JP2015200908A - imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging device for achieving a focus operation having higher accuracy.SOLUTION: The disclosed imaging device includes an imaging part for imaging a formed subject image through a focus lens to generate image data, an image processing part for calculating a value about a focus position on the basis of first image data generated by the imaging part when the focus lens is located at a first position, and second image data generated by the imaging part when the focus lens is located at a second position obtained by moving a first movement amount from the first position, a memory for previously storing a plurality of point spread functions corresponding to a plurality of subject distances, and a control part for selecting a plurality of necessary point spread function from many point spread functions to notify the image processing part of the plurality of selected point spread functions. The image processing part creates third image data by an operation using the first image data and the plurality of selected point spread functions, and uses the second image data and the third image data to calculate a value about the focus position.

Description

  The present disclosure relates to an imaging apparatus.

  Patent Document 1 discloses an imaging apparatus that employs a DFD (Depth From Defocus) system. The imaging apparatus performs focus control so that a plurality of types of blur can be obtained, and acquires a plurality of images having different blur sizes by the imaging element. Next, the imaging apparatus calculates the subject distance based on a plurality of images having different blur sizes. Then, the imaging apparatus performs focus control based on the calculated subject distance.

JP 2011-15163 A

  By adopting the DFD system, it is expected to increase the speed of the focusing operation on the subject. On the other hand, in order to employ the DFD system, there is a demand for higher focusing accuracy.

  The present disclosure provides an imaging apparatus that realizes a focusing operation with higher accuracy.

  The imaging device according to the present disclosure is generated by an imaging unit that captures a focus lens, an object image formed via the focus lens, and generates image data, and the focus lens is at a first position. Based on the first image data and the second image data generated by the imaging unit when the focus lens is at the second position moved the first movement amount from the first position, focusing is performed. An image processing unit that calculates values related to the position, a memory that stores a plurality of point spread functions corresponding to a plurality of object distances in advance, and a plurality of point spread functions that are required from a number of point spread functions stored in the memory are selected. And a control unit that notifies the image processing unit. The image processing unit creates third image data by calculation using the first image data and a plurality of selected point spread functions, and uses the second image data and the third image data to focus the position. Calculate the value for.

  ADVANTAGE OF THE INVENTION According to this invention, the imaging device which implement | achieves a highly accurate focusing operation can be provided.

1 is a block diagram showing an electrical configuration of a digital video camera in an embodiment Block diagram of operation of digital video camera in embodiment Image diagram of operation of digital video camera in embodiment Diagram of DFD operation of digital video camera in embodiment Flowchart of autofocus operation in digital video camera of embodiment An image diagram for explaining an autofocus operation in the digital video camera of the embodiment The figure which shows the zoom tracking table with respect to several object distance of the digital video camera in embodiment FIG. 3 is an image diagram for explaining the accuracy of DFD calculation in the digital video camera of the embodiment.

  Hereinafter, embodiments will be described in detail with reference to the drawings as appropriate. However, more detailed description than necessary may be omitted. For example, detailed descriptions of already well-known matters and repeated descriptions for substantially the same configuration may be omitted. This is to avoid the following description from becoming unnecessarily redundant and to facilitate understanding by those skilled in the art.

  In addition, the inventors provide the accompanying drawings and the following description in order for those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims. Absent.

  One of many methods for measuring the subject distance, which is the distance from the imaging device to the subject, is a technique called DFD (Depth from Defocus) that uses a correlation value of a blur amount (Defocus) generated in a captured image. . In general, the amount of blur that occurs in a captured image is uniquely determined for each imaging device in accordance with the relationship between the focus position at the time of shooting and the subject distance. In the DFD calculation, this characteristic is used to intentionally create two images with different amounts of blur by changing the focus position, and the subject distance is calculated from the difference in the amount of blur and the point spread function (PSF). measure. The imaging apparatus according to the embodiment performs autofocus control by measuring a subject distance using DFD calculation.

  Hereinafter, the configuration and operation of the imaging apparatus according to the embodiment will be described.

