the like, preprocessing such as AF exposure setting is performed.

  Once the preprocessing has been performed, next, a reference evaluation value is acquired (S13). Here, the first focus evaluation value is acquired after entering the AF operation. Specifically, based on the image signal from the image sensor 2, the focus evaluation value generation unit 11 calculates the focus evaluation value based on the outputs of the high-pass filter unit and the low-pass filter unit, and these values are set in the low-frequency range. They are stored in the filter evaluation value memory 20 and the high-pass filter evaluation value memory 21 (see FIG. 2). The reference evaluation value corresponds to the focus evaluation value at the lens position LA1 in the example shown in FIG. 6, and corresponds to the focus evaluation value at the lens position LB1 in the example shown in FIG.

  When the reference evaluation value is acquired, lens driving is then started (S15). Here, the lens control unit 6 drives the focus lens of the optical system 1 in a predetermined direction, that is, an infinite direction or a close direction.

  Once the lens driving is started, an evaluation value is acquired (S17). Here, every time an image signal for one frame is output from the image sensor 2, the focus evaluation value generation unit 11 generates a focus evaluation value and stores this value in the evaluation value memories 20 and 21.

  Once the evaluation value is acquired, it is next determined whether or not the direction is fixed (S19). Here, it is determined whether the direction is determined based on the focus evaluation value calculated based on the output of the low-pass filter unit. For example, it is determined whether the focus evaluation value continuously increases or decreases for a predetermined number of times from the acquisition of the reference evaluation value.

  If the result of determination in step S19 is that the direction has not been finalized, processing returns to step S17 and the aforementioned operation is performed. On the other hand, if the direction is fixed, it is next determined whether or not the direction is forward (S21). Here, the forward direction is a direction in which the focus evaluation value increases, while the reverse direction is a direction in which the focus evaluation value decreases.

  If the result of determination in step S21 is forward, the lens is driven as it is without changing the lens driving direction. On the other hand, if it is not the forward direction, the drive direction is reversed (S23). Since it is driven in the opposite direction, it is far from the peak of the focus evaluation value. Therefore, a control command is output to the lens control unit 6 so as to reverse the driving direction of the focus lens of the optical system 1.

  If the result of determination in step S21 is the forward direction, or if the drive direction is reversed in step S23, then an evaluation value is acquired (S25). Here, as in step S17, every time an image signal for one frame is output from the image sensor 2, the focus evaluation value generation unit 11 generates a focus evaluation value, and this value is used as the evaluation value memory 20, 21. To store.

  Once the evaluation value is acquired in step S25, it is next determined whether or not the maximum value has been updated (S27). Here, it is determined whether or not the maximum value is updated in the focus evaluation value based on the output of the high-pass filter unit stored in the high-pass filter evaluation value memory 21. The processing here corresponds to the function in the maximum value determination unit 30 in FIG.

  If the maximum value is updated as a result of the determination in step S27, a decrease threshold is calculated (S29). Here, the reduction threshold is calculated according to the reference evaluation value, the magnitude relationship between the focus evaluation values, and the continuity. Specifically, using the threshold coefficients A to D (see 26 to 29 in FIG. 2) and the reference decrease threshold (see the reference threshold table 31 in FIG. 2) corresponding to the maximum value determined in step S27, the peak is obtained. A reduction threshold value for determination is calculated. Detailed operation of the decrease threshold calculation will be described later with reference to FIGS. 4 and 5.

  If the decrease threshold value is calculated in step S29, or if the maximum value has not been updated as a result of the determination in step S27, then peak detection is performed (S31). Here, it is determined whether or not the peak has been detected using the decrease threshold value in the latest step S29. If the peak cannot be detected as a result of this determination, the process returns to step S25, the evaluation value is acquired again, and the above-described operation is executed.

  If the peak is detected as a result of the determination in step S31, then the in-focus position is calculated (S33). It is possible to determine that the peak has been detected after moving through the decrease threshold after passing the peak. Therefore, in this step, the in-focus position corresponding to the peak position is calculated by performing an interpolation calculation from the three points before and after.

  Once the in-focus position has been calculated, it is next driven to the in-focus position (S35). Here, a control command is output to the lens controller 6 so as to move to the in-focus position calculated in step S33. When this focus position driving is performed, the flow of the AF operation is ended.

