JP2018006801A - Imaging apparatus, control method therefor, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To further shorten the time required for detecting peak timing by changing photometric timing or a photometric interval for detecting the peak timing of a flicker.SOLUTION: An imaging apparatus comprises: photometric means 108; detection means 112 for detecting peak timing of a flicker from multiple photometric results obtained by the photometric means 108; and control means 112 by which, in the case where the peak timing of the flicker can be predicted from a detection result of the detection means 112, the photometric means 108 is caused to perform photometry multiple times within a shorter time than the case where the peak timing cannot be predicted, in the vicinity of the predicted peak timing.SELECTED DRAWING: Figure 5

Description

  The present invention relates to an imaging apparatus such as a digital camera, and more particularly to a technique for improving exposure unevenness due to flicker (periodic change in the amount of light from a subject).

  In an imaging apparatus such as a digital camera, with a recent increase in ISO, a high-speed shutter can be released even under an artificial light source in which flicker occurs. While high-speed shutter photography has the advantage of being able to take pictures without blurring, such as indoor sports photography, under flickering light sources, image exposure and color unevenness may occur for each frame or within one frame due to the effect of flicker. May occur.

  To solve this problem, there is a method of reducing the influence of flicker by detecting flicker and performing exposure at the peak position of the flicker where the change in brightness is the least. For example, a flicker detection operation in which photometry is performed 12 times continuously with a period of 1.667 ms is performed by the photometric sensor, and when the flicker is detected, the shutter is driven according to the flicker peak timing. At this time, if continuous shooting is being performed, the flicker frequency is considered to be unchanged from the detection result before the start of continuous shooting, and a technique for reducing the number of photometry from 12 to 6 has been proposed (Patent Document 1). ). In this proposal, by reducing the number of metering times to 6 times, the reduction in continuous shooting speed is suppressed, and even if the flicker frequency is subtly changed by detecting the peak timing of each frame and changing the power supply frequency, the flicker environment It is said that a good image can be acquired below.

JP 2014-220964 A

  Although the above Patent Document 1 discloses that the number of photometry for detecting the flicker peak timing is less than the number of photometry for detecting the presence or absence of flicker, the photometry timing and the photometry interval are changed. Is not disclosed or suggested.

  Accordingly, an object of the present invention is to further reduce the time required for detecting the peak timing by changing the photometric timing and the photometric interval for detecting the flicker peak timing.

  In order to achieve the above object, an image pickup apparatus according to the present invention includes a photometric unit, a detection unit that detects a flicker peak timing from a plurality of photometric results obtained by the photometric unit, When the peak timing can be predicted, a control unit is provided that performs photometry a plurality of times by the photometry unit in the vicinity of the predicted peak timing and in a shorter time than when the peak timing cannot be predicted. And

  According to the present invention, the time required for detecting the peak timing can be further shortened by changing the photometric timing and photometric interval for detecting the flicker peak timing.

1 is a schematic diagram illustrating a system configuration example of a digital single-lens reflex camera that is an example of an embodiment of an imaging apparatus of the present invention. It is a flowchart explaining the operation of the camera when performing continuous shooting while calculating the timing of the light amount change of the light from the subject and the timing when the light amount change satisfies a predetermined condition. It is a figure which shows the photometry output at the time of a flicker detection operation | movement. It is a figure explaining an example of the method of calculating the peak timing of a flicker. It is a timing chart figure which shows the sequence of the camera operation | movement during continuous shooting. It is a figure explaining the case where peak timing has shifted from prediction.

  Hereinafter, an example of an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a schematic diagram illustrating a system configuration example of a digital single-lens reflex camera having a continuous shooting function which is an example of an embodiment of an imaging apparatus of the present invention.

  As shown in FIG. 1, the digital single-lens reflex camera of this embodiment has a replacement lens unit 200 detachably attached to the camera body 100.

