JP6651741B2 - Imaging device - Google Patents

Description

  The present invention relates to an imaging device such as a camera, and more particularly, to a process for correcting a decrease in light amount around an image caused by lens characteristics.

  In a digital camera or the like, the amount of light decreases in the peripheral portion of a captured image due to the characteristics of a shooting lens. Therefore, shading correction (hereinafter also referred to as “peripheral light amount reduction correction”) is performed on the obtained image signal, and gain correction is performed on pixels in the light amount reduction portion.

  The degree of reduction in the peripheral light amount differs depending not only on the optical characteristics of the photographic lens used but also on the photographic conditions. Therefore, the image is divided into a plurality of blocks, and a series of correction coefficients different depending on the aperture value and the zoom magnification are stored in the memory for each block. Then, a shading correction is performed by selecting a correction coefficient according to an aperture value at the time of shooting, a zoom magnification, and the like (see Patent Document 1).

  In some shooting scenes, performing shading correction may increase the luminance level more than necessary and increase noise. In order to prevent this, it is determined whether or not the shooting scene is such that the decrease in the peripheral light amount is not noticeable, and in that case, the gain is suppressed. For example, the captured image is divided into a plurality of blocks, and the luminance value and color information are detected for each block, and the blue region is the subject in the blue region according to the sky where the peripheral light amount is conspicuously reduced. And shading correction is performed. When a green area such as a tree in which the decrease in peripheral light amount is inconspicuous is detected, no shading correction is performed (see Patent Documents 2 and 3).

JP 2003-110936 A JP 2006-148791 A JP 2007-27943 A

  The composition of the subject at the time of shooting is various, and there are cases where the shooting scene cannot be properly determined only by the exposure conditions and color information. If the peripheral light amount reduction correction that is inappropriate for the shooting scene is performed, the image quality of the peripheral portion of the shooting image is deteriorated.

  Therefore, it is required to perform appropriate peripheral light amount reduction correction for various shooting scenes.

  An imaging apparatus according to an aspect of the invention includes a detection unit that detects a luminance difference and a variation in luminance from a center side of an image to a peripheral direction with respect to a plurality of image regions of a captured image, based on a correction coefficient determined according to characteristics of an imaging optical system. And a correction unit that corrects a reduction in the amount of peripheral light in a captured image. Regarding a plurality of image regions, a plurality of image regions including the peripheral portion of the captured image may be set so that information on a decrease in the amount of peripheral light is easily obtained. For example, it is possible to define a plurality of image areas along a diagonal direction of a captured image. Further, an area setting unit for setting a plurality of image areas in the captured image may be provided.

  Then, the correction unit determines whether or not to execute the correction (hereinafter, referred to as “execution / non-execution”) according to at least one of the detected luminance variation and the luminance difference, and determines a correction coefficient. Make a uniform adjustment. Alternatively, only the execution / non-execution of the correction may be determined, or only the adjustment of the correction coefficient may be performed. For example, the correction unit adjusts the correction coefficient when the degree of luminance variation is smaller than the first threshold and the luminance difference is larger than the second threshold.

  The luminance characteristic detecting unit can detect a luminance difference and a luminance variation for a plurality of image areas. For example, it is possible to detect four image areas defined according to the four corners of the image, a peripheral area of the captured image other than the four corners, or an area closer to the center than the peripheral part. The correction unit may adjust the correction coefficient according to the number of image areas in which the degree of luminance variation is smaller than the first threshold and the luminance difference is larger than the second threshold.

  For example, when the correction unit has a plurality of image areas in which the degree of luminance variation is smaller than the first threshold value and the luminance difference is larger than the second threshold value, the correction unit can adjust the correction coefficient. Further, the correction unit sets the correction coefficient when the number of image areas in which the degree of luminance variation is smaller than the first threshold value and the luminance difference is larger than the second threshold value is plural and the image areas are adjacent to each other. It may be adjusted. In this case, the detection unit can detect a variation in luminance for an integrated area obtained by combining adjacent image areas. The correction unit may adjust the correction coefficient according to the degree of variation in the brightness of the integrated area.

  On the other hand, the correction unit determines whether the number of image areas in which the luminance variation is smaller than the first threshold and the luminance difference is larger than the second threshold is equal to or less than a predetermined number (for example, 1), or in which the luminance variation is smaller. Even when the number of image areas smaller than the first threshold value and the luminance difference is larger than the second threshold value is plural and the image areas are not adjacent to each other, it is possible not to perform the peripheral light amount reduction correction. .

  According to another aspect of the present invention, there is provided a program that causes an imaging device to detect a luminance difference and a luminance variation from an image center side to a peripheral direction with respect to an image area of a captured image, and a characteristic according to characteristics of the imaging optical system. Based on the determined correction coefficient, the correction unit is made to function as a correction unit that corrects a decrease in peripheral light amount of a captured image, and determination of execution / non-execution of correction and / or execution of correction is performed in accordance with at least one of the detected luminance variation and luminance difference. Alternatively, it is made to function as correction means so as to perform uniform adjustment of the correction coefficient.