  Next, the operation for calculating the decrease threshold in step S29 will be described with reference to FIGS. In this flow, the reduction threshold of the high-pass filter at the time of forward direction determination (maximum evaluation value is the current lens driving direction) is calculated.

  If the flow for calculating the decrease threshold value is entered, it is first determined whether or not the low-pass filter reference evaluation value is smaller than the next frame evaluation value (S41). Here, the reference evaluation value stored in the low-pass filter evaluation value memory 20 is compared with the next frame evaluation value (the value obtained most recently among the plurality of stored focus evaluation values) for determination. . When the focus lens is driven in the forward direction, the reference evaluation value is smaller than the evaluation value of the next frame.

  If the result of determination in step S41 is that the low-pass filter reference evaluation value is smaller than the next frame evaluation value, then a threshold coefficient A1 is calculated (S43). Here, the threshold coefficient A1 is calculated according to the increase / decrease rate of the evaluation value of the low-pass filter. The increase / decrease rate of the evaluation value is calculated by the low-frequency evaluation value increase / decrease detection unit 22. Once the threshold coefficient A1 is calculated, the threshold H is then calculated from the reference threshold × A1 (S45). As the reference threshold value, the reference decrease threshold value is stored in the reference threshold value table 31 in accordance with the maximum value of the evaluation value. Therefore, the read reference decrease threshold value is multiplied by the threshold coefficient A1 calculated in step S43.

  On the other hand, if the result of determination in step S41 is that the low-pass filter reference evaluation value is greater than the next frame evaluation value, next, a threshold coefficient A2 is calculated (S47). Here, as in step S43, the threshold coefficient A2 is calculated according to the increase / decrease rate of the evaluation value of the low-pass filter. Once the threshold coefficient A2 is calculated, the threshold H is then calculated from the reference threshold × A2 (S49). The reference decrease threshold stored in the reference threshold table 31 is multiplied by the threshold coefficient A2 calculated in step S47.

  The threshold coefficient A1 is smaller than the threshold coefficient A2. When the low-pass filter reference evaluation value is smaller than the next frame evaluation value, the focus lens is driven smoothly toward the peak because it is driven in the forward direction. In this case, it is possible to determine a peak when it is slightly reduced from the peak. However, when the low-pass filter reference evaluation value is larger than the next frame evaluation value, it is affected by disturbance and is in an unstable driving state, and it is difficult to say that the driving is smoothly proceeding toward the peak. Therefore, in order to determine that the peak has been exceeded, the decrease threshold is set to a large value.

  If the threshold value H is calculated in step S45 or S49, it is next determined whether or not the high-pass filter reference evaluation value is smaller than the next frame evaluation value (S51). Here, the reference evaluation value stored in the high-pass filter evaluation value memory 21 is compared with the next frame evaluation value (a plurality of values obtained most recently among the stored focus evaluation values) for determination. . When the focus lens is driven in the forward direction, the reference evaluation value is smaller than the evaluation value of the next frame in the focus evaluation value based on the output from the high-pass filter unit as in the case of the low-pass filter. .

  If the result of determination in step S51 is that the high-pass filter reference evaluation value is smaller than the next frame evaluation value, then a threshold coefficient B1 is calculated (S53). Here, the threshold coefficient B1 is calculated according to the increase / decrease rate of the evaluation value of the high-pass filter. The increase / decrease rate of the evaluation value is calculated by the high frequency evaluation value increase / decrease detection unit 24. Once the threshold coefficient B1 has been calculated, the threshold I is then calculated from the threshold H × B1 (S55). Here, the threshold value H calculated in step S45 or S49 is multiplied by the threshold coefficient B1 calculated in step S53.

  On the other hand, if the result of determination in step S51 is that the high-pass filter reference evaluation value is greater than the next frame evaluation value, then a threshold coefficient B2 is calculated (S57). Here, as in step S53, the threshold coefficient B2 is calculated according to the increase / decrease rate of the evaluation value of the high-pass filter. Once the threshold coefficient B2 is calculated, the threshold I is then calculated from the threshold I × B2 (S59). Here, the threshold value H calculated in step S45 or S49 is multiplied by the threshold coefficient B2 calculated in step S57.

