JP2009098436A - Focusing measurement device, method of focusing measurement and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technology with which the focusing level is accurately measured in the focusing measurement of an image even for an image having a large luminance difference. <P>SOLUTION: A focusing measurement device of a first invention comprises: an image input part which captures an imaged image; a filter part which generates an out-of-focus image by using a pair of filters which have respectively different frequency characteristics for the image; an extraction part which extracts a first amount of feature and a second amount of feature from the image and the out-of-focus image, respectively, at a plurality of points on the image; a calculation part which calculates a focusing quantity which shows the degree of focusing on the basis of the first amount of feature and the second amount of feature; and a measurement part which measures the focusing level at the plurality of points of the image on the basis of the calculated focusing quantity. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an in-focus measuring apparatus, an in-focus measuring method, and a program for evaluating an in-focus level indicating an autofocus of an imaging apparatus such as an electronic camera or a focus level in a captured image.

  Conventionally, whether or not the image is in focus is determined based on the contrast of the image in a specific region such as the entire image region or the autofocus selection region. That is, it is determined that the focus is higher as the contrast is higher and the focus is shifted as the contrast is lower.

  In Patent Document 1, high frequency components in a specific region of an image are extracted and integrated. It is disclosed that the integrated value is used as an evaluation value indicating the contrast in a specific region to determine whether or not the subject is in focus.

In Patent Document 2, two high-frequency components having different cutoff frequencies are extracted in a specific region of an image. Based on the evaluation value calculated from the two high-frequency components, the moving speed of the imaging lens of the imaging device is appropriately changed according to the deviation width from the in-focus position, thereby converging to the in-focus position. It discloses disclosing convergence time.
Japanese Patent No. 3840725 Japanese Patent Laid-Open No. 10-221594

  However, in the methods such as Patent Document 1 and Patent Document 2, in an image, the evaluation is performed in an area where the brightness difference is large but the focus difference is smaller than the evaluation value in the area where the brightness difference is small. The value may be larger. As a result, there is an erroneous determination that the image is in focus in a region where the luminance difference is large even though the image is not in focus.

  SUMMARY OF THE INVENTION In view of the above-described problems of the prior art, an object of the present invention is to provide a technique that enables accurate focus measurement even when measuring the focus level of an image having a large luminance difference.

  An in-focus measurement apparatus according to a first invention includes an image input unit that acquires a captured image, a filter unit that generates a blurred image using a pair of filters having different frequency characteristics with respect to the image, and a plurality of images , The extraction unit for extracting the first feature value and the second feature value from the image and each blurred image, and the first feature value and the second feature value are in focus. An arithmetic unit that calculates a focus amount indicating a degree, and a measurement unit that measures focus levels at a plurality of locations on the image based on the calculated focus amount.

  In a second aspect based on the first aspect, the computing unit calculates the difference between the first feature amount and the second feature amount, or the difference as the first feature amount, the second feature amount, or the first feature. A value obtained by dividing the amount by the average value of the amount and the second feature amount is used as a focusing amount.

  In a third aspect based on the first aspect or the second aspect, the first feature amount and the second feature amount are edge amounts obtained from a difference between the image and each blurred image or an absolute value of the difference. It is characterized by that.

  In a fourth aspect based on any one of the first aspect to the third aspect, the arithmetic unit calculates the amount of focus calculated at each of the portions in the predetermined region of the image among the plurality of portions of the image. And a reference value calculation unit that calculates a reference value of the focus level based on the measurement value, and the measurement unit measures the focus level at a plurality of locations of the image based on the magnitude of the value with respect to the reference value of the focus amount. And a reference value measuring unit to be performed.

  According to a fifth aspect, in the fourth aspect, the measurement unit further includes a reference value storage unit that is held to use the reference value for the measurement of the focus level of another image captured by the same imaging device. It is characterized by providing.

  In a sixth aspect based on the fourth aspect, the predetermined area is the entire image area.

  According to a seventh aspect, in the fourth aspect, the predetermined area is an autofocus selection area at the time of shooting.

  According to an eighth invention, in any one of the first to seventh inventions, the filter unit further includes a frequency characteristic adjusting unit that changes the frequency characteristic of the filter in accordance with the ISO sensitivity of the image sensor. .