  The camera body 100 has a camera CPU 101 that controls the entire camera, and a memory 102 such as a RAM or a ROM is connected to the camera CPU 101. The main mirror 105 and the sub mirror 111 guide the subject luminous flux that has entered the imaging optical path and passed through the lens unit 200 to the focus plate 106 during viewfinder observation, and retract from the imaging optical path to guide the subject luminous flux to the image sensor 103 during imaging. . The main mirror 105 is composed of a half mirror, and the sub mirror 111 reflects a part of the subject light flux that has passed through the main mirror 105 and guides it to the AF unit.

  The penta roof prism 109 converts the subject luminous flux formed on the focusing screen 106 into an erect image of the subject, and the converted subject image is guided to the AE sensor 108 and observed through the optical viewfinder. The display element 107 displays an AF distance measurement frame such as a PN liquid crystal and indicates a position at which the user is focused when looking into the optical viewfinder.

  The image sensor 103 is configured by a CCD sensor, a CMOS sensor, or the like, and includes an infrared cut filter, a low-pass filter, and the like. The image sensor 103 photoelectrically converts a subject image that has passed through the lens unit 200 during imaging and outputs an image signal. The shutter 104 is closed during non-shooting to shield the image sensor 103 and is opened during shooting to guide subject light to the image sensor 103.

  The AE sensor 108 uses not only photometry but also scene recognition such as face detection and flicker detection by using an image sensor such as a CCD sensor or a COMS sensor. In the present embodiment, a CMOS sensor having a QVGA resolution is used as the AE sensor 108. This CMOS sensor has a pixel addition mode for reading at a resolution obtained by adding two vertical pixels of the same color in addition to an all pixel mode for reading all the pixels of QVGA. Then, 2 ms is required for reading in the all pixel mode, and 1 ms is required for reading in the pixel addition mode. Note that the time shown here is an example, and is intended to indicate that the readout time in the pixel addition mode is shorter than the readout in the all pixel mode.

  The ICPU 112 is a CPU for driving control of the AE sensor 108 and image processing / calculation. Based on the photometry result of the AE sensor 108, the ICPU 112 performs a period of light amount change of the light from the subject to be described later and a timing at which the light amount change satisfies a predetermined condition in addition to face detection calculation, tracking calculation, photometry calculation, and the like. Calculation of light quantity change characteristics such as calculation is also performed. Note that the light amount change characteristic of light from the subject calculated by the ICPU 112 corresponds to the light amount change characteristic of a light source (flicker light source) whose light amount periodically changes according to the frequency of the commercial power supply. In the following description, A periodic change in the amount of light from the subject is also called flicker.

  A memory 113 such as a RAM or a ROM is connected to the ICPU 112. In the present embodiment, the dedicated ICPU 112 for the AE sensor 108 is used. However, all processing of the ICPU 112 may be performed by the camera CPU 101. The LPU 201 is a CPU in the lens unit 200, and sends distance information to the subject to the camera CPU 101.

  Further, in the camera of the present embodiment, when a release button (not shown) is pressed halfway, the release switch SW1 is turned on to perform shooting preparation operations such as AF and AE, and the release button is fully pressed. Then, the release switch SW2 is turned on and the photographing operation is performed.

  Next, an operation example of the digital single-lens reflex camera will be described with reference to FIGS. FIG. 2 is a flowchart for explaining the operation of the camera when performing continuous shooting while calculating the period of light amount change from the subject and the timing at which the light amount change satisfies a predetermined condition. 2 is executed by the ICPU 112 under the control of the camera CPU 101 by developing a program stored in the ROM or the like of the memory 113 in the RAM.

  In FIG. 2, in step S201, when the release switch SW1 is turned on and continuous shooting is started, the ICPU 112 determines whether or not flicker exists, and if so, the flicker light amount change period (frequency). Determine how many.

