JP5539490B2 - Image shake correction apparatus, optical apparatus including the same, imaging apparatus, and image shake correction apparatus control method - Google Patents

Description

The present invention relates to an image blur correction apparatus having an image blur correction function, an optical apparatus including the image blur correction apparatus , an imaging apparatus , and a method for controlling the image blur correction apparatus .

  In an imaging apparatus typified by a still camera and a video camera, there are an optical image shake correction system, an image sensor type image shake correction system, and the like as a system for correcting shake such as camera shake given to the apparatus from the outside.

  In these methods, a signal from a shake detection sensor that detects the degree of shake is subjected to digital signal processing through an A / D converter, a shake correction amount is calculated and D / A converted, and then image shake correction is performed. Correction means, specifically a shift lens or an image sensor.

  Here, an angular velocity sensor is often used to detect the degree of vibration. This angular velocity sensor vibrates a vibration material such as a piezoelectric element at a constant frequency, and converts a force generated by a Coriolis force generated by a rotational motion component into a voltage. To obtain an angular velocity signal.

  A microcomputer is used as an apparatus for performing A / D conversion, digital signal processing, and D / A conversion, and includes a filter that cuts off a plurality of predetermined frequencies and an integration filter. These filters include a recursive digital filter, and the recursive filter includes a feedforward unit and a feedback unit.

FIG. 9A is an overall block diagram of the recursive primary digital filter. The recursive digital filter is composed of a feedforward unit and a feedback unit. In the recursive digital filter, the intermediate value is a calculation result of the feedback unit, and here, Z [n] shown in the figure is an intermediate value in the current sampling. Here, Z ⁻¹ is a delay element, and the value after passing through the delay element represents the previous sampling. The order of the digital filter is determined by this delay element.

  FIG. 9B is a diagram in which the feedback unit of the recursive digital filter is extracted. When the number of times of sampling is represented by n, the intermediate value Z [n] in the current sampling is calculated from the input value X [n] in the current sampling and the intermediate value Z [n−1] in the previous sampling. FIG. 9C is a diagram in which the feedforward part of the recursive digital filter is extracted, and the output value Y in the current sampling from the intermediate value Z [n] in the current sampling and the intermediate value Z [n−1] in the previous sampling. [N] is calculated. FIG. 9D shows an arithmetic expression when the gain sections of the feedforward section and the feedback section are set as constants a, b, and c.

FIG. 9E is an overall block diagram of a non-recursive primary digital filter. It can be said that the non-recursive digital filter is composed of only the feed-forward unit compared to the recursive digital filter. When the input value in the current sampling is X [n], the input value in the previous sampling is X [n−1], and this value is an intermediate value in the non-recursive digital filter. That is, in the non-recursive digital filter, the value after passing through the delay element Z- ¹ corresponds to the intermediate value. FIG. 9F shows an arithmetic expression when the gain part of the feedforward part is set to constants a and b.

  In order to configure a filter having desired characteristics, the values and signs of the constants a, b, and c are appropriately set. By setting this constant, a digital high-pass filter or a digital low-pass filter can be obtained.

In addition, the second-order and higher-order digital filters are realized by increasing the delay element Z- ¹ , but the number of intermediate values increases according to the order.

  The optical image blur correction method corrects image blur on the image sensor by moving a shift lens, which is a correction means for image blur correction, in a plane orthogonal to the optical axis by the image blur correction amount (on the image sensor). The image blur is removed from the image formed on the image. The image sensor type image blur correction method corrects image blur on the image sensor by moving the image sensor, which is a correction unit for image blur correction, by a shake correction amount in a plane orthogonal to the optical axis. It is a method. Since any method can apply the present invention described later, the configuration of the optical image blur correction method will be described below as an example.

  In the imaging apparatus having the image blur correction function of the above method, when the shift lens driving unit is instructed to move by the shake correction amount and the shift lens to be controlled reaches the drive target position, the actual position of the shift lens To get. Then, feedback control is performed so that the deviation between the drive target position and the actual position is zero.

  The driving range of the shift lens in the optical image blur correction method is determined by a mechanical limit and a limit due to optical performance. When the image blur correction function is effective, it tries to correct blur due to panning operation other than camera shake. For this reason, when the shift lens is near the drive limit, correct image blur correction may not be performed. Therefore, in Patent Document 1, if the shift lens drive amount is large, the exposure is within the drive range before exposure. A technique for returning the shift lens to a predetermined position in is disclosed.

JP 7-199263 A

  According to the above-mentioned Patent Document 1, when the shift amount of the shift lens for image blur correction is large, the shift lens is returned to a predetermined position before exposure, so that the drive is performed even when the shift lens is near the drive limit. A range can be secured.