[1. Electrical configuration of imaging device]
FIG. 1 is a block diagram illustrating an electrical configuration of a digital video camera 100 that is an imaging apparatus according to an embodiment. The digital video camera 100 includes an optical system 110 including one or more lenses. The optical system 110 forms a subject image on the CMOS image sensor 140 by light from the subject. The formed subject image is picked up by a CMOS image sensor 140 which is an image pickup sensor. The CMOS image sensor 140 generates image data based on the captured subject image. Image data generated by the CMOS image sensor 140 is converted into a digital signal by an A / D converter (ADC) 150, and then subjected to various processes by the image processing unit 160 and stored in the memory card 200. Hereinafter, the configuration of the digital video camera 100 will be described in detail.

  The optical system 110 in the embodiment includes a zoom lens 111, a camera shake correction lens 112, a focus lens 113, and a diaphragm 114. The subject image can be enlarged or reduced by moving the zoom lens 111 along the optical axis 110X. Further, the focus of the subject image can be adjusted by moving the focus lens 113 along the optical axis 110X. The camera shake correction lens 112 is movable in a plane perpendicular to the optical axis 110X of the optical system 110. By moving the camera shake correction lens 112 in a direction in which the blur of the digital video camera 100 is canceled, the influence of the blur of the digital video camera 100 on the captured image can be reduced. The diaphragm 114 has an opening 114A located on the optical axis 110X, and adjusts the size of the opening 114A according to the setting of the user or automatically to adjust the amount of transmitted light.

  The lens driving unit 120 includes a zoom actuator that drives the zoom lens 111, a camera shake correction actuator that drives the camera shake correction lens 112, a focus actuator that drives the focus lens 113, and a diaphragm actuator that drives the diaphragm 114. The lens driving unit 120 controls the zoom actuator, the focus actuator, the camera shake correction actuator, and the aperture actuator.

  The CMOS image sensor 140 captures a subject image formed by the optical system 110 and generates analog image data that is an analog signal. The CMOS image sensor 140 performs various operations such as exposure, transfer, and electronic shutter.

  The A / D converter 150 converts the analog image data generated by the CMOS image sensor 140 into digital image data that is a digital signal.

  The image processing unit 160 performs various processes on the image data generated by the CMOS image sensor 140, generates image data to be displayed on the display monitor 220, and stores image data to be stored in the memory card 200. Or generate. For example, the image processing unit 160 performs various processes such as gamma correction, white balance correction, and flaw correction on the image data generated by the CMOS image sensor 140. The image processing unit 160 converts the image data generated by the CMOS image sensor 140 into the H.264 format. It compresses by the compression format etc. based on H.264 standard or MPEG2 standard. The image processing unit 160 can be realized by a DSP, a microcomputer, or the like.

  The control unit 180 controls the entire digital video camera 100. The control unit 180 can be realized by a semiconductor element or the like. The control unit 180 may be configured only by hardware, or may be realized by combining hardware and software. The control unit 180 can be realized by a microcomputer or the like.

  The buffer 170 functions as a work memory for the image processing unit 160 and the control unit 180. The buffer 170 can be realized by, for example, a DRAM or a ferroelectric memory.

  The card slot 190 holds the memory card 200 in a detachable manner. The card slot 190 can be mechanically and electrically connected to the memory card 200. The memory card 200 includes a flash memory, a ferroelectric memory, and the like, and can store data such as an image file generated by the image processing unit 160.

  The internal memory 240 is composed of a flash memory or a ferroelectric memory. The internal memory 240 stores a control program for controlling the entire digital video camera 100 and the like. The internal memory 240 stores a point spread function.

  The operation member 210 is a generic term for user interfaces that accept operations from the user. The operation member 210 includes, for example, a cross key that accepts an operation from the user, a determination button, and the like.