  Further, an image processing method according to another aspect of the present invention detects a luminance difference and a luminance variation from the image center side to a peripheral direction with respect to a plurality of image areas of a captured image, and determines the luminance difference according to characteristics of the imaging optical system. A method of performing a peripheral light amount reduction correction of a captured image based on a correction coefficient to be performed, and determining and / or performing correction execution / non-execution in accordance with at least one of a detected luminance variation and a luminance difference. Performs uniform adjustment of coefficients.

  According to the present invention, it is possible to perform peripheral light amount reduction correction suitable for a shooting scene.

It is a block diagram of a digital camera which is this embodiment. FIG. 4 is a diagram illustrating an image area defined for a luminance characteristic analysis with respect to a captured image. FIG. 4 is a diagram illustrating an image area along a diagonal direction of a captured image. It is a flowchart of a shading correction process. 9 is a flowchart of a process 1 for adjusting a correction coefficient. 9 is a flowchart of a process 2 for adjusting a correction coefficient. FIG. 11 is a diagram illustrating a luminance analysis target area of a captured image according to the second embodiment. 9 is a flowchart illustrating a correction process according to the second embodiment. 9 is a flowchart of a correction coefficient adjustment process according to the second embodiment. FIG. 9 is a diagram illustrating a process when adjusting a correction coefficient of a pixel other than a peripheral region.

  Hereinafter, a digital camera according to the present embodiment will be described with reference to the drawings.

  FIG. 1 is a block diagram of a digital camera according to the present embodiment.

  Here, the digital camera 100 is configured as a mirrorless digital camera, and shoots, reproduces a recorded image, reproduces a recorded image according to a user's input operation on an operation member 104 including a mode dial, a release button, a cross button, an execution button, and the like. Mode setting and the like are performed. A lens barrel 102 for accommodating a photographing lens 107 is detachably mounted on the front side of the camera, and an LCD 106 is provided on the rear side of the camera.

  A signal processing unit 101 including a CPU, a RAM, and the like performs operation control of the entire camera, such as exposure control, recording operation, and reproduction display operation, in response to a button operation on an operation member 104 such as a release button. The image processing apparatus further includes an image processing unit 132 and a correction unit 134 that execute image signal processing and the like. The camera operation control program is stored in a recording medium such as a ROM (not shown).

  When light from a subject enters the imaging lens 107, the image is formed on an image sensor 111 such as a CCD or a CMOS sensor, whereby a subject image is formed on the image sensor 111. On the light receiving surface of the image sensor 111, a color filter array (not shown) of R, G, B, etc. is arranged.

  In the case of a through image display, pixel signals for one field or one frame are output from the image sensor 111 at predetermined time intervals (for example, 1/60 seconds or 1/30 seconds) in accordance with a drive signal from a timing generator (TG) 120. Is read. The read pixel signal is subjected to amplification processing, digitization, and the like in the AFE circuit 130, and then sent to the image processing unit 132 of the signal processing unit 101.

  In the image signal processing unit 132, color interpolation processing, color conversion processing, white balance adjustment processing, and the like are performed on the pixel signals. Thus, the pixel values of R, G, and B are obtained for each pixel, and a color image signal is generated. The generated color image signal is temporarily stored in the image memory 103. The signal processing unit 101 drives the LCD 106 based on the color image signal, whereby a real-time moving image is displayed.

  When the release button is half-pressed, a contrast AF process is executed based on a pixel signal read from the image sensor 111. The signal processing unit 101 controls the lens driving unit 136 to drive the photographing lens 107, and detects an in-focus position where the contrast value reaches a peak. In addition to the contrast AF processing, the brightness of the subject image is detected based on the pixel signal read from the image sensor 111, and the exposure value is calculated.

  When the release button is fully pressed, the exposure control unit 138 drives a shutter (not shown) and a diaphragm (not shown) based on the exposure value to control the exposure. As a result, pixel signals for one frame are read from the image sensor 111.

  The image signal processing unit 132 generates still image data based on the read pixel signals for one frame. The still image data is recorded on the recording medium 105 such as a memory card in a compressed state or uncompressed. When the reproduction mode is set, the selected recorded image is displayed on the LCD 106.

  The correction unit 134 performs a shading correction process on the generated R, G, B color image signals. That is, the gain is corrected so as to compensate for a decrease in the amount of light in the peripheral portion of the captured image due to the lens characteristics and the like of the mounted imaging lens 107. However, the R, G, and B color image signals may be generated after the shading correction. The lens memory 108 of the lens barrel 102 stores data relating to lens design values (or measurement values, adjustment values, etc.), and a correction coefficient corresponding to a gain value is calculated based on the data and exposure conditions. Then, the pixel value is multiplied by the correction coefficient.

  In the present embodiment, shading correction is performed according to the shooting scene based on the luminance characteristics of the peripheral portion of the shot image. Specifically, execution / non-execution of shading correction is determined according to the variation in luminance and the degree of luminance reduction in the peripheral portion of the captured image, and the correction coefficient is adjusted / corrected according to the captured scene. Hereinafter, this will be described in detail with reference to FIGS.

  FIG. 2 is a diagram illustrating an image area defined for a luminance characteristic analysis with respect to a captured image. FIG. 3 is a diagram illustrating an image area along a diagonal direction of a captured image.