  In the present embodiment, in order to detect flicker, as shown in FIG. 3, the accumulation and readout of the pixels of the AE sensor 108 are continuously performed a plurality of times (here, 12 times) at a cycle of 600 fps and about 1.667 ms. . Generally, the frequency at which the brightness of the flicker light source changes is twice the frequency of the commercial power supply. Therefore, in the power supply region where the power supply frequency is 50 Hz, the frequency is 100 Hz, and the light quantity change period is 10 ms. Similarly, in a region where the power supply frequency is 60 Hz, the frequency is 120 Hz and the light amount change period is 8.33 ms.

  FIG. 3A is a graph showing the output of the AE sensor 108 when pixels are accumulated at intervals of 1.667 ms when there is a flicker light source with a power frequency of 50 Hz. In FIG. 3A, the accumulation of the nth pixel is “accumulation n”, the readout of the accumulation n is “read n”, and the photometric value obtained from the result of readout n is “AE (n)”. Regarding the acquisition time of each photometric value in FIG. 3A, since accumulation is performed in a finite time, it is represented by the median value during the accumulation period.

  As described above, the flicker light amount change period at the time of commercial power supply 50 Hz is 10 ms and 10 ÷ 1.667≈6. Therefore, as shown in FIG. Photometric values approximately equal in period can be obtained. That is, the relationship of AE (n) = AE (n + 6) is established.

  Similarly, the flicker light quantity change period at the commercial power supply of 60 Hz is 8.33 ms and 8.33 ÷ 1.667≈5 as described above, and therefore, as shown in FIG. Nearly equal photometric values are obtained in the rotation cycle, and the relationship is AE (n) = AE (n + 5).

  On the other hand, in an environment where flicker does not exist, AE (n) is constant regardless of n. As described above, the evaluation value F50 and the evaluation value F60 are respectively defined by the following expressions (1) and (2), and the threshold value F_th is used to determine whether or not flicker exists. It is possible to determine how many change periods (frequency) are.

  That is, when F50 <F_th and F60 <F_th are satisfied, it can be determined that the environment does not have flicker. Further, when F50 <F_th and F60 ≧ F_th are satisfied, it can be determined that the flicker environment has a light amount change period T = 10 ms (power supply frequency 50 Hz). Furthermore, when F50 ≧ F_th and F60 <F_th are satisfied, it can be determined that the flicker environment has a light amount change period T = 8.33 ms (power supply frequency 60 Hz).

  Further, there may be a case where both F50 and F60 exceed F_th due to panning or movement of the subject. In this case, the sizes of F50 and F60 are compared. If F50 is smaller, it is determined that the flicker environment has a light quantity change period T = 10 ms (power supply frequency 50 Hz). If F60 is smaller, it is determined that the flicker environment has a light quantity change period T = 8.33 ms (power supply frequency 60 Hz).

  That is, when F50 ≧ F_th and F60 ≧ F_th are satisfied, it is determined that the flicker environment has a light quantity change period T = 10 ms (power supply frequency 50 Hz) in F50 ≦ F60. In F50> F60, it is determined that the flicker environment has a light quantity change period T = 8.33 ms (power supply frequency 60 Hz).

  Alternatively, in such a case, the flicker detection may be performed again, assuming that the reliability of the flicker detection result is low. In this way, by calculating evaluation values such as F50 and F60, whether there is flicker in the shooting environment, if it exists, the power supply frequency is 50/60 Hz, and the light quantity change period at that time T can be calculated.

  FIG. 4 is a diagram for explaining an example of a method for calculating flicker peak timing. In FIG. 4, photometry by the AE sensor 108 is continuously performed 12 times at predetermined time intervals in an environment with flicker, and the results are defined as AE (1) to AE (12). The point where the maximum output is obtained among AE (1) to AE (12) is P2 (t (m), AE (m)), and the point of the previous photometric result is P1 (t (t ( m-1), AE (m-1)), and the point of the next photometric result is P3 (t (m + 1), AE (m + 1)).

  First, a straight line passing through two points of a point (point P3 in the example of FIG. 4) and point P2 that takes the smaller one of AE (m−1) and AE (m + 1) is obtained as L1 = at + b. Further, a straight line having a slope of −a is obtained as L2 through a point (P1 in the example of FIG. 4) that takes the larger of point P1 and point P3. When the calculated intersection of L1 and L2 is calculated, the flicker peak timing t (peak) can be obtained.