  The display monitor 220 includes a screen 220 </ b> A that can display an image indicated by the image data generated by the CMOS image sensor 140 or an image indicated by the image data read from the memory card 200. The display monitor 220 can also display various menu screens for performing various settings of the digital video camera 100 on the screen 220A. A touch panel 220B is arranged on the screen 220A of the display monitor 220. Touch panel 220B is touched by the user and can accept various touch operations. The instruction indicated by the touch operation on the touch panel 220B is notified to the control unit 180 and various processes are performed.

  The angular velocity sensor 250 detects an angular velocity generated in the digital video camera 100 due to camera shake or the like. The angular velocity detected by the angular velocity sensor 250 is notified to the control unit 180. The control unit 180 can drive the camera shake correction lens 112 based on the angular velocity notified from the angular velocity sensor 250 so as to eliminate image blurring based on the angular velocity generated in the digital video camera 100.

[2. Operation of digital video camera 100]
The digital video camera 100 performs an autofocus operation using the result of the DFD calculation. FIG. 2 is a block diagram for explaining focus lens control using the result of the DFD calculation of the digital video camera 100 according to the embodiment.

  The DFD operation circuit 161 is an operation circuit configured in the image processing unit 160, and performs a DFD operation to create a depth map. Specifically, the DFD arithmetic circuit 161 changes the focus position, thereby observing the observation image PA based on the observation image PA, the reference image PB, and the point spread function, which are two images with different amounts of blur created intentionally. A depth map indicating the subject distance at each pixel of the (reference image PB) is created.

  Next, the DFD arithmetic circuit 161 notifies the control unit 180 of the created depth map. The control unit 180 controls the lens driving unit 120 to drive the focus lens 113 based on the notified depth map.

  Details of the DFD calculation by the DFD calculation circuit 161 shown in FIG. 2 and details of determination of the subject distance by the control unit 180 will be described below.

  First, details of the DFD calculation performed by the DFD calculation circuit 161 will be described. FIG. 3 is an image diagram for explaining the movement of the focus lens 113 in the DFD calculation of the digital video camera 100 according to the embodiment. In the DFD calculation, the control unit 180 intentionally creates two images with different amounts of blur by changing the focus position. Specifically, as shown in FIG. 3, the control unit 180 controls the lens driving unit 120 to set the position of the focus lens 113 at the time t1 to the focus position L1. Similarly, the position of the focus lens 113 at time t2 is set to a focus position L2 different from the focus position L1. The CMOS image sensor 140 images a subject when the focus lens 113 is at the focus position L1, and creates an observation image PA. Similarly, the CMOS image sensor 140 images the subject when the focus lens 113 is at the focus position L2, and creates a reference image PB. Since the observation image PA and the reference image PB are different in the position of the focus lens 113 when imaged, the amounts of blur are different from each other although they are images of the same subject.

  FIG. 4 is a conceptual diagram illustrating calculation of the subject distance by DFD calculation of the digital video camera 100 according to the embodiment. The DFD calculation circuit 161 performs DFD calculation on each observation pixel SA that constitutes the observation image PA and each reference pixel SB that constitutes the reference image PB, and calculates a subject distance at each pixel SA (SB). The DFD arithmetic circuit 161 collates a plurality of observation pixels CA, which is a result obtained by convolving a plurality of point spread functions with the observation pixel SA, with a reference pixel SB having the same coordinates on the image as the observation pixel CA. The operation will be described below.

  The point spread function is a function indicating the response of the optical system to the point light source, and indicates a change in the amount of blur. By convolutioning the point spread function with an image corresponding to a set of point light sources (convolution), it is possible to artificially generate a blurred image. In the present embodiment, a large number of point spread functions are prepared in advance in the internal memory 240 corresponding to a large number of subject distances. The control unit 180 divides the subject distance from the closest side to the far side into 16 subject distances, and the 16 point spread functions PSF1 to PSF16 respectively corresponding to the subject distances are provided in a number of point spreads prepared in the internal memory 240. Select from functions. At this time, the resolution of the subject distance in 16 steps from the closest side to the far side is changed according to the movement amount of the focus lens 113. That is, as the movement amount of the focus lens 113 decreases, the control unit 180 narrows the interval of the subject distance that is decomposed into 16 steps from the closest side to the far side. On the other hand, as the movement amount of the focus lens 113 increases, the control unit 180 widens the subject distance interval that decomposes the closest side to the far side into 16 steps. Then, the control unit 180 notifies the DFD arithmetic circuit 161 of the selected 16 point spread functions PSF1 to PSF16.