  However, when a digital low-pass filter is used, even if the shake correction amount, which is the output of the digital filter, is changed by a predetermined amount, if the intermediate value is large, the output increases immediately after the predetermined amount is changed, resulting in a short The driving amount will increase with time. Further, when the intermediate value is large, the intermediate value of the digital low-pass filter acts to reduce the final digital filter output, and thus image blur correction may not be performed correctly.

(Object of invention)
An object of the present invention, by changing the intermediate value to a predetermined value the delay elements of the digital filter having a delay element to the exposure immediately before is held, to avoid that the driving amount of the correction means is larger (the panning operation An object of the present invention is to provide an imaging apparatus capable of performing appropriate image blur correction (excluding the influence of such a large shake).

In order to achieve the above object, the present invention provides a shake detection means for detecting shake, a first digital filter having a delay element that allows a shake signal from the shake detection means to pass through a predetermined frequency band, A second digital filter having a delay element that allows a shake signal from the shake detection means to pass a predetermined frequency band different from a pass band of the first digital filter; and the second digital filter until the start of exposure. A shake correction amount is calculated based on the output from the filter, and after exposure, a calculation means for calculating the shake correction amount based on the output from the first digital filter, and the correction means is driven using the shake correction amount. and, cut and control means for correcting the image blur, filters for calculating the shake correction amount from said second digital filter to the first digital filter The intermediate value changing means for changing the intermediate value held by the delay element of the first digital filter to a predetermined value, and the second value before the change at the timing when the intermediate value is changed to the predetermined value. calculating a difference between the output from the output from the first digital filter after the change from the digital filter, that has a discontinuous point connection means for adding the difference to the output from said first digital filter The image blur correction device is characterized.

According to the present invention, by changing the intermediate value to a predetermined value the delay elements of the digital filter having a delay element to the exposure immediately before is held, to avoid that the driving amount of the correction means increases, proper image An imaging device capable of performing shake correction can be provided.

It is a block diagram which shows the imaging device which concerns on each Example of this invention. It is a block diagram which shows circuit structures, such as a shift lens drive control part concerning each Example of this invention. It is a block diagram which shows the structure in the image stabilization control part which concerns on Example 1 of this invention. It is a wave form diagram which shows the shift lens target value which concerns on Example 1 of this invention. It is a flowchart which shows the operation | movement for shake correction amount calculation which concerns on Example 1 of this invention. It is a block diagram which shows the structure in the image stabilization control part which concerns on Example 2 of this invention. It is a wave form diagram which shows the shift lens target value which concerns on Example 2 of this invention. It is a flowchart which shows the operation | movement for shake correction amount calculation which concerns on Example 2 of this invention. It is a block diagram which shows a general recursive primary digital filter.

  The best mode for carrying out the present invention is as shown in Examples 1 and 2 below.

Example 1
FIG. 1 is a configuration diagram illustrating an image pickup apparatus having an image blur correction function according to the first embodiment of the present invention. In FIG. 1, reference numeral 101 denotes a zoom unit, which includes a zoom lens that performs zooming. Reference numeral 102 denotes a zoom drive control unit that controls the drive of the zoom unit 101. Reference numeral 103 denotes a shift lens which is an example of a correction unit for image blur correction capable of changing a position in a plane orthogonal to the optical axis 100. A shift lens drive control unit 104 controls the drive of the shift lens 103. During power saving, power supply to the shift lens drive control unit 104 is stopped by the control unit 119 described later.

  Reference numeral 105 denotes an aperture / shutter unit. Reference numeral 106 denotes an aperture / shutter drive control unit which controls driving of the aperture / shutter unit 105. A focus unit 107 includes a lens that performs focus adjustment. A focus drive control unit 108 controls the drive of the focus unit 107. Reference numeral 109 denotes an image pickup unit that uses an image pickup element such as a CCD, and converts an optical image that has passed through each lens group into an electrical signal. Reference numeral 110 denotes an imaging signal processing unit that converts an electrical signal output from the imaging unit 109 into a video signal. A video signal processing unit 111 processes the video signal output from the imaging signal processing unit 110 according to the application. Reference numeral 112 denotes a display unit that displays an image as necessary based on the video signal output from the video signal processing unit 111. A display control unit 113 controls the operation and display of the imaging unit 109 and the display unit 112.

  A shake detection unit 114 such as an angular velocity sensor detects the degree of shake given to the photographing apparatus. Reference numeral 115 denotes a power supply unit that supplies power to the entire system according to the application. An external input / output terminal unit 116 inputs / outputs communication signals and video signals to / from the outside. Reference numeral 117 denotes an operation unit for operating the system. Reference numeral 118 denotes a storage unit that stores various data such as video information. A control unit 119 controls the entire system.