  FIG. 2 shows a captured image PI including M × N pixels P arranged in a matrix. In the captured image PI, four areas (hereinafter, referred to as corner direction areas) C1 to C4 are defined along diagonal areas S1 and S2 along the diagonal direction from the center C. FIG. 3 shows an example.

  The pixels constituting the corner direction area C1 including the upper right corner of the captured image correspond to an area excluding the center area C0 in the diagonal area S2 along the direction of the upper right end E1 from the center C of the captured image PI. A peripheral area c1 near the upper right corner of the captured image is defined in the corner direction area C1, and an area near the image center C side (hereinafter, referred to as a center side area) CA1 is further included in the peripheral area c1. An area near the upper right end E1 (hereinafter, referred to as an end area) CB1 is defined. Here, the number of pixels forming the center side area CA1 and the end side area CB1 is substantially equal.

  Similarly to the corner direction area C1, the peripheral areas c2 to c4 are defined for the other corner direction areas C2 to C4, and the peripheral areas c2 to c4 are the center side areas CA2, CA3, CA4 and the end side areas CB2, CB3, CB4. Then, peripheral areas c1 to c4 corresponding to the four corners are determined as luminance analysis target areas.

  The decrease in the amount of light around the image, particularly around the four corners of the image, due to lens characteristics or the like may be noticeable or inconspicuous depending on the shooting scene. For example, when a subject such as a person or a building exists around the field of view (frame), the decrease in light amount is inconspicuous. On the other hand, when photographing a blue sky or the like, a decrease in the amount of light around the image is conspicuous.

  Further, depending on the composition, the degree of decrease in the peripheral light amount is not uniform at the four corners around the image. For example, in a case where the lower half of the captured image is a building such as a building and the upper half is a blue sky, a portion where the amount of light decreases is conspicuous and a portion where the light amount is not conspicuous are also mixed around the image.

  Therefore, by analyzing the luminance characteristics of each of the peripheral regions c1 to c4, the degree of the decrease in the peripheral light amount is detected. Specifically, when the variation in the luminance is large in the peripheral regions c1 to c4, or when the degree of the luminance decrease from the image center side to the image end (peripheral) side is small, the decrease in the light amount becomes inconspicuous. Based on the above, the degree of luminance variation and the luminance difference are detected for each peripheral area.

  Then, it is determined whether or not to perform the shading correction according to the luminance variation and the degree of the luminance difference in each of the peripheral areas c1 to c4. When the shading correction is performed, the correction coefficient is adjusted in consideration of the variation in the entire peripheral areas c1 to c4 and the degree of luminance reduction.

  FIG. 4 is a flowchart of the shading correction process.

  In step S1, the luminance value Lv of each pixel belonging to the diagonal areas S1 and S2 is calculated. There are various methods of calculating the luminance value. For example, it is possible to calculate the luminance and color difference signals based on the R, G, and B image signals and obtain the luminance value of each pixel. Further, it is also possible to obtain a pixel value of G corresponding to the luminance level as a luminance value.

  In step S2, a standard deviation sd is obtained for each of the peripheral regions c1 to c4 based on the luminance value Lv in that region. In step S3, the standard deviation obtained by integrating the peripheral regions (c1, c2), (c2, c3), (c3, c4), and (c4, c1) adjacent to each other in the positional relationship of the four corners (hereinafter, referred to as an integrated standard deviation) tsd is calculated. This is used when determining the degree of correction of a correction coefficient described later.

  In step S4, a luminance difference ΔL between the center side area CA1 and the end side area CB1 with respect to the peripheral areas c1 to c4 is obtained. Here, the luminance average value of each of the center side area CA1 and the end side area CB1 is obtained, and the difference (absolute value) is calculated as the luminance decrease amount from the center side to the peripheral side. However, a luminance difference between the center side area CA1 and the end side area CB1 (such as a luminance difference between arbitrary pixels) may be obtained by other methods.

  In step S5, the standard deviation sd is lower than a threshold value (first threshold value) TA, and the luminance difference ΔL is directed to the peripheral direction with respect to the standard deviation sd and the luminance difference ΔL calculated for the peripheral regions c1 to c4. It is determined whether the magnitude is larger than a threshold (second threshold) TB. When the standard deviation sd is lower than the threshold value TA and the luminance difference ΔL is higher than the threshold value TB, there is little variation in luminance and a decrease in luminance occurs. The threshold value TA and the threshold value TB are stored in a memory or the like in advance, and the values can be set according to various conditions.

  When there are three or four peripheral areas where the standard deviation sd is lower than the threshold TA and the luminance difference ΔL is larger than the threshold TB, in step S6, the correction coefficient may be adjusted according to each standard deviation and the luminance difference. It is determined (hereinafter, the process performed in this case is referred to as “process 1”). However, when the correction ratio application rate becomes 100% as described later, the correction coefficient is used as it is and is not corrected. On the other hand, when there are two peripheral regions where the standard deviation sd is lower than the threshold value TA and the luminance difference ΔL is larger than the threshold value TB, it is determined whether or not the peripheral regions are adjacent to each other (S7).

  If the peripheral regions are adjacent to each other, that is, if the two peripheral regions are any combination of (c1, c2), (c2, c3), (c3, c4), and (c4, c1), in step S8, It is determined that the correction coefficient is adjusted according to each standard deviation and the luminance difference (hereinafter, the processing performed in this case is referred to as “processing 2”).