  The DFD operation circuit 161 convolves the 16 point spread functions PSF1 to PSF16 corresponding to the distance to the subject at each observation pixel SA into each observation pixel SA, thereby obtaining the distance to the subject at each observation pixel SA. Corresponding 16 observation pixels CA1 to CA16 are generated. Since the 16 observation pixels CA1 to CA16 have different point spread functions that are convoluted, different blurred images are formed.

  Subsequently, the DFD arithmetic circuit 161 collates the observation pixels CA1 to CA16 with the reference pixel SB, and determines an observation pixel CAn having a minimum difference value from the reference pixel SB among the observation pixels CA1 to CA16. Then, the subject distance corresponding to the point spread function convolved with the observation pixel CAn is determined to be the subject distance of the observation pixel SA. For example, when the difference value between the observation pixel CA3 and the reference pixel SB is minimum compared to the difference value between each of the other observation pixels CA1 to CA2 and CA4 to 16 and the reference pixel SB, the DFD arithmetic circuit 161 The object distance corresponding to the point spread function PSF3 convoluted with the observation pixel SA when the observation pixel CA3 is created is determined to be the object distance at the observation pixel SA.

  By performing the above operation for each pixel of the observation image PA and the reference image PB, the DFD arithmetic circuit 161 completes a Depth map that is a distribution of subject distances for each pixel. In the embodiment, since 16 point spread functions corresponding to the distance to each subject are used, the Depth map indicates a subject distance of 16 gradations.

  Subsequently, an autofocus operation using the DFD calculation result will be described with reference to FIGS. 5 and 6. FIG. 5 is a flowchart of the autofocus operation in the digital video camera 100. FIG. 6 is an image diagram for explaining an autofocus operation in the digital video camera 100. FIG. 6 shows an image of the contrast value with respect to the subject distance at each position P1 to P4 of the focus lens 113. In FIG. 6, positions P <b> 1 to P <b> 4 indicate the position of the focus lens 113. The movement amounts D1 to D3 indicate the movement amount of the focus lens 113. The subject distances S1 to S5 indicate the distance to the subject. Further, the contrast curves CC1 to CC4 are contrast curves in which the contrast value peaks when the focus lens 113 is at the positions P1 to P4 at the subject distances S1 to S4, respectively. The subject distance S5 indicates the subject distance where the subject to be focused is present.

  As shown in FIG. 5, the digital video camera 100 performs an autofocus operation in the shooting mode. First, the image processing unit 160 acquires an image captured by the CMOS image sensor 140 at the current position P1 of the focus lens 113 (step S300). Subsequently, the control unit 180 controls the lens driving unit 120 to move the focus lens 113 to the position P2 shifted from the position P1 by the first movement amount D1 (step S301). When the focus lens 113 moves to the position P2, the image processing unit 160 acquires an image captured by the CMOS image sensor 140 (step S302).

  Here, the image processing unit 160 performs the above-described DFD calculation using the image acquired in step S300 as the observation image PA and the image acquired in step S302 as the reference image PB (step S303). The image processing unit 160 calculates the subject distance (information indicating the in-focus position) by DFD calculation.

  Next, the control unit 180 determines whether or not the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 is within a predetermined range (step S304). At this time, it is determined whether the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 is the subject distance corresponding to any of the 16 point spread functions PSF1 to PSF16. As described above, the point spread functions PSF1 to PSF16 correspond to subject distances that are resolved in 16 steps from the closest side to the far side. At this time, the point spread function PSF1 corresponds to the closest subject distance among the point spread functions PSF1 to PSF16. On the other hand, the point spread function PSF16 corresponds to the farthest subject distance among the point spread functions PSF1 to PSF16. Therefore, specifically, the control unit 180 determines whether the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 corresponds to the point spread function PSF1 or the point spread function PSF16 (predetermined range). It is determined whether it corresponds to the point spread functions PSF2 to PSF15 (within a predetermined range).