  Next, the operation of the imaging apparatus having the above configuration will be described.

  The operation unit 117 includes a shutter release button configured such that the first switch (SW1) and the second switch (SW2) are sequentially turned on according to the amount of pressing. The first switch is turned on when the shutter release button is depressed approximately half, and the second switch is turned on when the shutter release button is depressed to the end. When the first switch is turned on, the control unit 119 drives the focus unit 107 via the focus drive control unit 108 to perform focus adjustment. At the same time, the diaphragm / shutter unit 105 is driven via the diaphragm / shutter drive controller 106 to set an appropriate exposure amount. When the second switch is further turned on, the control unit 119 causes the storage unit 118 to store image data obtained from the light image exposed to the imaging unit 109.

  At this time, if there is an instruction to enable the image blur correction function from the operation unit 117, the control unit 119 instructs the shift lens drive control unit 104 to perform an image blur correction operation. Then, the shift lens drive control unit 104 that has received the instruction drives the shift lens 103 until an instruction to disable the image blur correction function is made, that is, the shift lens 103 in a direction to cancel the shake in a plane orthogonal to the optical axis 100. Is moved to perform image blur correction.

  When the operation unit 117 has not been operated for a certain period of time, the control unit 119 shuts off the power to the display unit 112 and the shift lens drive control unit 104 in order to save power.

  Further, in this imaging apparatus, one of the still image shooting mode and the moving image shooting mode can be selected by the operation unit 117, and the operation condition of each drive control unit can be changed in each mode.

  When the zoom unit 101 is instructed by the operation unit 117, the control unit 119 drives the zoom unit 101 via the zoom drive control unit 102 to move the zoom unit 101 to the instructed zoom position. At the same time, based on the image information processed by the signal processing units 110 and 111 sent from the imaging unit 109, the focus unit 107 is driven via the focus drive control unit 108 to perform focus adjustment.

  FIG. 2 is a block diagram showing the internal configuration of the shift lens drive control unit 104 and the circuit configuration of the preceding stage.

  First, the configuration on the front stage side of the shift lens drive control unit 104 will be described. 114a is a vertical shake detection unit that detects vertical (pitch direction) shakes of the imaging device in the normal posture, and 114b is a horizontal shake detection unit that detects horizontal (yaw direction) shakes of the imaging device in the normal posture. is there. Reference numerals 209a and 209b denote anti-vibration control units included in the control unit 119, which calculate shake correction amounts in the pitch direction and the yaw direction, determine the drive target position of the shift lens 103, and output it to the shift lens drive control unit 104. To do.

  Next, the configuration within the shift lens drive control unit 104 will be described. Reference numerals 301a and 301b denote PID units as feedback control units in the pitch direction and the yaw direction, and a control amount is obtained from a deviation between the above drive target position and an actual position signal to be described later indicating the current position of the shift lens 103. A command signal is output. Reference numerals 302a and 302b denote pitch direction and yaw direction drive units, which drive the shift lens 103 based on position command signals sent from the PID units 301a and 301b. Reference numerals 303a and 303b denote position detectors in the pitch direction and the yaw direction, which detect current positions of the shift lens 103 in the respective directions and output actual position signals to the PID units 301a and 301b.

  Next, position control of the shift lens 103 by the shift lens drive control unit 104 will be described.

  In the position control of the shift lens 103, the shift lens 103 is driven in the pitch direction and the yaw direction based on signals indicating the shake of the imaging device from the shake detection unit 114a and the shake detection unit 114b. A magnet is attached to the shift lens 103, and the position detection units 303a and 303b detect the magnetic field of the magnet, and the actual position signal of the shift lens 103 is output to the PID units 301a and 301b. The PID units 301a and 301b perform feedback control such that the input actual position signal converges to the drive target position sent from the image stabilization control units 209 and 209b. At this time, the PID units 301a and 301b perform PID control that selectively combines proportional (P) control, integral (I) control, and differential (D) control.

  As described above, even if a shake such as a hand shake occurs in the image pickup apparatus, the image shake can be corrected appropriately.

  FIG. 3 is a block diagram showing details of the image stabilization control unit 209 (209a, 209b) shown in FIG.