  On the other hand, if the standard deviation sd is lower than the threshold value TA and the peripheral difference ΔL is larger than the threshold value TB, or if there is only one peripheral region among the four peripheral regions c1 to c4, the peripheral region is viewed as a whole of the captured image. Since it can be regarded as a photographing scene in which the decrease in light amount is inconspicuous, it is determined not to perform the shading correction (S9). Also, it is determined not to perform the shading correction even in the case where the areas are not adjacent to each other in the corner direction (S9).

  FIG. 5 is a flowchart of processing 1 for adjusting the correction coefficient. Here, correction processing of a correction coefficient in a case where there are three or four peripheral areas in which a luminance variation and a luminance difference have occurred is shown.

When the average value Asd of the standard deviation is calculated in Step S11, a provisional value (here, referred to as a provisional correction ratio application rate) for correcting the correction coefficient of the shading correction is calculated in Step S12. The provisional correction ratio application rate is obtained by the following equation. Here, Sqrtmax indicates the maximum standard deviation, Sqrtmin indicates the minimum standard deviation, and Ratioref indicates the provisional correction ratio application rate.

Ratioref = (Sqrtmax-Asd) ÷ (Sqrtmax-Sqrtmin)
... (1)

  In step S13, the average value Atsd of the standard deviation (hereinafter, referred to as an integrated standard deviation) of the luminance values for the adjacent corner direction areas is calculated. If there are four target corner direction regions, four integrated standard deviation average values Atsd are calculated. In the case of three, two integrated standard deviation average values Atsd are calculated.

In step S14, the provisional correction ratio application rate Ratioref calculated in step S12 is corrected based on the integrated standard deviation average value Atsd calculated in step S13.

Ratiofair =
Ratioref × (Sqrtmax−Atsd) ÷ (Sqrtmax−Sqrtmin)
... (2)

  The calculation of the correction rate application rate that is actually used from the provisional correction rate application rate by equation (2) is for calculating the correction amount of the correction coefficient suitable for the shooting scene. If the sky and the other half show buildings, etc., the shading correction is appropriate for the four surrounding areas where the sky is projected, but the amount of light decreases in the areas where the buildings are projected Is not noticeable.

  In the present embodiment, the degree of correction of the correction coefficient (correction rate application rate Ratiorefair) is applied uniformly regardless of the position of the peripheral area. For this reason, there is a possibility that the shading correction may be strongly applied to a peripheral area where the light quantity decrease is not conspicuous. Therefore, the provisional correction ratio application ratio is corrected based on the integrated standard deviation average value calculated for the adjacent corner direction regions, and the correction ratio application ratio actually used is obtained (S14).

  For example, when the degrees of the standard deviation of the adjacent corner direction areas are different, that is, when the degree of the variation of the luminance is different, the correction ratio application corrected to a value smaller than the provisional correction ratio application ratio by the equation (2). Rate is obtained. As a result, excessive shading correction is not applied to a peripheral portion of a captured image in which the light amount is not conspicuous.

  In step S15, the value of the correction coefficient according to the lens characteristics is adjusted. Specifically, the correction coefficient is multiplied by the correction ratio, and the correction ratio is multiplied by the correction ratio application ratio. Here, the correction ratio is a value for adjusting the degree to which the correction coefficient is reflected in the pixel value. The correction ratio is set to a distance from the optical axis center so that the gain is excessively applied to the pixel value so as not to be an overgain (1 or more). It is determined accordingly. It is set arbitrarily according to the shooting conditions. When the correction ratio is 100%, the correction coefficient value is used as it is, and as the correction ratio value decreases, the correction coefficient value decreases. When the correction ratio is 0%, the correction coefficient is calculated as a value of 1 and multiplied by the pixel value. That is, no shading correction is applied.

  On the other hand, the correction ratio application rate Raitorefpair calculated by the equation (2) is a correction value for adjusting the value of the correction ratio. By multiplying the correction ratio by the correction ratio application ratio Raitorefpair, the correction coefficient value is consequently obtained. Is corrected (S14). Then, shading correction is performed based on the corrected correction coefficient.

  Various methods can be applied to the shading correction. Here, the correction coefficient data for the pixels spaced at a fixed interval from the center C along the corner direction areas C1 and C2 is calculated from the lens data, and is linearly interpolated along the radial direction and the concentric line around the center C of the captured image. , A correction coefficient is calculated for pixels that are not targeted for lens data. Then, the value of the correction coefficient of each pixel is multiplied by a correction ratio obtained by multiplying the correction ratio application rate, and shading correction is performed by the corrected correction coefficient value. However, it may be obtained by another method such as an interpolation operation other than the linear interpolation. Further, when the correction ratio application rate is uniform on the screen, the correction value is determined according to the distance from the center, so interpolation may be performed based on the correction values calculated from lens data at two pixel positions sandwiching the target pixel.

  FIG. 6 is a flowchart of processing 2 for adjusting the correction coefficient. Here, correction processing of a correction coefficient in a case where two corner-direction areas in which a luminance variation and a luminance difference occur are adjacent to each other is shown.