  When it is determined that the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 is not within the predetermined range (corresponding to the point spread function PSF1 or the point spread function PSF16) (in step S304). “No”), the control unit 180 sets the second movement amount D2 that is larger than the first movement amount D1 as the movement amount of the focus lens 113 in the DFD calculation (step S305). Then, the control unit 180 controls the lens driving unit 120 to move the focus lens 113 from the position P2 to the position P3 shifted by the second movement amount D2 (step S301 in the second round). When the focus lens 113 moves to the position P3, the image processing unit 160 acquires an image captured by the CMOS image sensor 140 (step S302 in the second round).

  Here, the image processing unit 160 performs the DFD operation again using the image acquired in step S302 of the first round as the observation image PA and the image acquired in step S302 of the second round as the reference image PB ( Step S303 of the second round). Then, the image processing unit 160 calculates the subject distance (information indicating the in-focus position) by DFD calculation. Again, the control unit 180 determines whether or not the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 is within a predetermined range (step S304 in the second round). The control unit 180 performs steps S301 to S305 until the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 falls within a predetermined range (corresponding to the point spread functions PSF2 to PSF15). Repeat the operation. The technical significance of increasing the amount of movement of the focus lens 113 when the subject distance calculated by the DFD calculation in step S303 is not within the predetermined range will be described later.

  If it is determined in step S304 that the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 is within a predetermined range (“Yes” in step S304), the control unit 180 may As the amount of movement of the focus lens 113 in the DFD calculation, a third amount of movement D3 that is smaller than the amount of movement of the previous time (first amount of movement D1 or second amount of movement D2) is set (step S306). ). Subsequently, the control unit 180 controls the lens driving unit 120 to move the focus lens 113 from the position P3 to the position P4 shifted by the third movement amount D3 (step S307). When the focus lens 113 moves to the position P4, the image processing unit 160 acquires an image captured by the CMOS image sensor 140 (step S308).

  Here, the image processing unit 160 performs the DFD calculation using the image acquired in the previous step S302 as the observation image PA and the image acquired in step S308 as the reference image PB (step S309). The image processing unit 160 calculates the subject distance (information indicating the in-focus position) by DFD calculation.

  Next, the control unit 180 determines whether or not the subject distance (information indicating the focus position) calculated by the DFD calculation satisfies a predetermined resolution (step S310). In the present embodiment, the resolution that does not cause blur on the image is set as a predetermined resolution.

  When the subject distance (information indicating the focus position) calculated by the DFD calculation does not satisfy the predetermined resolution (“No” in step S310), the control unit 180 causes the image processing unit 160 to perform the DFD calculation again. At this time, the control unit 180 sets a movement amount obtained by further reducing the movement amount from the third movement amount D3 as the movement amount of the focus lens 113 in the DFD calculation (step S306 in the second round). Then, the control unit 180 repeats the processing from step S306 to step S310 until the subject distance (information indicating the focus position) calculated by the DFD calculation in step S309 satisfies a predetermined resolution (“No” in step S310). "). The technical significance of reducing the amount of movement of the focus lens 113 when the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S309 is equal to or less than a predetermined resolution will be described later.

  In step S310, when it is determined that the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S309 satisfies the predetermined resolution (“Yes” in step S310), the control unit 180 performs auto In the focus operation, the focus position where the focus lens 113 should be finally moved is determined based on the subject distance determined by the DFD calculation (step S311). Specifically, the control unit 180 refers to the tracking table and calculates a focus position based on the obtained subject distance and the current position of the zoom lens 111. FIG. 7 shows a zoom tracking table for a plurality of object distances of the digital video camera 100 according to the embodiment. As shown in FIG. 7, the zoom tracking table is a focus position with respect to the position of the zoom lens 111 corresponding to several typical subject distances DL (1 m, 2 m, 3 m, and infinity (INF) in FIG. 7). Including curves DM1 to DM4. The control unit 180 interpolates the curve described in the zoom tracking table with respect to the subject distance, thereby calculating a focus position that is a position to move the focus lens 113 for subject distances other than the representative subject distance DL. Can do.