  According to a ninth aspect of the present invention, there is provided an in-focus determination method for generating a blurred image using a pair of filters having different frequency characteristics with respect to a captured image, and the image and each blurred image at a plurality of locations of the image. Extracting a first feature value and a second feature value from the first feature value, calculating a focus amount indicating a degree of focus from the first feature value and the second feature value, respectively. And a step of measuring focus levels at a plurality of locations of the image based on the calculated focus amount.

  A program of a tenth invention causes a computer to function as the focus measurement device of any of the first to eighth inventions.

  According to the present invention, by extracting a feature amount of an image using a pair of filters having different frequency characteristics and calculating the difference between them, the focus level can be accurately adjusted even in an image having a large luminance difference. It becomes possible to measure.

<< First Embodiment >>
FIG. 1 is a configuration diagram of an electronic camera 10 according to the first embodiment of the present invention. The electronic camera 10 includes an imaging lens 1, an imaging device 2, an A / D conversion unit 3, a buffer memory 4, a storage unit 5, a lens driving unit 6, a CPU 7, an operation member 8, and a monitor 9. The lens driving unit 6 and the imaging element 2 of the imaging lens 1 are connected to the CPU 7 respectively. On the other hand, the buffer memory 4, the storage unit 5, the CPU 7, the operation member 8, and the monitor 9 are connected via a bus 11 so that information can be transmitted.

  Light from the subject is imaged on the imaging surface of the imaging device 2 by the imaging lens 1. The image sensor 2 captures images according to instructions from the CPU 7. As the image sensor 2, a CCD or CMOS semiconductor image sensor or the like can be appropriately selected and used.

  The A / D converter 3 performs A / D conversion on the output of the image signal from the image sensor 2 and outputs a digital signal of the image.

  The buffer memory 4 temporarily stores and holds an image of the subject image captured by the image sensor 2 and A / D converted.

  The lens driving unit 6 moves the imaging lens 1 to an optimal position in focus based on an instruction from the CPU 7.

  The storage unit 5 stores the image held in the buffer memory 4 via the bus 11 according to an instruction from the CPU 7. The image is stored in various formats such as a bitmap format, a jpeg format, and a tiff format. The storage unit 5 stores a focus measurement program executed by the CPU 7. Images, programs, and the like stored in the storage unit 5 can be appropriately referred from the CPU 7 via the bus 11. The storage unit 5 can select and use a storage device such as a general hard disk device, a magneto-optical disk device, or a removable memory card.

  The CPU 7 executes a focus measurement program stored in the storage unit 5 and appropriately reads an image stored in the buffer memory 4 or the storage unit 5 to perform focus measurement. Based on the measurement result, the CPU 7 transmits a command for moving the imaging lens 1 to the optimum position in focus to the lens driving unit 6. At the same time, the CPU 7 stores the measurement result in the storage unit 5 or outputs it to the monitor 9. As the CPU 7, a CPU of a general computer can be used.

  The operation member 8 sends an operation signal corresponding to the content of the member operation by the user to the CPU 7. The operation member 8 includes operation members such as a mode setting button such as a shooting mode, a zoom setting button, and a release button.

  The monitor 9 displays a captured image, a mode setting screen, and the like. As the monitor 9, a liquid crystal monitor or the like can be appropriately selected and used.

  Next, the procedure for focusing measurement according to the first embodiment will be described with reference to the flowchart of FIG.

  Specifically, when the user presses the release button of the operation member 8 halfway, a focus measurement command for imaging the subject is transmitted to the CPU 7. The CPU 7 executes a focus measurement program stored in the storage unit 5. As a result, the processing from step S101 to step S108 in FIG. 2 is performed. In the first embodiment, the subject to be imaged by the electronic camera 10 is a landscape around the soccer goal as shown in FIG. An autofocus selection area 20 which is a reference for focusing is set on the grass under the soccer goal.

  In step S <b> 101, the CPU 7 causes the image pickup device 2 to pick up a subject image formed on the image pickup surface of the image pickup device 2 by the image pickup lens 1 at the current position. The captured image is held in the buffer memory 4 via the A / D conversion unit 3. The CPU 7 reads the captured image directly from the buffer memory 4.