In step S21, the one having the larger standard deviation sd is selected from two target corner direction areas. In step S22, the provisional correction ratio application rate is calculated by the following equation based on the selected standard deviation sd. Here, Sqrtmax indicates the maximum standard deviation, Sqrtmin indicates the minimum standard deviation, and Ratioref indicates the provisional correction ratio application rate.

Ratioref = (Sqrtmax-sd) ÷ (Sqrtmax-Sqrtmin)

... (3)

Then, based on the integrated standard deviation tsd of the luminance values obtained for the two adjacent corner direction regions, the correction ratio application ratio Ratioorpair is obtained from the provisional correction ratio application ratio Ratioorref (S23, S24).

Ratiofair
= Ratiooref × (Sqrtmax−tsd) ÷ (Sqrtmax−Sqrtmin)
... (4)

  The execution of step S25 is the same as the execution of step S15 in FIG. 5, in which the correction coefficient is multiplied by the correction rate obtained by multiplying the correction coefficient by the correction rate application rate. As a result, the shading correction process is performed based on the calculated correction ratio application rate Ratiorefair.

  In the present embodiment, the correction of the peripheral light amount is performed by multiplying the correction coefficient by the correction ratio and the correction ratio application ratio. However, the correction coefficient value calculated from the lens data without multiplying the correction ratio is used. A configuration in which only the correction ratio application rate is multiplied may be adopted. Further, when the correction ratio application rate obtained by the equation (4) is 1, that is, 100%, only the correction ratio is multiplied by the correction coefficient, and correction of the correction coefficient according to the luminance difference and the variation in the luminance results in Not done.

  As described above, according to the present embodiment, the peripheral regions c1 to c4 corresponding to the four corners of the captured image are defined, and the standard deviation and the luminance difference (luminance reduction amount) from the center to the peripheral direction in the peripheral regions c1 to c4 are determined. To detect. If the value of the standard deviation is equal to or less than the threshold value TA and the luminance difference is equal to or more than the threshold value TB, it is determined that there is no luminance difference and there is a light amount decrease in the peripheral portion of the captured image, that is, it is a scene where the peripheral portion light amount decrease is conspicuous. Perform the correction. Then, in the case of correction, the correction coefficients determined according to the lens characteristics are uniformly adjusted for the peripheral regions c1 to c4 according to the detected luminance difference and the degree of the luminance variation.

  When performing the peripheral light amount reduction correction, a correction coefficient for a discrete image height is calculated due to a limitation of the data capacity of the stored lens data, and the correction coefficient calculated for each pixel by interpolating the correction coefficient is continuously calculated. Multiplied by the pixel of the photographed image having a high luminance distribution. Therefore, when correction is performed in an image region where the decrease in peripheral light amount is not conspicuous, the image quality is likely to deteriorate due to a phenomenon in which the luminance change between pixels is conspicuous.

  However, in the present embodiment, only the photographing scene that needs to be corrected is analyzed by analyzing the variation in luminance and the degree of the luminance difference in the peripheral part of the photographed image in the peripheral area along the diagonal direction where the information is easily obtained. In addition, by suppressing the degree of the correction (gain) based on the luminance analysis, it is possible to execute an appropriate correction according to the shooting scene.

  In particular, by performing luminance analysis on the image areas c1 to c4 at the four corners and determining whether or not to perform correction in accordance with the number of image areas that satisfy the above conditions, various types of subjects are included in the composition. Appropriate peripheral light amount reduction correction can be performed on a scene.

  In addition, by obtaining a standard deviation obtained by integrating adjacent image areas, the degree of correction can be reduced in a shooting scene in which a decrease in light amount is conspicuous only in a part of the periphery of a captured image, while a decrease in light amount is not conspicuous in other peripheral parts. Can be suppressed.

  The manner of defining the peripheral areas c1 to c4 is not limited to the area position and the size as shown in FIGS. Further, the area for calculating the standard deviation and the area for calculating the luminance difference in the peripheral areas c1 to c4 are not limited to the areas as shown in FIGS. For example, areas D1 to D4 around the periphery other than the four corners as shown in FIG. 2 may be set as the luminance analysis target area, and the standard deviation calculation area and the luminance difference calculation area may be separately or partially overlapped. May be determined. As for the luminance difference, the degree of luminance decrease may be detected including the central area such as the image central area D0. In particular, a plurality of luminance analysis target areas that do not include the peripheral portion may be set in order to cope with a shooting scene in which a variation in luminance and a luminance difference occur in other than the peripheral portion.

  Regarding the determination of the execution / non-execution of the correction, the number of the regions satisfying the above condition among the peripheral regions c1 to c4 is three or more, the correction is executed when two are the adjacent regions, and the two are the non-adjacent regions, Correction is not executed when the number is one or less. However, the number of cases is not limited to this, and the number may be appropriately set in a plurality of image areas to divide the cases. Also, for a plurality of image regions, it is possible to arbitrarily set regions other than the four corners so as to include the peripheral portion.

  The variation in luminance may be detected by a value indicating a degree of variation other than the standard deviation. For example, it is also possible to make a determination based on the number of pixels for which the sum of the absolute values of the luminance differences is equal to or greater than a threshold. As for the luminance difference, a value that clarifies the degree of luminance decrease from the center to the periphery may be obtained. In addition, the setting of a threshold value for determining a variation in luminance and a luminance difference may be appropriately set according to the situation. Further, it is also possible to detect either a variation in luminance or a luminance difference, and determine whether to execute correction or not, and adjust the correction coefficient based on the detection.