  In FIG. 3, reference numeral 201 denotes an A / D converter that converts analog data, which is a shake signal detected by the shake detector 114 (114a, 114b), into digital data. Reference numeral 202 denotes a digital high-pass filter (digital HPF), which is a filter that passes a predetermined high frequency band. As an example, the constants a and b in the configuration of the primary filter shown in FIG. 9 are positive signs, and the constant c is a negative sign filter. Reference numeral 203 denotes a digital low-pass filter (digital LPF), which is a filter that passes a predetermined low frequency band. As an example, the constants a and b in the configuration of the primary filter shown in FIG. 9 are positive signs, and the constant c is a positive sign filter. The data detected by the shake detection unit 114 is an angular velocity, and when the shift lens drive control unit 104 controls the shift lens 103 as an angle, the digital low-pass filter 203 functions as an integrator.

  A shake correction amount calculation unit 204 performs sign inversion on the shake amount as an output result of the digital low-pass filter 203, and performs zoom position information from the zoom drive control unit 102 and the focus drive control unit 108 in FIG. The amount of shake correction reflecting the focus information is calculated. Reference numeral 205 denotes a digital filter intermediate value changing unit that presets the intermediate value held by the digital low-pass filter 203 in accordance with the exposure start signal (second switch ON signal) input from the operation unit 117. Change to a predetermined value. A discontinuous point connection unit 206 continuously connects discontinuous points of the output result of the digital low-pass filter 203 generated by the digital filter intermediate value changing unit 205.

  The calculation result (drive target position) of the shake correction amount calculation unit 204 is output to the shift lens drive control unit 104, and the drive of the shift lens 103 is started.

  4A and 4B show the X-axis as the time axis, the Y-axis as the target value of the shift lens 103, and the intersection of the X-axis and Y-axis as the time zero and shift lens when changing the digital filter intermediate value. FIG. 10 is a waveform diagram with a center point (drive center) of a drive range 103; The center point of the driving range is usually the optical axis 100.

  The shift lens target value is a value obtained by inverting the sign of the output of the digital low-pass filter 203 and multiplying the zoom and focus information.

  The dotted waveform in FIG. 4A represents a target value calculated when the intermediate value of the digital low-pass filter 203 is not changed even after the exposure is started. At this time, since the influence of the intermediate value of the previous sampling on the output value of the current sampling is accumulated, the waveform rises to the right. As a result, the drive limit of the shift lens 103 is approached (away from the center point of the drive range).

  On the other hand, the solid line waveform after the start of exposure in FIG. 4A is a case where the intermediate value of the digital low-pass filter 203 is changed at the start of exposure. At this time, since the influence of the intermediate value of the previous sampling on the output value of the current sampling is once cleared, the influence of the accumulation is also eliminated, and a flat waveform is shown instead of a rising waveform. In this way, when the intermediate value is changed to a predetermined value, unlike the case where the offset at the start of exposure is subtracted as a constant at the output stage of the digital low-pass filter 203, it is possible to eliminate the influence of the previous accumulation. is there.

  In the case of control as shown in FIG. 4A, the intermediate value of the digital low-pass filter 203 is changed to a predetermined value at the start of exposure, and the output value is directly used as the shift lens target value. Therefore, it is convenient in that the shift lens 103 is driven near the center point, but discontinuous points occur in the waveform at the time of change. Here, the discontinuous point is a point generated by a difference, that is, an offset between the immediately preceding shift lens target value and the shift lens target value immediately after changing the intermediate value.

  In the control of FIG. 4B, the discontinuous points generated in FIG. 4A are continuously connected, and therefore, the offset is raised. Specifically, it is realized by temporarily calculating and storing an offset when changing the intermediate value to a predetermined value, and adding this offset to the output of the digital low-pass filter 203. The operation of raising the offset is continuously performed during exposure. Thereafter, the addition of the offset ends at the timing of the end of exposure. By controlling in this way, it is possible to prevent the occurrence of discontinuous points, and thus it is possible to prevent the shooting angle of view from being changed before and after the start of exposure.

  Here, as an example of specific numerical values for the constants a, b, and c of the digital low-pass filter 203, assuming that the sampling period is 10 KHz, a = 0.999993, b = 0. 000003, c = 0.000003 is preferred.

  Next, when the shake correction amount is calculated based on the shake signal from the shake detection unit 114, the digital filter intermediate value change unit 205 and the discontinuous point connection unit 206 act in accordance with the exposure start signal from the operation unit 117. The operation when doing this will be described with reference to the flowchart of FIG.

  In step S701, sampling for calculating a shake correction amount is started. First, in step S702, the A / D conversion unit 201 converts analog data, which is a shake signal detected by the shake detection unit 114, into digital data. Get the value that was recorded. In the next step S703, digital high-pass filter calculation is performed. This calculation is performed by the digital high-pass filter 202, and the constant c in the configuration of the primary filter shown in FIG. 9 is a negative sign filter. The digital high-pass filter 202 is a filter that passes a predetermined high frequency band, and serves as an example to block a low frequency band as a temperature drift component of the shake detection unit 114.