In step S102, the CPU 7 extracts two edge amounts from the captured image. The CPU 7 uses the two Gaussian blur filters P (r) ∝exp (−r ² / 2σ ² ) having different parameters σ indicating the frequency characteristics to generate a blurred image by performing smoothing processing on each of the images. To do. The CPU 7 generates two edge images indicating the distribution of edge amounts by taking the difference between the captured image and each blurred image. The CPU 7 holds two edge images in the storage unit 5.

  The values of the two parameters σ can be arbitrarily selected. It is preferable to select values close to each other. The reason for this is that, even if the parameter σ of the blur filter P is slightly changed, the value of the edge amount in the region where the image is not in focus hardly changes. On the other hand, in the region where the image is in focus, when the parameter σ of the blur filter P is slightly changed, the value of the edge amount changes greatly. In other words, the focus amount representing the displacement of the edge amount is a difference between two edge images obtained in step S104, which will be described later, in the in-focus area, which is larger than the non-focus area, and the focus level is measured. Becomes easier. Therefore, in the first embodiment, σ = 0.5 and 0.6.

  In step S103, the CPU 7 divides each edge image obtained in step S102 into block areas each having a size of 30 pixels × 30 pixels, for example. For each block region of each edge image, the CPU 7 calculates an average value of the edge amounts obtained by integrating the edge amounts of all the pixels included in the block region and dividing the sum by the total number of pixels. Thereby, the feature of each block area can be expressed by one value.

  In step S104, the CPU 7 divides, for each block area, the difference between the average values of the two edge amounts obtained in step S103 by the average value of the edge amounts obtained with σ = 0.5 in the same block area. The CPU 7 holds the calculated value in the storage unit 5 as the focusing amount in each block area. In the first embodiment, the focus amount is a value obtained by dividing the difference between the average values of the two edge amounts by the average value of the edge amounts obtained with σ = 0.5 of the same block region. An average value of edge amounts obtained with σ = 0.6 in the same block region or a value obtained by dividing them by an average value may be used.

  In step S105, the CPU 7 calculates an average value of the focusing amount in the autofocus selection area 20 by using the focusing amount of the block area included in the autofocus selection area 20 among the focusing amounts calculated in step S104. . The CPU 7 holds the obtained average value in the storage unit 5 as a reference value for the focusing amount.

  FIG. 4 shows the distribution of positive values three-dimensionally by performing a difference of (focus amount−reference value) for each block region. By using the focus amount in the first embodiment as a value obtained by dividing the difference between the average values of the two edge amounts by the average value of the edge amounts obtained with σ = 0.5 of the same block region, the luminance difference The values are evenly distributed not only in the vicinity of the soccer goal, but also on the grass under the low-contrast goal where you want to focus. In other areas where the focus is not in focus, as described above, the focus amount is smaller than the reference value, and thus is not shown. In the plane where the focus amount = reference value in FIG. 4, an area corresponding to the autofocus selection area 20 in FIG.

  In step S <b> 106, the CPU 7 obtains an integrated value obtained by integrating only positive values among the differences of (focus amount−reference value) in each block area in the autofocus selection area 20.

  In step S107, the CPU 7 uses the integrated value obtained in step S106 as an evaluation value for in-focus level determination, and determines whether or not the focus is achieved from a comparison with the previously obtained evaluation value. In this case, the reference value used in step S106 may be the same value for each image. The higher the degree of focus, the greater the amount of focus. Therefore, when the integrated value obtained in step S106 is maximized in the target area, that is, the current evaluation value and the previous evaluation value match. When this is done, it means that the imaging lens 1 is in the optimum position for focusing. If the current evaluation value matches the previous evaluation value, the CPU 7 determines that the imaging lens 1 is at the optimum position for focusing. In order to notify the user that the focus has been achieved, the CPU 7 displays, for example, on the monitor 9 using symbols or characters. Then, the CPU 7 ends the focus measurement. On the other hand, if they do not match, the process proceeds to step S108 (NO side). In the first embodiment, immediately after the start of measurement, since there is no previous evaluation value, step S107 is skipped and the process proceeds directly to step S108.