  In this method, the flicker peak timing is calculated using the point at which the maximum output is obtained among AE (1) to AE (12) and the output once before and after that point. Therefore, when AE (1) or AE (12) is the maximum output, peak timing cannot be calculated because AE (0) and AE (13) do not exist. However, the flicker light quantity change period is known at this point. Therefore, for example, if it is known that the light quantity change period T = 10 ms, AE (0) = AE (6), AE (13) = AE (7) from the relationship of AE (n) = AE (n + 6). ) Can calculate AE (0) and AE (13), and the peak timing can be obtained.

  As described above, the processing in step S201 shows the presence / absence of flicker, if there is flicker, the light amount change period (frequency), and its peak timing. Therefore, if the flicker periods T and k are natural numbers, it can be seen that after step S201, flicker peaks come every t (peak) + kT.

  If the power frequency of the flicker light source is determined to be 50 Hz or 60 Hz, the ICPU 112 proceeds to step S203, and if it is determined that there is no flicker, the process proceeds to step S202.

  In step S202, the ICPU 112 proceeds to step S204 while the release switch SW2 is turned on, performs the imaging operation once in the normal sequence, and repeats this to realize continuous shooting.

  FIG. 5 is a timing chart showing a sequence of camera operations during continuous shooting. In the normal sequence, as shown in FIG. 5A, charge accumulation and readout of the AE sensor 108 are performed in the all pixel mode. The hatched portion in the figure represents readout, and as already described, the readout time in the all pixel mode of the AE sensor 108 in this embodiment is 2 ms.

  Thereafter, the main mirror 105 and the sub mirror 111 are raised and retracted from the photographing optical path, and after performing a photographing operation, the main mirror 105 and the sub mirror 111 are again lowered and returned to their original positions. Here, as an example, when the charge accumulation time of the AE sensor 108 is 9 ms, the AE time including reading is 11 ms. Although illustration is omitted, the AF sensor is also accumulated in parallel with the AE time, and the AF / AE operation is performed.

  In the AF / AE operation, a scene analysis result is fed back from a high resolution image acquired by the AE sensor 108. After that, assuming that it takes 30 ms for up and down mirrors 105 and 111 and 1 ms for exposure, and the rear curtain travel time for shutter 104 is 3 ms, the cycle time required to capture one image is taken. Becomes 75 ms, and the frame speed becomes 13.3. The ICPU 112 repeats the normal sequence of step S204 until the release switch SW2 is turned off, and the continuous shooting ends when the release switch SW2 is turned off.

  On the other hand, in step S203, the ICPU 112 proceeds to step S205 as long as the release switch SW2 is turned on even when flicker is detected, performs a single shooting operation in the flicker sequence, and repeats this to realize continuous shooting. Is done. The flicker peak timing has already been detected in step S201, and it is expected that a peak timing will come every t (peak) + kT.

  However, the power supply frequency is not always 50 Hz or 60 Hz, and a certain amount of deviation may be considered, and there may be an error in the clock that measures the time inside the camera, and the expected peak timing may be displaced. Therefore, during continuous shooting operation when flicker is detected, flicker peak timing is detected for each frame. However, since the flicker frequency is determined by the power supply frequency, it is considered very rare that the flicker frequency changes during continuous shooting. Therefore, during continuous shooting, frequency determination is performed in step S201 when continuous shooting is started, and peak timing is detected in step S205 during continuous shooting.

  FIG. 5B is a time chart showing an example of the flicker sequence in step S205 of FIG. Here, as an example, a light amount change of flicker (light amount change period T = 10 ms) driven by a 50 Hz power source is also shown. Since only the flicker peak timing is detected during continuous shooting, as shown in FIG. 5B, charge accumulation / readout is continuously performed 6 times at a period of about 1.667 ms. This high-speed driving cannot be performed because the readout time is 2 ms in the all-pixel mode, and can be realized only in the pixel addition mode with a readout time of 1 ms.