  In the next step S704, it is determined whether or not exposure is in progress. The moment when the second switch is turned on by operating the shutter release button arranged on the operation unit 117 is the start of exposure, and the time when the second switch is turned on is under exposure. If the result of this determination is that exposure is not currently in progress, processing proceeds from step S704 to step S713, and digital low-pass filter calculation is performed. This calculation is performed by the digital low-pass filter 203, and the constant c in the configuration of the primary filter shown in FIG. 9 is a positive sign filter. The digital low-pass filter 203 is a filter that passes a predetermined low-frequency band, and serves as an integrator as an example.

  On the other hand, if it is determined in step S704 that the exposure is currently being performed, the process proceeds to step S705, where it is determined whether or not it is the first sampling after the start of exposure. If it is the first sampling, the process advances from step S704 to step S706 to hold the digital low-pass filter output in the previous sampling. This output is LPFOOut (n-1). In the next step S707, the digital low pass filter calculation intermediate value is changed to a predetermined value. The intermediate value is a variable generated in the feedback unit of FIG. 9B and appears in common in the simultaneous equations of the feedback unit and the feedforward unit of FIG. In the first embodiment, changing the intermediate value to a predetermined value means setting the intermediate value to zero. Note that the predetermined value other than zero may be a value obtained by subtracting a value corresponding to a predetermined offset amount from the current value. As shown in FIG. 4, the offset amount changes moment by moment at the start of exposure, but at least the margin (constant) that can be expected to prevent collision with the mechanical drive end is an intermediate value. By subtracting from, a similar effect can be obtained.

  In the next step S708, a digital low-pass filter operation is performed reflecting that the intermediate value has been changed to a predetermined value. This output is defined as LPFOOut (n). Here, when the intermediate value is larger than the predetermined value in the stage before step S707, the shake correction amount also increases, and as a result, the drive amount of the shift lens 103 also increases. In addition, when the intermediate value is large, the feedforward unit of the digital low-pass filter acts to reduce the output, and thus correct image blur correction may not be performed. However, in the first embodiment, since the intermediate value is changed to the predetermined value (zero) as described above, the above problem is improved.

  In the next step S709, the difference between LPFOOut (n) and LPFOOut (n−1) that are discontinuous points is calculated. Here, this difference is called an offset (see FIG. 4A). In the next step S710, the offset is subtracted from the digital low-pass filter output. Thereby, as shown in FIG.4 (b), the discontinuous point which arose after the intermediate value change can be connected continuously.

  If it is determined in step S705 that it is not the first sampling, the process proceeds to step S712 to perform digital low-pass filter calculation. Thereafter, using the offset calculated in advance, the offset is subtracted from the digital low-pass filter output in the same manner as described above in the next step S710.

  As described above, the digital low-pass filter calculation during exposure is performed.

  When the exposure is completed, the process proceeds to step S713 as described above, the digital low-pass filter operation is performed, and the offset is not subtracted.

  In the next step S711, the shake correction amount is calculated using the calculated output of the digital low-pass filter. In step S714, the sampling for calculating the shake correction amount is terminated.

  According to the first embodiment, the intermediate value of the digital low-pass filter 203, which is a digital low-pass filter means, is changed to a predetermined value immediately before the start of exposure. Therefore, it is possible to avoid an increase in the driving amount of the correction unit, in other words, it is possible to eliminate the influence of a large shake such as a panning operation and perform an appropriate image shake correction.

  Further, at the timing when the intermediate value is changed to a predetermined value (zero), the discontinuous points of the output of the digital low-pass filter 203 generated after the change are continuously reconnected. Therefore, discontinuous points generated after changing the intermediate value can be continuously connected.

(Example 2)
Next, an image pickup apparatus according to Embodiment 2 of the present invention will be described. Note that the configuration of the imaging apparatus is the same as the configuration shown in FIGS. 1 and 2, and a description thereof will be omitted.

  FIG. 6 is a block diagram showing details of the image stabilization control unit 209 (209a, 209b) according to the second embodiment of the present invention. The difference from FIG. 3 is that when the digital high-pass filter and the digital low-pass filter are in one series, there are two series in the second embodiment of the present invention. In addition, the same part as FIG. 3 attaches | subjects the same code | symbol, and abbreviate | omits the description.

  In FIG. 6, reference numeral 202 denotes a first series digital high-pass filter (digital HPF), and reference numeral 207 denotes a second series digital high-pass filter (digital HPF). Reference numeral 203 denotes a first series digital low-pass filter (digital LPF), and 208 denotes a second series digital low-pass filter (digital LPF). Each series can characterize which frequency band is the target characteristic by changing the cutoff frequency of each filter.