  Note that the threshold coefficient B1 is smaller than the threshold coefficient B2. In the case of the evaluation value of the high-pass filter, as in the case of the evaluation value of the low-pass filter, if the reference evaluation value of the high-pass filter is larger than the evaluation value of the next frame, the drive state is unstable and the peak is steadily It is hard to say that it is driven towards. Therefore, in order to determine that the peak has been exceeded, the decrease threshold is set to a large value.

  Once the threshold value I is calculated in step S55 or S59, it is next determined whether or not the low-pass filter evaluation value is continuously increased (S61). Here, the continuity of the focus evaluation value, for example, the continuity up to the maximum value during driving in the forward direction is confirmed for a predetermined period from the low-pass filter reference focus evaluation value. The continuity of the evaluation value is detected by the low-frequency evaluation value continuity detection unit 23 (see FIG. 2). In the example shown in FIG. 6, the focus evaluation value continuously increases from the focus evaluation value LVR1 at the lens position LA1 that is the reference evaluation value to the focus evaluation value LVR5 at the lens position LA5 that is the maximum value without decreasing the focus evaluation value. In such a case, it is determined that there is a continuous increase. On the other hand, in the example shown in FIG. 7, the focus evaluation value may decrease between the focus evaluation value LV2 at the lens position LB1 that is the reference evaluation value and the focus evaluation value LV5 at the lens position LB6 that is the maximum value. (Between LB1 and LB2, and between LB3 and LB4) In such a case, it is determined that there is no continuous increase.

  If the result of determination in step S61 is that the low-pass filter evaluation value has increased continuously, a threshold coefficient C1 is calculated (S63). Here, the threshold coefficient C1 is calculated according to the increase rate of the evaluation value of the low-pass filter. Once the threshold coefficient C1 is calculated, the threshold J is then calculated from the threshold I × C1 (S45). The threshold value I is calculated in step S55 or S59, and this threshold value I is multiplied by the threshold coefficient C1 calculated in step S63.

  On the other hand, if the result of determination in step S61 is that the low-pass filter evaluation value has not increased continuously, a threshold coefficient C2 is calculated (S67). In this case, the evaluation value does not continuously increase from the reference evaluation value to the maximum value. In this step, the threshold value C2 is calculated from the number of evaluation value reductions and the reduction rate of the low-pass filter. Once the threshold coefficient C2 is calculated, the threshold J is then calculated from the threshold I × C2 (S69). Here, the threshold value I calculated in step S55 or S59 is multiplied by the threshold coefficient C2 calculated in step S67.

  As in the case of the threshold coefficients A1, A2, B1, and B2, the threshold coefficient C1 is smaller than the threshold coefficient C2. When the low-pass filter evaluation value continuously increases, it can be said that the focus lens is being driven smoothly toward the peak. In this case, it is possible to determine a peak when it is slightly reduced from the peak. However, when the low-pass filter evaluation value does not continuously increase, it is in an unstable state, and it is difficult to say that the low-pass filter evaluation value is being smoothly driven toward the peak. Therefore, in order to determine that the peak has been exceeded, the decrease threshold is set to a large value.

  Once the threshold value J is calculated in step S65 or S69, it is next determined whether or not the high-pass filter evaluation value is a continuous increase (S71). Here, the continuity of the focus evaluation value for a predetermined period from the high-pass filter reference focus evaluation value, for example, the continuity up to the maximum value during driving in the forward direction is confirmed. The continuity of the evaluation value is detected by the high frequency evaluation value continuity detection unit 25 (see FIG. 2). In the example shown in FIG. 6, the focus evaluation value continuously increases from the focus evaluation value HVR1 at the lens position LA1 that is the reference evaluation value to the focus evaluation value HVR5 at the lens position LA5 that is the maximum value without decreasing the focus evaluation value. In such a case, it is determined that there is a continuous increase. On the other hand, in the example shown in FIG. 7, the focus evaluation value may decrease between the focus evaluation value HV1 at the lens position LB1 that is the reference evaluation value and the focus evaluation value HV6 at the lens position LB6 that is the maximum value. (Between LB3 and LB4), in such a case, it is determined that there is no continuous increase.