  Then, the control unit 180 controls the lens driving unit 120 to move the focus lens 113 to the in-focus position (step S312).

  With the autofocus operation described above, the digital video camera 100 can ensure focusing accuracy.

  Subsequently, when the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S303 is not within a predetermined range (corresponding to the point spread function PSF1 or the point spread function PSF16), the focus lens 113 is used. The technical significance of increasing the amount of movement will be described.

  FIG. 6 shows the relationship of the contrast value with respect to the subject distance when the focus lens 113 is located at positions P1 to P4. In the DFD calculation, the subject distance is calculated from at least two images acquired at different focus lens positions. Specifically, as described above, the image processing unit 160 collates the observation pixel CA that is the result of convolving the point spread function with the observation pixel SA of the observation image PA and the reference pixel SB of the reference image PB. Then, the subject distance is calculated. In this collation, in order to calculate a significant subject distance, the reference image PB needs to have a contrast value equal to or greater than a predetermined value at the target subject distance. The case where the target subject distance does not have a contrast value equal to or greater than the predetermined value means that the contrast value of the subject image located at the target subject distance cannot be detected significantly in the reference image PB. . Therefore, it is necessary to further move the focus lens 113 to a position where the contrast value of the subject image located at the target subject distance can be detected significantly. In the example shown in FIG. 6, the target subject exists at the subject distance S5. At this time, since the contrast value at the subject distance S5 of the contrast curve CC2 having a peak at the subject distance S2 is very small, a significant contrast value at the subject distance S5 cannot be detected from the contrast curve CC2. That is, when the focus lens 113 is located at the position P2, a significant contrast value cannot be detected for the subject at the subject distance S5. At this time, it is possible to move the focus lens 113 by the same first movement amount D1 as in the previous step (Step S301) without increasing the movement amount in Step S305 of FIG. However, by moving the focus lens 113 by the second movement amount D2 that is larger than the first movement amount D1, it is possible to increase the possibility that the contrast value of the subject image located at the target subject distance can be detected significantly. it can. As shown in FIG. 6, by moving the focus lens 113 by the movement amount of the second movement amount D2 that is larger than the first movement amount D1 to be positioned at the position P3, a significant contrast value C1 at the subject distance S5. Can be detected. That is, the contrast curve CC3 having a peak at the subject distance S3 has a significant contrast value C1 at the subject distance S5.

  Next, the technical significance of reducing the amount of movement of the focus lens 113 when the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S309 is lower than a predetermined resolution will be described.

  Decreasing the amount of movement of the focus lens 113 means bringing the reference image PB closer to the observation image PA. In the DFD calculation, if the contrast value indicated by the observed image PA and the contrast value indicated by the reference image PB are slightly overlapped, some kind of collation result can be obtained. The accuracy in matching the observation image PA with the reference image PB will be described with reference to FIG. FIG. 8 is an image diagram for explaining the accuracy of the DFD calculation. As shown by contrast curves CC3 and CC4 in FIG. 8, the slope becomes steeper as it approaches the peak position from the vicinity of the base, but the slope is gentle in the vicinity of the base. Therefore, even if there is a range in which the contrast value indicated by the observed image PA and the contrast value indicated by the reference image PB are superimposed, if the overlapped range is near the base, the subject with respect to the predetermined difference ΔC in the contrast value The difference in distance (for example, the difference ΔS3 in the subject distance with respect to the predetermined difference ΔC in the contrast value at the subject distance S5 in the contrast curve CC3) becomes large, and sufficient subject distance detection accuracy cannot be ensured. On the other hand, when the overlapping range is closer to the peak position side from the base side, the difference of the subject distance with respect to the predetermined difference ΔC of the contrast value (for example, the predetermined difference ΔC of the contrast value at the subject distance S5 of the contrast curve CC4). The difference ΔS4) in the subject distance with respect to can be reduced, and sufficient subject distance detection accuracy can be ensured.