  In the present embodiment, the execution / non-execution of the correction is determined first, and then the degree of the correction is determined. However, when only the execution / non-execution of the peripheral light amount reduction correction is determined and the correction is executed, the correction coefficient is used. Can be used as is without adjusting. Conversely, for a shooting scene that does not require correction, it is possible to deal with only correction coefficient adjustment, such as setting a correction ratio application rate that does not determine gain execution without determining whether to execute peripheral light reduction correction. It is.

  Next, a digital camera according to a second embodiment will be described with reference to FIGS. In the second embodiment, a captured image is divided into a plurality of blocks, a target region for luminance analysis is determined in block units, and a plurality of partial regions each configured in block units (here, as an example, four corners of the captured image are used). ), The correction coefficient is adjusted independently. Also, it is determined whether or not the correction of the light amount reduction results in a situation where the saturation state and the improvement of the brightness level in the peripheral portion are conspicuous.

  FIG. 7 is a diagram illustrating a luminance analysis target area of a captured image according to the second embodiment.

  The captured image PI is divided into M × N blocks composed of a plurality of pixels, and a luminance value (such as an average luminance value) of each block is calculated. Then, peripheral regions c1 to c4 corresponding to the four corners of the captured image are determined as target regions for luminance analysis. Each of the peripheral areas c1 to c4 is configured by 3 × 3 blocks.

  As in the first embodiment, the standard deviation and the luminance difference along the peripheral direction from the center of the captured image are calculated for the peripheral area c1. The standard deviation is obtained from the luminance value calculated for each block with respect to the entire peripheral area c1. On the other hand, the luminance difference is defined as image areas CA1 and CB2, each of which is composed of 2 × 2 blocks and overlapped by one block. The same applies to the other peripheral areas c2 to c4.

  On the other hand, in the captured image PI, peripheral areas D1 to D4 located between the peripheral areas c1 to c4 are defined. The peripheral regions D1 to D4 are provided to determine whether saturation occurs due to correction for a decrease in peripheral light amount, or whether improvement in the luminance level of the peripheral portion is conspicuous.

  For example, in image signal processing for generating color image data, WB adjustment processing and color conversion processing are performed on R, G, and B pixel values, and coefficients related to image signal processing such as WB gain and color matrix coefficients are calculated. An arithmetic process for multiplying the pixel value is performed. If there is an area having a relatively large brightness level in a peripheral portion of a captured image, multiplying a correction coefficient for shading correction in addition to a series of correction coefficients causes pixel saturation beyond the number of bits.

  On the other hand, when the shooting scene is dark, multiplying the R, G, and B pixel values by the shading correction coefficient amplifies noise in the peripheral portion of the image, and further improves the brightness level in the peripheral portion. Therefore, the maximum luminance value is extracted in the peripheral regions D1 to D4, and it is determined based on the maximum luminance value whether the pixel saturation occurs or the improvement in the luminance level of the peripheral portion is conspicuous. Since pixel saturation is more likely to occur in the four corners c1 to c4, the maximum luminance value may be extracted in the peripheral areas c1 to c4 in the four corners.

  FIG. 8 is a flowchart illustrating a correction process according to the second embodiment.

  After dividing the photographed image (here, M × N), peripheral areas c1 to c4 and D1 to D4 to be analyzed for a decrease in peripheral light amount are determined (steps S31 and S32). When the standard deviation is calculated, it is determined whether the standard deviation is smaller than the threshold value TA (Steps S33, S34). If the standard deviation is smaller than the threshold value TA (the degree of luminance variation is small), the process proceeds to step S35. On the other hand, if the standard deviation is equal to or larger than the threshold value TA (the degree of luminance variation is large), no shading correction is performed.

  In step S35, a brightness difference (brightness reduction amount) between the image center side and the image peripheral side in the peripheral areas c1 to c4 is calculated. Then, it is determined whether or not the luminance difference is larger than the threshold value TB. If the luminance difference is larger than the threshold value TB, the process proceeds to step S37. on the other hand. If the luminance difference is equal to or smaller than the threshold value TB, no shading correction is performed.

  In the case of a decrease in peripheral light quantity (S36), in step S37, a maximum luminance value is extracted for the peripheral areas D1 to D4. Then, in step S38, when the maximum luminance value calculated in the peripheral areas D1 to D4 is multiplied by a coefficient relating to a series of image signal processing such as a WB gain coefficient and a color conversion matrix coefficient and a shading correction coefficient, the pixel value becomes It is determined whether the value exceeds the number, that is, whether or not pixel saturation occurs. If pixel saturation does not occur, the process proceeds to step S39. On the other hand, when pixel saturation occurs, no shading correction is performed. However, the correction coefficient may be adjusted so that saturation does not occur.

  In step S39, when the maximum luminance value is multiplied by a coefficient relating to a series of image signal processing, whether the value is larger than a predetermined threshold value TC, that is, the luminance level of a peripheral portion is corrected by a correction for a dark shooting scene or the like. It is determined whether the improvement becomes noticeable. If the calculated value (the value before being multiplied by the correction coefficient) is larger than the threshold value TC, the process proceeds to step S40. On the other hand, if the calculated value is equal to or smaller than the threshold value TC, no shading correction is performed.