  In step S108, the CPU 7 moves the position of the imaging lens 1 through the lens driving unit 6 so that the evaluation value in the autofocus selection area 20 is maximized. Specifically, if the current evaluation value> the previous evaluation value, the imaging lens 1 is moved in the same direction as the previous time. If the current evaluation value <the previous evaluation value, the reverse of the previous value. The imaging lens 1 is moved in the direction. The amount by which the lens driving unit 6 moves the imaging lens 1 is determined, for example, according to the difference between the current evaluation value and the previous evaluation value. Then, the process proceeds to step S101, and the next image is captured and in-focus measurement is performed. Steps S101 to S108 are performed until the current evaluation value = the previous evaluation value. When the evaluation value becomes the maximum, the CPU 7 determines that the imaging lens 1 is in an optimum position in focus and ends the focus measurement.

As described above, in the first embodiment, by using two blur filters P having different parameters σ indicating frequency characteristics for an image, two edge images are generated, and the focusing amount is determined by the difference between them. By calculating, even when there is a large luminance difference, it is possible to accurately measure the in-focus level for imaging of the subject.
<< Second Embodiment >>
FIG. 5 is a configuration diagram of the focus measuring apparatus 30 according to the second embodiment of the present invention. The focus measurement device 30 according to the second embodiment includes a CPU 31, a memory 32, a storage device 33, an image input unit 34, and an input / output interface 36. The CPU 31, the memory 32, the storage device 33, the image input unit 34 and the input / output interface 36 are connected via a bus 35 so that information can be transmitted. Further, an input device 37 and an output device 38 are connected via the input / output interface 36.

  The CPU 31 receives a start command and initial setting from the input device 37 by the user via the input / output interface 36 and executes a focus measurement program held in the memory 32. The CPU 31 appropriately reads image data stored in the storage device 33 or the image input unit 34 and performs in-focus measurement. The result of this focus measurement is stored in the storage device 33 and is simultaneously displayed on the output device 38 via the input / output interface 36. As the CPU 31, a CPU of a general computer can be used.

  The memory 32 is used for holding a focus measurement program and temporarily storing processing results during the focus measurement performed by the CPU 31 and various data from each component. As the memory 32, a general semiconductor memory can be used.

  The storage device 33 stores and holds image data to be processed by the CPU 31. Image data is held in various formats such as a bitmap format, a jpeg format, and a tiff format. The data held in the storage device 33 can be referred to as appropriate from the CPU 31 via the bus 35. As the storage device 33, a general hard disk device, a magneto-optical disk device, or a storage device such as a removable memory card can be selected and used.

  The image input unit 34 is used when inputting image data to be processed. The image input unit 34 is provided with an insertion port such as a USB cable or a memory card. Then, an image is input to the focus measurement device 30 from a device (a scanner, a digital camera, a digital video, or the like) connected by a USB cable or the like, or a memory card inserted in an insertion slot. The image data input from the image input unit 34 may be transferred and held in the mass storage device 33 via the bus 35 or may be directly read out by the CPU 31.

  The input / output interface 36 is used to transmit a start command and initial setting from the input device 37 to the CPU 31, and conversely, send processing results from the CPU 31 to the output device 38.

  The input device 37 is used for an instruction to start execution of an image processing program by the user, an initial setting operation, and the like. As the input device 37, a general keyboard, mouse, or the like can be used.

  The output device 38 displays the progress and the result of the process sent from the CPU 31 as information that can be confirmed by the user. As the output device 38, a monitor, a printer, or the like can be used.

  Hereinafter, the focus measuring apparatus 30 according to the second embodiment will be described with reference to the flowchart of FIG.

  Specifically, using the input device 37, the user inputs a command for the focus measurement program together with initial settings and issues a start command. The CPU 31 of the focus measurement device 30 receives the command through the input / output interface 36 and executes the focus determination program stored in the memory 32. As a result, the processing from step S201 to step S206 in FIG. 6 is performed. In the following description, it is assumed that a plurality of image data are stored in advance in the storage device 33 of the focus measurement device 30.

  In step S <b> 201, the CPU 31 reads one image from the storage device 33 and holds it in the memory 32.