  In the step of reducing the movement amount of the focus lens 113 in step S306 shown in FIG. 5, the subject distance (information indicating the in-focus position) calculated by the DFD calculation in step S304 is already within a predetermined range (points). It corresponds to the spread functions PSF2 to PSF15). This is because the contrast value C1 at the subject distance S5 of the contrast curve CC3 shown in FIG. 6 indicates the contrast value for the target subject in the reference image PB acquired at step S302 (observed image PA in the DFD calculation at step S309). However, it has shown that it exists in the peak position side comparatively rather than the base vicinity. Therefore, as in the contrast curve CC4 shown in FIGS. 5 and 6, the contrast value for the target subject in the reference image PB acquired in step S308 is made closer to the peak position side, so that the DFD calculation in step S309 is performed. In addition, the detection accuracy of the subject distance can be increased. Thus, since the observation image PA at the time of the DFD calculation in step S309 is already on the peak position side from the vicinity of the base, the amount of movement of the focus lens 113 is reduced, so that the contrast of the target subject in the reference image PB is reduced. The value can be further asymptotically closer to the peak position. In the example shown in FIG. 6, by moving the focus lens 113 to the position P4 with a smaller third movement amount D3, a more significant contrast value C2 can be detected at the subject distance S5. Thereby, the precision of the focusing operation based on the DFD calculation result can be improved.

[3. Effect etc.)
As described above, in the digital video camera 100, the focus lens 113, the CMOS image sensor 140 that captures the subject image formed through the focus lens 113 and generates image data, and the focus lens 113 are in the first position. The first image data (observation image PA) generated by the CMOS image sensor 140 and the focus lens 113 in the second position moved by the first movement amount from the first position. Based on the second image data (reference image PB) generated by the image sensor 140, an image processing unit 160 that calculates a value related to the in-focus position, and a plurality of point spread functions corresponding to a plurality of subject distances in advance. Select the required multiple point spread functions from the memory to be stored and the many point spread functions stored in the memory. And a control unit 180 notifies the image processing unit 160 and, the. The image processing unit 160 creates third image data by calculation using the first image data and a plurality of selected point spread functions, and focuses using the second image data and the third image data. Calculate the position value. Thereby, the precision of the value regarding the already calculated in-focus position can be further improved.

  In general, in an autofocus method called a contrast autofocus method, only a high-frequency component of a captured image at a certain focus lens position is extracted, and the focus position cannot be determined by that alone. On the other hand, in the autofocus method using the DFD calculation, the result of the focus position has already been calculated corresponding to the subject distance. In the digital video camera 100 of the present disclosure, the in-focus position has already been calculated using the DFD calculation. However, by changing the movement amount of the focus lens 113, the accuracy of the in-focus position calculation result can be further improved. it can.

  In addition, the control unit 180 determines whether or not the value related to the in-focus position is within a predetermined range. Then, when the control unit 180 determines that the value related to the in-focus position is not within the predetermined range, the control unit 180 shifts the next movement amount of the focus lens 113 to the second movement amount that is increased from the first movement amount. Then, the value relating to the in-focus position is calculated again. Thereby, the focusing operation can be performed more reliably so that the value relating to the in-focus position falls within a predetermined range.

  On the other hand, when it is determined that the value related to the in-focus position is within the predetermined range, the control unit 180 shifts the next movement amount of the focus lens 113 to the third movement amount that is reduced from the first movement amount. Then, the value relating to the in-focus position is calculated again. Thereby, the precision of the value regarding the focus position which is already within the predetermined range can be further increased.

  Then, when the control unit 180 determines that the value relating to the calculated focus position is within the second predetermined range (satisfies a predetermined resolution), the control unit 180 moves the focus lens 113 to the calculated focus position. Move. Thereby, a highly accurate focusing operation can be realized.