  If the result of determination in step S71 is that the high-pass filter evaluation value has increased continuously, a threshold coefficient D1 is calculated (S73). Here, the threshold coefficient D1 is calculated according to the increase rate of the evaluation value of the high-pass filter. Once the threshold coefficient D1 is calculated, the threshold K is then calculated from the threshold J × D1 (S75). The threshold value J is calculated in step S65 or S69, and this threshold value J is multiplied by the threshold coefficient D1 calculated in step S73.

  On the other hand, if the result of determination in step S71 is that the high-pass filter evaluation value has not increased continuously, a threshold coefficient D2 is calculated (S77). In this case, the evaluation value does not continuously increase from the reference evaluation value to the maximum value. In this step, the threshold value D2 is calculated from the number of evaluation value reductions and the reduction rate of the high-pass filter. Once the threshold coefficient D2 has been calculated, the threshold value K is then calculated from the threshold value J × D2 (S79). Here, the threshold value J calculated in step S65 or S69 is multiplied by the threshold coefficient D2 calculated in step S77.

  Note that the threshold coefficient D1 is smaller than the threshold coefficient D2. In the case of the high-pass filter evaluation value, as in the case of the low-pass filter evaluation value, when the high-pass filter evaluation value does not continuously increase, the drive state is unstable and the drive is smoothly performed toward the peak. It ’s hard to say. Therefore, in order to determine that the peak has been exceeded, the decrease threshold D2 is set to a large value.

  When the threshold value K is calculated in step S75 or S79, the threshold value K is determined (S81). Here, the threshold value K calculated in step S75 or S79 is determined. When the threshold value K is determined, the process returns to the original flow. When returning to the original flow, as described above, peak detection is performed in step S31 (FIG. 3). At this time, peak detection is performed using the threshold value K determined in step S81. The determined threshold value K is used as the decrease threshold value ThA in FIG. 6 and the decrease threshold value ThB in FIG.

  Next, the operation in the flow of FIGS. 4 and 5 will be described in the case where disturbance due to camera shake or the like as shown in FIG. 7 occurs. In step S17 of FIG. 3, the reference evaluation values of the low-pass and high-pass filters are acquired. Further, in step S25 of FIG. 3, the evaluation of the low-pass and high-pass filters is performed each time an image signal for one frame is output. The value is obtained.

  If the flow for calculating the reduction threshold shown in FIG. 4 is entered, it is first determined whether or not the low-pass filter reference evaluation value is smaller than the next frame evaluation value (S41). In the example shown in FIG. 7, since the low-pass filter reference evaluation value LV2 is larger than the low-pass filter evaluation value LV1 at the lens position LB2, the process proceeds to Steps S47 and S49 as a result of determination, and threshold H = reference threshold × A2 is set. calculate.

  Subsequently, in the determination in step S51, since the high-pass filter reference evaluation value HV1 is smaller than the high-pass filter evaluation value HV2 at the lens position LB2, the process proceeds to steps S53 and S55, and threshold I = threshold H × B1 is calculated. .

  Subsequently, in the determination in step S61, the low-pass filter evaluation value decreases from the reference evaluation value to the next evaluation value (evaluation value LV1 at the lens position LB2), and then reaches the maximum value. Since there is a part where the value decreases (evaluation value LV2 at the lens position LB4), the process proceeds to steps S67 and S69, and threshold value J = threshold value I × C2 is calculated.

  Subsequently, in the determination in step S71, the high-pass filter evaluation value increases from the reference evaluation value to the next evaluation value (evaluation value HV2 at the lens position LB2), but thereafter the evaluation value decreases to the maximum value. Since there is a spot (evaluation value HV3 at the lens position LB4), the process proceeds to steps S77 and S79, and threshold value K = threshold value J × D2 is calculated.