  Here, in accordance with the exposure start signal input from the operation unit 117, switching from the series 2 (second series) targeting the high frequency band to the series 1 (first series) targeting the low frequency band is performed. The case where it went is demonstrated. Series 1 performs “performance-oriented” filter processing with a high vibration suppression rate in the whole region of body shake + hand shake (for example, 1 to 15 Hz), and has a high suppression effect on the low frequency side such as body shake. Yes, used during exposure. The problem of this series 1 is that the shift lens 103 goes to the mechanical drive end. On the other hand, the series 2 performs a “view-oriented” filter process having the highest vibration suppression rate in the vicinity of a specific frequency (for example, 5 Hz), and is used except during exposure.

  In the digital filter intermediate value changing unit 205, the intermediate value of the first series digital low pass filter 203 is changed to a preset predetermined value in accordance with the exposure start signal input from the operation unit 117. The discontinuous point connection unit 206 continuously connects discontinuous points between the series 2 before switching and the series 1 after changing the intermediate value. There is no need for two series, and the same contents can be applied to three or more series. Note that the above-described series switching is performed by the discontinuous point connection unit 206 with respect to the shake correction amount calculation unit 204. Specifically, the shake correction amount calculation unit 204 switches whether the input for calculating the shake correction amount is the series 1 or the series 2 according to an instruction from the discontinuous point connection unit 206. As will be described later with reference to FIGS. 7 (a) and 7 (b), there are advantages and disadvantages when the discontinuous points are connected and when the discontinuous points are not connected, but the discontinuous point connecting unit 206 has the discontinuity points. Even when it is not connected, it functions as a sequence switching unit.

  FIGS. 7A and 7B are waveform diagrams similar to FIGS. 4A and 4B according to the second embodiment of the present invention. That is, when changing the digital low-pass filter intermediate value, the X axis is the time axis, the Y axis is the shift lens target value, the intersection of the X axis and the Y axis is time zero, and the shift lens target value is the center point of the drive range. It is a waveform diagram.

  The series 1 and series 2 waveforms represented by the solid lines in FIG. 7A represent the output waveforms of the above-described digital low-pass filters 208 and 203, respectively. In the control shown in FIG. 7A, at the start of exposure, the input for calculating the shake correction amount from the series 2 to the series 1 is switched, and at the same time, the intermediate value of the series 1 is set to zero. When the exposure is completed, the sequence is switched back (returned) to series 2. Accordingly, the shift lens target value during this period has a waveform represented by a broken line.

  In the solid line waveforms of series 1 and series 2, since the influence of the intermediate value of the previous sampling on the output value of the current sampling is accumulated, the waveform rises to the right. As a result, the drive limit of the shift lens 103 is approached (away from the center point of the drive range).

  Since the influence of the intermediate value of the previous sampling on the output value of the current sampling is once cleared, the broken line waveform eliminates the influence of accumulation, and shows a flat waveform instead of a rising waveform. In this way, when the intermediate value is changed to a predetermined value, unlike the case where the offset at the start of exposure is subtracted as a constant at the output stage of the digital low-pass filter 203, it is possible to eliminate the influence of the previous accumulation. is there.

  In the case of control as shown in FIG. 7A, switching from the series 2 to the series 1 at the start of exposure, the intermediate value of the digital low-pass filter 203 of the series 1 is changed to a predetermined value, and the output value is directly used as the shift lens target. Value. Therefore, it is convenient in that the shift lens 103 is driven near the center point, but discontinuous points occur in the waveform at the time of change. Here, the discontinuous point is a point generated when the shift lens target value is different when switching from the series 2 to the series 1.

  In the control of FIG. 7B, the discontinuous points generated in FIG. 7A are continuously connected, so that the offset generated at the discontinuous points is increased. Specifically, when switching from series 2 to series 1, the offset generated by the output value obtained by changing the intermediate value of the digital low-pass filter 203 of series 1 to a predetermined value and the output value of series 2 at the time of switching (FIG. 7 ( a) Offset) is once calculated and stored. The offset is added to the output of the digital low-pass filter 203. The operation of raising the offset is continuously performed during exposure. Thereafter, the addition of the offset is terminated at the timing of the end of exposure, and the sequence is switched back to Series 2 again.

  In FIG. 7B, the broken line waveform represents the output waveform when the intermediate value is set to zero in the series 1, similar to the broken line in FIG. The solid line waveform represents a state in which the waveform is increased by an offset amount and connected to the series 2 during the exposure period. Therefore, after the exposure period is over, since switching to the series 2 occurs, a discontinuous point is generated here.