  In step S202, the CPU 31 extracts two edge amounts from the image. The CPU 31 performs a smoothing process on each image using two Gaussian blur filters P having different parameters σ indicating frequency characteristics to generate a blurred image. CPU31 produces | generates two edge images which show distribution of edge amount by taking the difference of an image and each blur image. The CPU 31 holds two edge images in the memory 32. Note that, as in the first embodiment, σ = 0.5 and 0.6.

  In step S203, the CPU 31 divides each edge image obtained in step S202 into block areas each having a size of 30 pixels × 30 pixels, for example. For each block region of each edge image, the CPU 31 calculates an average value of edge amounts obtained by adding up the edge amounts of all the pixels included in the block region and dividing the sum by the total number of pixels.

  In step S204, for each block region, the CPU 31 divides the difference between the average values of the two edge amounts obtained in step S203 by the average value of the edge amounts obtained with σ = 0.5 in the same block region. The CPU 31 holds the calculated value in the memory 32 as the focusing amount in each block area.

  In step S205, the CPU 31 obtains an average value of the focus amount in the entire image using the focus amount obtained in step S204. The CPU 31 holds the obtained average value in the memory 32 as a focus amount reference value.

In step S <b> 206, the CPU 31 outputs the image to the output device 38 via the input / output interface 36. At the same time, the CPU 31 illustrates, on the image output to the output device 38, a focusing amount area that is a value larger than the reference value obtained in step S205 as a focused area by a line or the like. The focus determination ends. FIG. 7 shows, in a broken line, an area with a focusing amount larger than the reference value based on the distribution of FIG. 4 when the image read in step S201 is the soccer goal of FIG.
<< Third Embodiment >>
FIG. 8 is a flowchart of processing in the focus determination apparatus according to the third embodiment of the present invention.

  The focus measurement apparatus according to the third embodiment is basically the same as the focus measurement apparatus 30 according to the second embodiment in FIG. 5, and description of the operation of each component is omitted. The third embodiment is different from the second embodiment in that an image with the highest focus level is based on the reference value of one reference image among a plurality of images taken by continuous shooting or the like. Is to choose.

  Specifically, using the input device 37, the user inputs a command for the focus measurement program together with initial settings and issues a start command. The CPU 31 of the focus measurement device 30 receives the command through the input / output interface 36 and executes the focus determination program stored in the memory 32. As a result, the processing from step S301 to step S309 in FIG. 8 is performed. In the following description, it is assumed that a plurality of images obtained by continuously shooting the subject in FIG. 3 are stored in the storage device 33 of the focus measurement device 30 in advance.

  In step S <b> 301, the CPU 31 reads one image from the storage device 33 and holds it in the memory 32. In the third embodiment, the first image is the reference image.

  In step S302, the CPU 31 extracts two edge amounts from the image. The CPU 31 performs a smoothing process on each image using two Gaussian blur filters P having different parameters σ indicating frequency characteristics to generate a blurred image. The CPU 31 generates two edge images indicating the distribution of edge amounts by taking the difference between the captured image and each blurred image. The CPU 31 holds two edge images in the memory 32. Note that, as in the second embodiment, σ = 0.5 and 0.6.

  In step S303, the CPU 31 divides each edge image obtained in step S302 into block areas having a size of 30 pixels × 30 pixels, for example. For each block area of each edge image, the CPU 31 calculates the average value of the edge amounts by adding up the edge amounts of all the pixels included in the block region and dividing the result by the total number of pixels.

  In step S304, for each block region, the CPU 31 divides the difference between the average values of the two edge amounts obtained in step S303 by the average value of the edge amounts obtained with σ = 0.5 for the same block region. The CPU 31 holds the calculated value in the memory 32 as the focusing amount in each block area.

  In step S305, the CPU 31 determines whether or not the processing target image is a reference image. If it is a reference image, the process proceeds to step S306 (YES side). If it is not a reference image, the process proceeds to step S307 (NO side).

  In step S306, the CPU 31 uses the focus amount of the block area included in the autofocus selection area 20 out of the focus amount of the reference image obtained in step S304, and averages the focus amount in the autofocus selection area 20. Find the value. The CPU 31 holds the obtained average value in the memory 32 as a reference value for the focusing amount in the reference image.

  In step S <b> 307, the CPU 31 obtains an integrated value obtained by integrating only positive values among the differences of (focus amount−reference value) in each block area in the autofocus selection area 20. The CPU 31 holds the integrated value in the memory 32.