[4. Other Embodiments]
As described above, the embodiments have been described as examples of the technology disclosed in the present application. However, the technology in the present disclosure is not limited to this, and can also be applied to an embodiment in which changes, replacements, additions, omissions, and the like are appropriately performed. Moreover, it is also possible to combine each component demonstrated in the said embodiment and it can also be set as a new embodiment.

  In the above embodiment, the point spread function is prepared in the internal memory 240. The present invention is not limited to this, and the point spread function may be prepared in a memory in the image processing unit 160, for example. In the digital video camera 100 according to the embodiment, 16 point spread functions are selected, but the number of point spread functions to be selected is 16 or more according to the number of gradations of the Depth map. May be less than 16.

  The digital video camera which is the imaging device in the embodiment is a non-lens interchangeable digital video camera, but is not limited thereto, and may be a lens interchangeable digital video camera.

  In the above embodiment, the result of convolution of the point spread functions PSF1 to PSF16 into the observation pixel SA is collated with the reference pixel SB, but the present disclosure is not limited to this. From the observation pixel SA and the reference pixel SB, an omnifocal image whose depth is wide and in focus is once generated, and the result obtained by convolving the point spread functions PSF1 to PSF16 with the omnifocal image is collated with the reference pixel SB. It may be. According to the omnifocal image, since the depth is widely focused, the depth has a wide range of contrast information, and the S / N of the DFD calculation can be improved. Therefore, the calculation accuracy of the in-focus position can be further improved. Furthermore, the result of convolving the point spread functions PSF1 to PSF16 with the omnifocal image may be collated with the observation pixel SA instead of the reference pixel SB. The observation pixel SA and the reference pixel SB have different contrast information with respect to the subject because the positions of the focus lenses are different from each other when an image is captured. Therefore, even if the observation pixel SA is a subject having contrast information, a subject having no contrast information may occur in the reference pixel SB. In such a case, the dynamic range in which the in-focus position can be determined by the DFD calculation can be expanded by collating the result obtained by convolving the point spread functions PSF1 to PSF16 with the omnifocal image with the observation pixel SA. .

  As described above, the embodiments have been described as examples of the technology in the present disclosure. For this purpose, the accompanying drawings and detailed description are provided.

  Accordingly, among the components described in the accompanying drawings and the detailed description, not only the components essential for solving the problem, but also the components not essential for solving the problem in order to illustrate the above technique. May also be included. Therefore, it should not be immediately recognized that these non-essential components are essential as those non-essential components are described in the accompanying drawings and detailed description.

  Moreover, since the above-mentioned embodiment is for demonstrating the technique in this indication, a various change, substitution, addition, abbreviation, etc. can be performed in a claim or its equivalent range.

  The imaging device according to the present disclosure can be applied to a digital video camera, a digital still camera, a mobile phone with a camera function, a smartphone with a camera function, and the like.

100 Digital video camera (imaging device)
DESCRIPTION OF SYMBOLS 110 Optical system 111 Zoom lens 112 Camera shake correction lens 113 Focus lens 114 Aperture 120 Lens drive part 140 CMOS image sensor (imaging sensor)
150 A / D converter 160 Image processing unit 170 Buffer 180 Control unit 190 Card slot 200 Memory card 210 Operation member 220 Display monitor 240 Internal memory 250 Angular velocity sensor

Claims (2)

A focus lens,
An imaging unit that captures a subject image formed via the focus lens and generates image data;
The first image data generated by the imaging unit when the focus lens is at the first position and the second position where the focus lens has moved the first movement amount from the first position. An image processing unit that calculates a value related to the in-focus position based on the second image data generated by the imaging unit at times.
A memory for storing a plurality of point spread functions corresponding to a plurality of subject distances;
A control unit that selects a plurality of necessary point spread functions from a number of point spread functions stored in the memory and notifies the image processing unit;
The image processing unit creates third image data by calculation using the first image data and the selected plurality of point spread functions, and the second image data and the third image data are generated. An imaging apparatus that calculates a value related to the in-focus position.   The imaging apparatus according to claim 1, wherein the number of point spread functions stored in the memory is larger than the number of point spread functions selected by the control unit.

Images