  However, when photometry is performed, a scene analysis result such as face detection can be better performed with an image having as high a resolution as possible. Therefore, charge accumulation for photometry and scene analysis is performed for 9 ms in the all-pixel mode as in the normal sequence of FIG. 5A, and thereafter, accumulation for flicker peak timing detection is performed six times in the pixel addition mode. . The accumulation for six times in the 1.667 ms period for peak detection is accumulated for a total of 10 ms, which is longer than the flicker period of 10 ms or 8.33 ms. For this reason, the peak timing is always included in the six accumulations regardless of the power supply frequency of 50 Hz / 60 Hz.

  Using these six photometric results AE (1) to AE (6) and using the method shown in FIG. 4, the peak timing can be calculated. In the case of FIG. 5B, the time required for AE is 22 ms, which is about 11 ms longer than the case of no flicker in FIG. In this manner, a method of accumulating over one period of flicker is referred to as “accumulating for one period”.

  Next, with reference to FIG. 5C, consider shortening the AE time as compared with FIG. 5B. Here, when flicker is detected in step S201 in FIG. 2, the fact that peak timing can be predicted is used. The flicker peak timing calculation method shown in FIG. 4 interpolates and calculates a total of three photometric values, that is, the photometric value at the peak and one photometric value before and after that. Therefore, for example, when the predicted time P of the flicker peak timing is known as the time P1, as shown in FIG. 5C, it is not necessary to perform charge accumulation for flicker detection as many as six times. In this case, accumulation is performed only three times with a period of 1.667 ms, and the center of the second accumulation may be time P1.

  In the case of FIG. 5C, the time required for AE is 17 ms, which can be shortened by 5 ms compared to FIG. In this way, the method of accumulating charges only three times near the predicted peak timing is referred to herein as “limited accumulation near the peak”.

  However, it is not always necessary to perform “limited accumulation near the peak”. For example, as shown in FIG. 5D, in the case of the flicker peak timing prediction time P = P2 (> P1), the time required for AE is 23.5 ms, which is more than that in the case of FIG. Will also extend by about 1.5 ms. Thus, the AE time is shorter than in the case of FIG. 5B when the peak timing prediction time P is 14 ms ≦ P ≦ 19 ms.

  That is, when 14 ms ≦ P ≦ 19 ms is satisfied, “limited accumulation near peak” in FIG. 5C is used, and when 14 ms ≦ P ≦ 19 ms is not satisfied, “accumulation for one period” in FIG. It can be seen that the AE time can be shortened by using.

  As described above, in step S205 of FIG. 2, the peak timing detection process is performed at a high speed by selectively switching between “limited accumulation near peak” and “accumulation for one period” according to the flicker peak timing prediction time P. Can be The ICPU 112 repeats the flicker sequence in step S206 until the release switch SW2 is turned off. When the release switch SW2 is turned off, the continuous shooting ends.

  5 (c) and 5 (d), the case where the charge accumulation is performed only three times in the 1.667 ms cycle as “limited accumulation near the peak” has been described, but the case where the cycle and the number of times are changed will be described. .

  In the example of FIG. 5C and FIG. 5D, an ideal state in which the peak timing comes in accordance with the peak timing prediction time P is described. However, in reality, there may be a case where the power supply frequency is slightly deviated from 50 Hz / 60 Hz, or the clock frequency used for timing in the camera is deviated from an assumed value due to individual variation, temperature, or the like.

  For example, when the actual power supply frequency is 50.2 Hz, the camera determines that the power supply is 50 Hz in step S201 in FIG. 2, and the peak timing is predicted with the flicker frequency set at 100 Hz. However, since the actual flicker frequency is 100.4 Hz, the deviation from the predicted peak timing increases as time elapses from the peak timing detected in step S201.