  In step S308, if the CPU 31 has processed all the images captured by continuous shooting, the process proceeds to step S309. If an unprocessed image still exists, the process proceeds to step S301 (YES side). The CPU 31 performs the processing from step S301 to step S308 on all images obtained by continuous shooting using the reference value of the reference image obtained in step S307, obtains the integrated value of each image, and holds it in the memory 32. .

In step S309, the CPU 31 reads from the memory 32 the integrated value of all images obtained by continuous shooting obtained in step S307. The CPU 31 outputs the image that gives the largest integrated value as the most focused image to the output device 38 via the input / output interface 36 and ends the focus measurement.
≪Supplementary items regarding the embodiment≫
In the first embodiment to the third embodiment, the edge amount is a value obtained from the difference between the image and each blurred image, but the present invention is not limited to this. For example, the edge amount can be a value obtained from the absolute value of the difference between the image and each blurred image.

  In the first to third embodiments, the in-focus amount is the difference between the average values of the two edge amounts in each block region, and the edge amount obtained by σ = 0.5 in the same block region. Although the value is divided by the average value, the present invention is not limited to this. For example, the difference between the average values of the two edge amounts may be the average value of the edge amounts obtained by σ = 0.6 in the same block area, or a value obtained by dividing them by the average value. May be used as the focusing amount. FIG. 9 shows, for example, the case where the difference between the average value of the two edge amounts based on the image of FIG. 3 is used as the in-focus amount in step S105 of the first embodiment, and the difference between the difference from the reference value is shown in FIG. FIG. 4 shows the distribution of the focusing amount that is a positive value in the same manner as in FIG. Note that an area corresponding to the autofocus selection area 20 in FIG. 3 is indicated as 22 on the plane where the focusing amount = reference value in FIG.

  In the first to third embodiments, the size of each block area is 30 pixels × 30 pixels, but the present invention is not limited to this. The size of the block area is suitably selected according to the size of the image to be processed, the processing speed, the size of the memory area, the accuracy of focus measurement, and the like.

  In the first to third embodiments, the average value of the edge amounts obtained from the respective edge images is used in each block region in order to obtain the focusing amount. It is not limited. In each block region, a dispersion value of the edge amount obtained from each edge image may be used.

  In the first to third embodiments, the focus amount reference value is a value obtained from a read image or a designated image among a plurality of images obtained by continuous shooting. However, the present invention is not limited to this. For example, a value directly input by the user can be used as the reference value.

  In the first embodiment to the third embodiment, the parameter σ of the blur filter P is set to 0.5 and 0.6, but the present invention is not limited to this. In addition to the accuracy of focus measurement and the like, as the ISO sensitivity of the image sensor increases, spike-like noise from the pixels of the image sensor is added to the image. It is preferable to appropriately select the parameter σ of the blur filter P according to the sensitivity.

  In the first embodiment or the third embodiment, the autofocus selection area 20 is set to an area near the center of the field of view to be imaged, but the present invention is not limited to this. The autofocus selection area 20 may be set in an area outside the center of the visual field. Alternatively, the autofocus selection area 20 may be set as a plurality of areas, and all the areas may be used for in-focus measurement, or some areas may be appropriately selected and used for in-focus measurement.

  In the first embodiment or the third embodiment, the reference value is obtained using the focus amount of the block area in the autofocus selection area 20, but the present invention is not limited to this. The reference value can be obtained by using the focus amounts of all the block areas of the image.

  In the second embodiment, the reference value is obtained using the focus amounts of all the block areas, but the present invention is not limited to this. For example, the reference value can be obtained by using the focusing amount of the block area included in the autofocus selection area or an arbitrary size area.

  In the third embodiment, the first image is the reference image, but the present invention is not limited to this. The reference image can be arbitrarily determined from a plurality of continuously shot images.

  It should be noted that the present invention is also applicable to having a program for realizing each step in the focus measurement method according to the present invention and causing a computer to function as a focus measurement device that measures the focus level of an image.

  Note that the present invention is also applicable to a recording medium that stores a computer program for realizing each step in the focus measurement method according to the present invention.