  In step S40, a process of adjusting / correcting the correction coefficient calculated based on the lens data is performed. Hereinafter, the adjustment process of the correction coefficient will be described in detail with reference to FIG.

  FIG. 9 is a flowchart of a correction coefficient adjustment process according to the second embodiment.

In step S41, the correction ratio application rate based on the standard deviation is calculated by the following equation. Here, sd indicates a standard deviation, Sqrtmax indicates a predetermined standard deviation maximum value, Sqrtmin indicates a predetermined standard deviation minimum value, and RatiorefS indicates a correction ratio application rate of the standard deviation. For each of the image regions C1 to C4, the correction ratio application rate RatiorefS of the standard deviation is obtained.

RatiorefS = (Sqrtmax-sd) ÷ (Sqrtmax-Sqrtmin)
... (5)

In step S42, the correction ratio application rate based on the luminance difference is calculated by the following equation. Here, DY is a luminance difference. Ymax indicates a predetermined maximum threshold of the luminance difference, Ymin indicates a predetermined minimum threshold of the luminance difference, and RatiorefY indicates a correction rate application rate of the luminance difference. As in step S41, the correction ratio application rate RatiorefY of the luminance difference is calculated for each of the image areas C1 to C4.

RatiorefY = (DY−Ymin) ÷ (Ymax−Ymin)
... (6)

  In step S43, the correction coefficient calculated based on the lens characteristic data is multiplied by the correction ratio determined for each region, and also multiplied by the correction ratio application ratio calculated by the equations (5) and (6). At this time, the correction rate application rate calculated by the equations (5) and (6) for each image area is multiplied by the corresponding correction coefficient.

  In step S44, the adjustment of the correction coefficient for the pixels in the area between the lines for which the correction values have been calculated is performed along the concentric lines. Specifically, pixels on concentric lines having the same distance from the center of the captured image (optical axis) are linearly interpolated based on the correction correction coefficients of pixels on the same line and adjacent peripheral regions. I do.

  FIG. 10 is a diagram illustrating a process when adjusting the correction coefficient of a pixel in a region between lines for which a correction value has been calculated. Here, the peripheral regions c1 to c4 and the like are not shown.

  As described above, the correction correction coefficient reflecting the correction ratio application rate obtained for each of the peripheral areas c1 to c4 is provided at predetermined distance intervals along the diagonal direction from the optical axis (image center C) to the four corners. The correction coefficient between pixels along the diagonal direction to which no correction coefficient is given is obtained by linear interpolation based on the image height (distance from the center). Desired. When, for example, the peripheral areas D1 to D4 are set instead of the peripheral areas c1 to c4, the correction values along the line from the optical axis to the peripheral areas D1 to D4 serving as the luminance determination area are corrected.

  Then, the target pixel is obtained by linear interpolation using the correction coefficient obtained for the pixels on the concentric line. For example, when the correction correction coefficients of the pixels PK1 and PK2 having the same image height from the center along the upper left direction and the upper right direction are calculated, the pixels PK1 and PK1 having the same image height along the concentric line are displayed. It is obtained by linear interpolation using the correction correction coefficient of PK2.

  When the value of the correction ratio application rate differs depending on the image area, the values of the correction correction coefficients of the pixel PK and the pixel PJ do not match. Therefore, when the correction correction coefficient of each pixel other than the peripheral regions c1 to c4 obtained by the linear interpolation is reflected, a large gain is applied by the shading correction in the image region requiring the shading correction, and the image region in which the decrease in the peripheral light amount is not noticeable. Then, the gain is suppressed.

  As described above, according to the second embodiment, the standard deviation and the center side to the peripheral side are set for the peripheral areas c1 to c4 defined in block units in the captured image divided into M × N blocks. A luminance difference (a luminance reduction amount) is detected. At the same time, the maximum luminance value is detected in the peripheral areas D1 to D4. Then, the execution / non-execution of the correction is determined according to the luminance variation and the luminance difference, and the correction coefficient is adjusted. In addition, the saturation of the pixel is determined based on the maximum luminance value, and the luminance level of the peripheral portion is conspicuously improved depending on whether a calculated value obtained by multiplying the pixel having the maximum luminance value by a coefficient relating to a series of pixel signal processing is below a threshold value. Then, whether or not correction is to be performed is determined, or the correction coefficient is adjusted so as not to saturate.

  By performing / non-performing the correction and adjusting the correction coefficient for each image region, it is possible to perform the peripheral light amount reduction correction more suitable for the shooting scene. In particular, when the correction coefficient is uniformly adjusted, the adjustment may be performed to an area that does not need to be adjusted. However, by adjusting each image area, the adjustment can be performed only to the area that requires the adjustment. Further, when it is determined that the conspicuousness of the pixel saturation / brightness improvement occurs in the peripheral portion of the image, it is determined that the correction is not performed with priority. As a result, it is possible to prevent a situation in which the image quality is conspicuous, such as saturation at the peripheral portion of the image, and unnecessary noise increase due to excessive unevenness in luminance level, resulting in unnecessary noise increase.