  In the example illustrated in FIG. 7, the final threshold value K (corresponding to the decrease threshold value ThB) is threshold value K = reference threshold value × A2 × B1 × C2 × D2. Although not described in detail, in the example shown in FIG. 6, since the evaluation value monotonously increases, threshold value K = reference threshold value × A1 × B1 × C1 × D1. As described above, since there is a relationship of A1 <A2, B1 <B2, C1 <C2, and D1 <D2, the threshold value K is greater in the example shown in FIG. 7 than in the example shown in FIG. Is also a large value. As a result, in the example shown in FIG. 6, it is determined that the peak is slightly lowered after passing the peak, whereas in the example shown in FIG. It is determined that the peak is reached when the voltage drops.

  As described above, even immediately after the hill-climbing AF is started, the shooting environment is determined from the change in the focus evaluation value, and the threshold corresponding to the shooting environment is set. For this reason, the amount of overrun after passing the peak can be optimized.

  As described above, in one embodiment of the present invention, a high-pass filter unit that passes a high-frequency component of an image signal, and a low-pass filter unit that passes a lower-frequency component of an image signal than a high-pass filter unit, A focus evaluation value calculation unit for calculating a focus evaluation value based on the output of the low-pass filter unit or the high-pass filter unit, and using two types of focus evaluation values of a low-frequency component and a high-frequency component of the image signal A threshold for performing peak determination is set. Since the change in focus evaluation value differs between the low-frequency component and the high-frequency component, it is possible to quickly and stably detect the in-focus point even at the start of the focus adjustment operation by determining by combining the two types of focus evaluation values. It becomes possible to perform highly accurate focus adjustment.

  In one embodiment of the present invention, the focus evaluation value calculated based on the image signal acquired at the position where the focus adjustment operation is started and the focus evaluation value calculated based on the image signal acquired thereafter. Increase / decrease detection unit (see low frequency evaluation value increase / decrease detection unit 22 and high frequency evaluation value increase / decrease detection unit 24 in FIG. 2), and increase / decrease direction and increase / decrease detected by the increase / decrease detection unit. Based on the rate, a threshold for determining the peak of the focus evaluation value is set, and the focus adjustment operation based on the peak of the focus evaluation value is performed (the threshold coefficient A calculation unit 26, the threshold coefficient B calculation unit in FIG. 2). 28, see S45, S49, S55, S59, etc. in FIG. 4). Since the threshold is set according to the increase / decrease direction and the increase / decrease rate of the focus evaluation value, it is possible to perform focus adjustment quickly and with high stability even at the start of the focus adjustment operation. Become.

  In one embodiment of the present invention, continuity detection is performed to detect continuity of a plurality of focus evaluation values calculated based on a plurality of image signals acquired in a predetermined period from a position at which a focus adjustment operation is started. 2 (see low-frequency evaluation value continuity detection unit 23 and high-frequency evaluation value continuity detection unit 25 in FIG. 2), and determines the peak of the focus evaluation value based on the continuity detected by the continuity detection unit And a focus adjustment operation based on the peak of the focus evaluation value is performed (threshold coefficient C calculating unit 27, threshold coefficient D calculating unit 29 in FIG. 2, S65, S69, S75, S79 in FIG. 5). Etc.). Since the threshold value is set according to the continuity of the focus evaluation values, it is possible to perform focus adjustment quickly and with high stability even when the focus adjustment operation is started.

  In one embodiment of the present invention, the threshold value is set based on both the increase / decrease rate of the focus evaluation value and the continuity. However, although the speed and stability are somewhat reduced, the threshold may be set based on one of them. In addition, although the focus evaluation value is calculated using two types of filters, a low-pass filter and a high-pass filter, one or three or more types of filters may be provided to calculate the threshold value.

  In one embodiment of the present invention, in determining the continuity of the focus evaluation value, the continuity from when the reference evaluation value is acquired until the focus evaluation value becomes maximum is determined. However, the present invention is not limited to this, and other criteria may be applied, for example, until the focus evaluation value is acquired a predetermined number of times.

  In one embodiment of the present invention, the focus evaluation value immediately after the start of the AF operation is used as the reference evaluation value (S13 in FIG. 3). However, the timing is not limited to immediately after the start, and may be a reference timing.

  In the present embodiment, the digital camera is used as the photographing device. However, the camera may be a digital single-lens reflex camera or a compact digital camera, and may be used for moving images such as video cameras and movie cameras. It may be a camera, or may be a camera built into a mobile phone, a smart phone, a personal digital assistant (PDA), a game machine, or the like. In any case, the present invention can be applied to any device that performs focus adjustment using the focus evaluation value.