  By controlling as shown in FIG. 7B, it is possible to prevent a discontinuous point from occurring at the start of exposure, and thus it is possible to prevent the shooting angle of view from being changed before and after the start of exposure.

  Here, numerical values suitable as constants a, b, and c of the first-order recursive digital low-pass filter are given as examples. Assuming that the sampling period of the digital low-pass filter 203 is 10 KHz, when the cutoff frequency is 0.1 Hz, a = 0.999993, b = 0.00030003, and c = 0.00030003. Also, assuming that the sampling period of the digital low-pass filter 208 is 10 KHz, a = 0.999993, b = 0.000031, c = 0.000031 when the cutoff frequency is 0.1 Hz.

  Next, when the shake correction amount is calculated based on the shake signal from the shake detection unit 114, the digital filter intermediate value change unit 205 and the discontinuous point connection unit 206 match the exposure start signal from the operation unit 117. The operation when acting will be described with reference to the flowchart of FIG. Although the flowchart of FIG. 5 is for the one-sequence filter, the flowchart of FIG. 8 according to the second embodiment shows the operation for the two-sequence filter shown in FIG.

  In step S801, sampling for calculating the shake correction amount is started. First, in step S802, the A / D conversion unit 201 converts the analog data, which is the shake signal detected by the shake detection unit 114, into digital data. Get the value. Then, in the next step S803, it is determined whether or not exposure is in progress. As a result, if it is determined that the exposure is not currently being performed, the process proceeds to step S813 to perform the digital high-pass filter 2 calculation, and in the subsequent step S814, the digital low-pass filter 2 calculation, that is, the second series of calculations is performed.

  If it is determined in step S803 that exposure is currently in progress, the process proceeds to step S804, and the digital high-pass filter 1 is calculated. From here on, the first series of calculations is performed. In the next step S805, it is determined whether or not it is the first sampling after the start of exposure. If it is determined that the sampling is the first sampling, the process proceeds to step S806, and the digital low-pass filter 2 output in the previous sampling is held. This output is LPFOOut2 (n-1). In the subsequent step S807, the intermediate value of the digital low-pass filter 1 calculation is changed to a predetermined value (see FIG. 7). Here, changing the intermediate value to a predetermined value means changing the predetermined value to zero.

  In the next step S808, the digital low-pass filter 1 operation is performed reflecting that the intermediate value has been changed to a predetermined value. This output is LPFOOut1 (n). Then, in the next step S809, the difference between LPFOOut1 (n) and LPFOOut2 (n−1) that are discontinuous points is calculated. Here, this difference is called an offset (see FIG. 7). In the subsequent step S810, the offset is subtracted from the output of the digital low-pass filter 1. Thereby, as shown in FIG. 7, the discontinuous point produced after the intermediate value change can be continuously connected.

    If it is determined in step S805 that it is not the first sampling, the process proceeds to step S812 to perform digital low-pass filter 1 calculation. Thereafter, using the offset calculated in advance, the offset is subtracted from the digital low pass filter output in the next step S810.

  As described above, the digital low-pass filter calculation during exposure is performed.

  When the exposure is completed, the process proceeds to step S813 as described above, and the digital high-pass filter 2 calculation is performed. In the next step 814, the digital low-pass filter 2 calculation is performed and the offset is not subtracted.

  In the next step S811, the shake correction amount is calculated using the calculated output of the digital low-pass filter. In step S815, sampling for calculating the shake correction amount is terminated.

  According to the second embodiment, a plurality of series (series 1 and series 2) from the shake detection unit 114 to the calculation of the shake correction amount through the digital low-pass filter unit are provided. The series for calculating the shake correction amount is switched from the series 2 to the series 1 immediately before the exposure at the imaging unit (imaging device) 119. At this switching timing, the intermediate value of the digital low-pass filter means held as the calculation result of the feedback unit of the switching destination series (series 1) is changed to a predetermined value. Therefore, it is possible to avoid an increase in the driving amount of the correction unit, in other words, it is possible to eliminate the influence of a large shake such as a panning operation and perform an appropriate image shake correction.

  Further, at the timing when the intermediate value is changed to a predetermined value (zero), the discontinuous points of the output of the digital low-pass filter 203 generated after the change are continuously reconnected. Therefore, discontinuous points generated after changing the intermediate value can be continuously connected.

In the first embodiment and the second embodiment, the present invention has been described using the first-order recursive digital low-pass filter. However, the present invention is not limited to the first-order recursive digital low-pass filter, and can be applied to a non-recursive digital low-pass filter. As described with reference to FIG. 9, in the non-recursive digital low-pass filter, the output of the delay element Z- ¹ is an intermediate value, and this value is set to a predetermined value such as zero at the start of exposure. An effect is obtained.