  For example, assume that it takes 400 ms to detect flicker of a 50 Hz power source in step S201 and to perform the flicker sequence in step S205. Since the predicted peak timing comes at an interval of 10 ms from the peak detected in step S201, it is a timing after 400 ms that is a multiple of 10 ms from the peak timing detected in step S201.

  On the other hand, if the flicker is 100.4 Hz, the actual peak timing comes after about 398.4 ms, which is slightly earlier than predicted, by the flicker frequency ratio. That is, the prediction shifts from the predicted peak timing by about 1.6 ms, and “peak vicinity limited accumulation” is performed at the timing shown in FIG.

  In the example of FIG. 6, if there is a peak timing deviation of 1.6 ms from the prediction, the first time out of the three times of metering will be the maximum value, so that the interpolation calculation using the algorithm described in FIG. The timing cannot be detected. In this case, it is possible to avoid the detection of the peak timing by increasing the interval of charge accumulation performed three times from 1.667 ms or increasing the number of accumulations from three times.

  Regarding the deviation of the clocking clock inside the camera, the deviation from the peak timing of prediction can be calculated by considering the maximum error of the clock in the same manner as the deviation of the power supply frequency. That is, the deviation from the peak timing prediction can be calculated from the elapsed time since the last peak timing, the assumed maximum power supply frequency deviation, and the maximum error of the time clock. Then, it is possible to avoid the peak timing from becoming uncalculated by increasing the accumulation interval or increasing the number of accumulations according to the deviation amount.

  As described above, in the present embodiment, the time required for detecting the peak timing can be further shortened by changing the photometry timing and photometry interval for detecting the flicker peak timing.

  In addition, this invention is not limited to what was illustrated to the said embodiment, In the range which does not deviate from the summary of this invention, it can change suitably.

  The present invention can also be realized by executing the following processing. That is, the present invention supplies a program that realizes one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in the computer of the system or apparatus read the program. It can also be realized by processing to be executed. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.

104 Shutter 108 AE sensor 112 ICPU

Claims (7)

Photometric means;
Detecting means for detecting a flicker peak timing from a plurality of photometric results obtained by the photometric means;
When the peak timing of the flicker can be predicted from the detection result of the detection unit, the photometry unit performs a plurality of times of photometry in the vicinity of the predicted peak timing and in a shorter time than when the peak timing cannot be predicted. An imaging device comprising: a control means for performing the operation.   Whether the control means performs photometry a plurality of times in the vicinity of the predicted peak timing in a shorter time than the case where the peak timing cannot be predicted in accordance with the predicted peak timing. The imaging apparatus according to claim 1, wherein whether to perform photometry a plurality of times by the photometry unit over one cycle or more of the flicker is selected.   The control means, when performing the photometry by the photometry means a plurality of times in the vicinity of the predicted peak timing and in a shorter time than when the peak timing cannot be predicted, The imaging apparatus according to claim 1, wherein an interval at which photometry is performed a plurality of times by the photometry unit is changed according to an error.   The control means, when performing the photometry by the photometry means a plurality of times in the vicinity of the predicted peak timing and in a shorter time than when the peak timing cannot be predicted, The imaging apparatus according to claim 1, wherein the number of times of performing photometry a plurality of times by the photometry unit is changed according to an error.   The control means performs a plurality of times of light metering by the light metering means when performing the light metering by the light metering means in the vicinity of the predicted peak timing and in a shorter time than when the peak timing cannot be predicted. 5. The imaging apparatus according to claim 1, wherein a central timing of a period to be performed is matched with the predicted peak timing. 6. A method for controlling an imaging apparatus including photometric means,
A detection step of detecting a flicker peak timing from a plurality of photometric results obtained by the photometric means;
When the peak timing of the flicker can be predicted from the detection result in the detection step, the photometry means performs a plurality of times of photometry in the vicinity of the predicted peak timing and in a shorter time than when the peak timing cannot be predicted. And a control step for performing an imaging method.   A program for causing a computer to execute each step of the method for controlling an imaging apparatus according to claim 6.

Images