  By determining the luminance variation, luminance difference, and the luminance improvement analysis area of the saturation / peripheral portion for each block, areas that are easy to analyze for the luminance variation, luminance difference, and luminance improvement of the saturation / periphery portion are separately and appropriately. It can be set. In particular, by making the luminance variation analysis area different from the luminance difference analysis area, the luminance analysis can be performed effectively, and the saturation / peripheral partial luminance improvement area is defined as an area different from the luminance difference and the luminance variation. By doing so, the determination of correction is made for the peripheral area where the luminance decrease is not relatively large, and the image quality can be reliably prevented from lowering.

  Note that it is also possible to determine a luminance difference and a luminance variation detection area instead of a block unit as in the first embodiment. Further, the peripheral regions c1 to c4 may be determined to have an area shape other than the rectangular shape. The area subject to pixel saturation / noise analysis may be set to the same area as the luminance variation and luminance difference target area.

  In the present embodiment, it is determined whether or not correction is to be performed for the luminance variation, the luminance difference, and the luminance improvement of the saturation / peripheral portion. In each case, the correction is performed only when the correction is necessary, and the correction coefficient is determined. Has been adjusted. However, execution of correction / non-correction for each of these may be determined for each image region, and / or the correction coefficient may be adjusted. Alternatively, the determination may be made based on a combination of any two or more of these. As for the adjustment of the correction coefficient, the correction ratio application rate may be calculated based on only the standard deviation as in the first embodiment.

  In the analysis of the photographed scene, the standard deviation, the luminance difference, and the like are obtained based on the luminance signal. However, it is also possible to use the values of the R, G, and B signals themselves, and the pixel values of the color difference signal. In this embodiment, the correction processing is applied to the digital camera. However, the correction processing can be applied to other imaging devices or image processing devices. Further, the photographing scene may be analyzed by analyzing the luminance variation and the luminance difference in the peripheral portion of the image, and the image processing or the photographing operation other than the shading correction may be controlled. That is, an invention that does not use the configuration of the correction unit can be applied to an imaging device or the like.

100 Digital camera (photographing device)
101 signal processing unit 132 image signal processing unit 134 correction unit

Claims (10)

For a plurality of image areas in the captured image, a detection unit that detects a luminance difference and a luminance variation from the image center side to the peripheral direction,
Based on a correction coefficient determined according to the characteristics of the imaging optical system, a correction unit that performs peripheral light intensity reduction correction of the captured image,
The imaging apparatus according to claim 1 , wherein the correction unit uniformly adjusts a correction coefficient when a degree of variation in luminance is smaller than a first threshold and a luminance difference is larger than a second threshold . The imaging apparatus according to claim 1, wherein the correction unit detects a standard deviation as a value indicating a degree of luminance variation . The detection unit detects, for a plurality of image areas, a luminance difference and a variation in luminance,
2. The correction unit according to claim 1, wherein the correction unit adjusts a correction coefficient according to the number of image areas in which a degree of luminance variation is smaller than a first threshold and a luminance difference is larger than a second threshold. 3. Imaging device.   4. The imaging apparatus according to claim 3, wherein the correction unit adjusts a correction coefficient when there are a plurality of image regions in which a degree of luminance variation is smaller than a first threshold and a luminance difference is larger than a second threshold. 5. apparatus.   When the number of image areas in which the degree of luminance variation is smaller than the first threshold value and the luminance difference is larger than the second threshold value is plural and the image areas are adjacent to each other, The imaging device according to claim 3, wherein the adjustment is performed. The detection unit detects a variation in luminance with respect to an integrated region in which image regions adjacent to each other are combined,
The imaging device according to claim 3, wherein the correction unit adjusts a correction coefficient according to a degree of variation in luminance of the integrated area.   The correction unit may be configured such that the degree of luminance variation is smaller than the first threshold value and the number of image areas whose luminance difference is greater than the second threshold value is equal to or less than a predetermined number, or the degree of luminance variation is smaller than the first threshold value. 7. The method according to claim 3, wherein the peripheral light amount reduction correction is not performed when the number of image areas having a small brightness difference is larger than the second threshold value but not adjacent to each other. An imaging device according to claim 1. The imaging device according to claim 1, wherein the detection unit detects a variation in luminance and a luminance difference in a plurality of image regions each including a peripheral portion in the captured image. Imaging device,
Detecting means for detecting a luminance difference and a luminance variation from the image center side to the peripheral direction with respect to a plurality of image regions of the captured image;
Based on a correction coefficient determined according to the characteristics of the photographic optical system, function as a correction unit that performs peripheral light amount reduction correction of the captured image,
When the degree of variation in luminance is smaller than a first threshold value and the luminance difference is larger than a second threshold value , the program functions as the correction means so as to uniformly adjust the correction coefficient . For a plurality of image regions of the captured image, a luminance difference and a luminance variation from the image center side to the peripheral direction are detected,
A method for performing peripheral light amount reduction correction of a captured image based on a correction coefficient determined according to characteristics of an imaging optical system,
An image processing method , wherein a correction coefficient is uniformly adjusted when a degree of luminance variation is smaller than a first threshold and a luminance difference is larger than a second threshold .