Further, the order of the digital low-pass filter is not limited to the first order. As the delay element Z ⁻¹ increases, the order increases and the intermediate value also increases. However, if the plurality of intermediate values are uniformly changed to a predetermined value at the start of exposure, the same effect as in the above embodiment can be obtained.

(Correspondence between the present invention and the embodiment)
The shake detection unit 114 corresponds to a shake detection unit that detects shake in the present invention. The digital low-pass filter 203 corresponds to digital low-pass filter means having a delay element that allows a predetermined frequency to pass through the shake signal from the shake detection means of the present invention. Further, the shake correction amount calculation unit 204 corresponds to a calculation unit that calculates a shake correction amount based on an output from the digital low-pass filter unit of the present invention. Further, the shift lens 103 corresponds to the correcting means of the present invention. The shift lens drive control unit 104 corresponds to a control unit that drives the correction unit using the shake correction amount and corrects the image blur according to the present invention. The digital filter intermediate value changing unit 205 corresponds to the intermediate value changing means for changing the intermediate value of the digital low-pass filter means to a predetermined value immediately before exposure according to the present invention.

  The digital low-pass filter 203 corresponds to the first digital low-pass filter means having a delay element that allows a predetermined frequency to pass through the shake signal from the shake detection means of the present invention. The second digital low-pass filter having a delay element that allows the digital low-pass filter 208 to pass a predetermined frequency different from the pass band of the first digital low-pass filter with respect to the shake signal from the shake detection means of the present invention. It corresponds to the filter means.

DESCRIPTION OF SYMBOLS 103 Shift lens 104 Shift lens drive control part 109 Imaging part 114 Shake detection part 117 Operation part 201 A / D conversion part 202,207 Digital high pass filter 203,208 Digital low pass filter 204 Shake correction amount calculation part 205 Digital filter intermediate value change part 206 Discontinuous point connection unit 209a, 209b Anti-vibration control unit 301a, 301b PID unit 302a, 302b Drive unit 303a, 303b Position detection unit

Claims (7)

Shake detection means for detecting shake;
A first digital filter having a delay element that passes a predetermined frequency band with respect to a shake signal from the shake detection means;
A second digital filter having a delay element that allows a shake signal from the shake detection means to pass a predetermined frequency band different from a pass band of the first digital filter; and the second digital filter until the start of exposure. Calculating means for calculating a shake correction amount based on an output from the filter, and calculating a shake correction amount based on an output from the first digital filter after exposure;
Using the shake correction amount by driving the correcting means, and control means for correcting the image blur, when the filter for calculating the shake correction amount is switched to the first digital filter from the second digital filter, Intermediate value changing means for changing an intermediate value held by the delay element of the first digital filter to a predetermined value;
In the timing changing the intermediate value to the predetermined value, it calculates the difference between the output from the output from the first digital filter after the change from the second digital filter before the change, the said differential second An image blur correction apparatus comprising discontinuous point connecting means for adding to the output from one digital filter.   The image blur correction apparatus according to claim 1, wherein the first digital filter is a recursive type including a feedforward unit and a feedback unit, and a calculation result of the feedback unit is set to an intermediate value. 3. The difference according to claim 1, wherein the difference is continuously added to an output from the first digital filter generated after the intermediate value is changed to the predetermined value during exposure. 4. Image shake correction device. The predetermined value image blur correction device according to any one of claims 1 to 3, characterized in that it is zero. An optical apparatus, comprising an image shake correction apparatus according to any one of claims 1 to 4. Imaging apparatus characterized by comprising an image shake correction apparatus according to any one of claims 1 to 4. A control method for an image shake correction apparatus that drives the correction means based on the shake signal detected by the shake detection means to correct image shake,
A first filtering step having a delay step, and allowing a shake signal from the shake detection means to pass a predetermined frequency band;
A delay step, and a second filter step of passing a different frequency band than the pass band in the first filtering step with respect to the shake signal from the shake detection unit, the exposure to the start of the second A calculation step of calculating a shake correction amount based on an output from the filter step, and calculating a shake correction amount based on an output from the first filter step after exposure;
When the correction unit is driven using the shake correction amount to correct the image shake and the filter for calculating the shake correction amount is switched from the second filter step to the first filter step, the first step is performed. An intermediate value changing step for changing the intermediate value held in the delay step in one filter step to a predetermined value;
In the timing changing the intermediate value to the predetermined value, calculates the difference between the output from the first filter step after the output and the change from the second filter step before the change, the difference after the change And a discontinuous point connecting step for adding to the output from the first filter step that has occurred.

Images