  It should be noted that the present invention can be implemented in various other forms without departing from the spirit or main features thereof. For this reason, the above-described embodiment is merely an example in all respects and should not be construed in a limited manner. The present invention is shown by the scope of the claims, and the present invention is not limited to the text of the specification. Further, all modifications and changes belonging to the equivalent scope of the claims are within the scope of the present invention.

  As described above, the present invention is a technique that can be applied to an image processing apparatus for an image such as an electronic camera or a video camera, for which accurate calculation and determination of an in-focus level is required.

The figure which shows the structure of the electronic camera which concerns on this 1st Embodiment. The flowchart which shows the procedure of the focus measurement which concerns on this 1st Embodiment. The figure which shows that the soccer goal and turf are arrange | positioned in the imaging | photography visual field in the 1st Embodiment The figure which shows three-dimensionally the distribution of the difference which becomes a positive value among the difference with respect to the reference value of the amount of focus among each block area | region of the image which imaged FIG. The figure which shows the structure of the focus determination apparatus which concerns on the 2nd embodiment. The flowchart which shows the procedure of the focus measurement which concerns on the 2nd embodiment. The figure which shows the area | region of the focusing amount of a larger value than a reference value with a line in the imaged image of FIG. The flowchart which shows the procedure of the focus measurement which concerns on the 3rd embodiment. The figure which shows three-dimensionally the distribution of the difference which becomes a positive value among the differences with respect to the reference value of the amount of focus newly defined among each block area | region of the image which imaged FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Imaging lens, 2 Imaging device, 3 A / D conversion part, 4 Buffer memory, 5 Storage part, 6 Lens drive part, 7 CPU, 8 Operation member, 9 Monitor, 10 Electronic camera, 11 Bus

Claims (10)

An image input unit for acquiring a captured image;
A filter unit that generates a blurred image using a pair of filters each having a different frequency characteristic with respect to the image;
An extraction unit that extracts a first feature amount and a second feature amount from the image and each of the blurred images at a plurality of locations of the image;
A calculation unit that calculates a focus amount indicating a degree of focus from the first feature amount and the second feature amount;
A measurement unit that measures a focus level at a plurality of locations of the image based on the calculated focus amount;
An in-focus measurement device comprising: The in-focus measurement device according to claim 1,
The computing unit is
The difference between the first feature value and the second feature value, or the difference between the first feature value, the second feature value, or the first feature value and the second feature value. A value obtained by dividing by an average value is used as the focusing amount. In the focus measuring apparatus according to claim 1 or 2,
The in-focus measuring apparatus, wherein the first feature amount and the second feature amount are edge amounts obtained from a difference between the image and each blurred image or an absolute value of the difference. In the focus measurement apparatus according to any one of claims 1 to 3,
The computing unit is
A reference value calculation unit that calculates a reference value of a focus level based on the focus amount calculated at each of the locations in the predetermined region of the image among the plurality of locations of the image. ,
The measuring unit is
A focus measurement apparatus, further comprising: a reference value measurement unit that measures a focus level at a plurality of locations of the image based on a value of the focus amount with respect to the reference value. The in-focus measurement device according to claim 4,
The measuring unit is
A focus measurement apparatus, further comprising a reference value storage unit that holds the focus level of another image captured by the same imaging apparatus in order to use the reference value. The in-focus measurement device according to claim 4,
The in-focus measuring apparatus, wherein the predetermined area is the entire area of the image. The in-focus measurement device according to claim 4,
The in-focus measuring apparatus, wherein the predetermined area is an autofocus selection area during photographing. In the focus measurement apparatus according to any one of claims 1 to 7,
The filter unit is
A focus measurement apparatus, further comprising: a frequency characteristic adjustment unit that changes the frequency characteristic of the filter according to an ISO sensitivity of the image sensor. Generating a blurred image using a pair of filters each having a different frequency characteristic with respect to the captured image;
Extracting a first feature amount and a second feature amount from the image and each of the blurred images at a plurality of locations of the image, respectively;
Calculating an in-focus amount indicating a degree of focus from the first feature amount and the second feature amount;
Measuring focus levels at a plurality of locations of the image based on the calculated focus amount;
A focus measurement method comprising: A program for causing a computer to function as the focus measurement device according to any one of claims 1 to 8.

Images