  In addition, regarding the operation flow in the claims, the specification, and the drawings, even if it is described using words expressing the order such as “first”, “next”, etc. It does not mean that it is essential to implement in this order.

  The present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, you may delete some components of all the components shown by embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

DESCRIPTION OF SYMBOLS 1 ... Optical system, 2 ... Image sensor, 3 ... AGC / CDS, 4 ... A / D, 5 ... Signal processing part, 6 ... Lens control part, 7 ... Imaging control unit, 8 ... AE processing unit, 9 ... Image processing unit, 10 ... CPU, 10a ... Focus determination unit, 11 ... Focus evaluation value generation unit, 12 ... Bus

Claims (5)

An image pickup unit that generates an image signal by forming an optical image by an optical system on an image pickup device, obtains an image signal by repeatedly taking an image while moving the optical system, and focuses on the basis of the image signal. In the focus adjustment device that performs the adjustment operation,
A high-pass filter that passes high-frequency components of the image signal;
A low-pass filter that passes a lower frequency component of the image signal than the high-pass filter, and
A focus evaluation value calculation unit that calculates a focus evaluation value based on the output of the low-pass filter unit or the high-pass filter unit;
Increase / decrease for detecting the increase / decrease direction and the increase / decrease rate between the focus evaluation value calculated based on the image signal acquired at the position where the focus adjustment operation is started and the focus evaluation value calculated based on the image signal acquired thereafter. A detection unit;
A continuity detection unit that detects continuity of a plurality of focus evaluation values calculated based on a plurality of image signals acquired in a predetermined period from a position at which a focus adjustment operation is started;
Based on the increase / decrease direction and the increase / decrease rate detected by the increase / decrease detection unit and the continuity detected by the continuity detection unit, a threshold for determining the peak of the focus evaluation value is set and the peak of the focus evaluation value is set. A control unit for performing a focusing operation based on
A focus adjusting apparatus comprising: The increase / decrease detection unit detects an increase / decrease direction and an increase / decrease rate of the focus evaluation value based on the output of the low pass filter unit,
The continuity detection unit detects continuity of the focus evaluation value based on the output of the low-pass filter unit,
The control unit, based on the increase / decrease direction and the increase / decrease rate detected by the increase / decrease detection unit, and the continuity detected by the continuity detection unit, the peak of the focus evaluation value based on the output of the high-pass filter unit A threshold value for detecting the focus evaluation value and performing a focus adjustment operation based on the peak of the focus evaluation value.
The focus adjusting apparatus according to claim 1, wherein: The increase / decrease detection unit detects an increase / decrease direction and an increase / decrease rate of the focus evaluation value based on the output of the high pass filter unit,
The continuity detection unit detects continuity of the focus evaluation value based on the output of the high-pass filter unit,
The control unit, based on the increase / decrease direction and the increase / decrease rate detected by the increase / decrease detection unit, and the continuity detected by the continuity detection unit, the peak of the focus evaluation value based on the output of the high-pass filter unit A threshold value for determining the focus evaluation value, and performing a focus adjustment operation based on the peak of the focus evaluation value.
The focus adjusting apparatus according to claim 2, wherein: The control unit sets a reduction amount from the maximum value of the focus evaluation value, and determines that the maximum value is a peak when the focus evaluation value decreases from the maximum value beyond the reduction amount.
4. The focus adjusting apparatus according to claim 1, wherein The focus evaluation value calculated at the start of the focus adjustment operation by calculating a focus evaluation value from the high frequency component of the image signal acquired by the image sensor at the start of the focus adjustment operation and the lower frequency component than the high frequency component. Store the value as the reference evaluation value,
After the start of the focus adjustment operation, each time an image signal is acquired by the image sensor, a focus evaluation value is calculated from the high frequency component and the low frequency component,
In accordance with the magnitude relationship and continuity between the reference evaluation value and the focus evaluation value, a threshold for peak detection of the focus evaluation value is calculated,
The peak of the focus evaluation value is determined using the threshold for peak detection.
A focus detection method characterized by the above.