JP2008022502A - Imaging apparatus, imaging method, imaging program and image processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an imaging apparatus capable of appropriately performing red-eye correction, for example, and reducing a user-side processing burden. <P>SOLUTION: As shown in Fig.2, the imaging apparatus comprises: a red-eye detection unit 67 which detects a plurality of red-eye correction candidate regions as candidate regions to perform red-eye correction thereon from a flash photographed image and generates red-eye detection reliability indicating the certainty as red eyes of the detected red-eye correction candidate regions for each of the red-eye correction candidate regions; a determination unit 71 which determines a degree of detection success for each red-eye correction candidate region based on at least the red-eye detection reliability; and a display unit which displays, together with the image, deletion candidates as red-eye correction candidate regions to become candidates to be deleted from a target of red-eye correction in the plurality of red-eye correction candidate regions in order from the lowest degree. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an imaging apparatus, an imaging method, an imaging program, and an image processing apparatus.

  Conventionally, an imaging apparatus that performs red-eye correction by batch processing has been proposed (see, for example, Patent Document 1).

An imaging device that performs red-eye correction by interactive processing has also been proposed.
JP 2002-247596 A

  In the technique of Patent Document 1 (a technique for performing red-eye correction by batch processing), an iris area and other feature areas are extracted from a captured image using hue, saturation, brightness, and the like. Then, based on the arrangement relationship between the iris region and other feature regions, the red-eye region that is the target region for performing red-eye correction is specified.

  However, in the technique of Patent Document 1, if the red-eye area cannot be correctly specified, for example, red-eye correction may be performed on an inappropriate red-eye area, and thus there is a possibility that the red-eye correction cannot be performed appropriately.

  On the other hand, in the technique of performing red eye correction by interactive processing, even when a plurality of red eye areas are detected, it is necessary to specify whether or not to perform red eye correction for each of all red eye areas. This tends to complicate processing on the user side in red-eye correction.

  An object of the present invention is to provide an imaging apparatus that can appropriately perform red-eye correction, for example, and can reduce the processing burden on the user side.

  The imaging apparatus according to the present invention detects a plurality of red-eye correction candidate areas that are candidates for red-eye correction from an image captured with flash, and indicates the red-eye likeness of the detected red-eye correction candidate area. A red-eye detection unit that generates red-eye detection reliability for each of the red-eye correction candidate regions, and a degree of success in detection based on at least the red-eye detection reliability for each of the red-eye correction candidate regions A determination unit that determines, and a display unit that displays, together with the image, deletion candidates that are red-eye correction candidate regions that are candidates to be deleted from the red-eye correction target in the plurality of red-eye correction candidate regions in order of decreasing degree. It is characterized by having.

  In the imaging apparatus according to the present invention, for example, red-eye correction can be performed appropriately, and the processing burden on the user side can be reduced.

  An imaging apparatus according to a preferred first embodiment of the present invention will be described with reference to FIGS.

  First, a schematic configuration of the imaging apparatus will be described with reference to FIG.

  The imaging device 1 is a device having a red-eye correction function. The imaging device 1 mainly includes an input unit 5, an optical system 10, a solid-state imaging device 20, an analog signal processing unit 30, an A / D conversion unit 40, a flash 50, and a digital signal processing unit 60. The imaging apparatus 1 includes an image display unit 70, an internal storage unit 80, and an external I / F 90.

  The input unit 5 has buttons. The buttons include, for example, a shutter button 5a, a selection button 5b, a delete button 5c, and a confirm button 5d (see FIG. 15) provided on the main body.

  The optical system 10 includes optical devices such as a lens and a diaphragm, and a driving system thereof. The optical system 10 is provided between the subject O and the solid-state imaging device 20.

  The solid-state imaging device 20 includes a CCD sensor, a CMOS sensor, and the like. The solid-state imaging device 20 is connected to the analog signal processing unit 30.

  The analog signal processing unit 30 includes a clamp circuit and an amplifier circuit. The analog signal processing unit 30 is connected to the solid-state imaging device 20 and the A / D conversion unit 40.

  The A / D conversion unit 40 includes an A / D converter and the like. The A / D conversion unit 40 is connected to the analog signal processing unit 30 and the digital signal processing unit 60.

  The flash 50 has a strobe or the like. The flash 50 is connected to the digital signal processing unit 60.

  The digital signal processing unit 60 includes a filter circuit, a detection circuit, and the like. The digital signal processing unit 60 is connected to the A / D conversion unit 40, the flash 50, the image display unit 70, the internal storage unit 80, and the external I / F 90.

  The image display unit 70 includes a display device, a display control circuit, and the like. The image display unit 70 is connected to the digital signal processing unit 60 and the internal storage unit 80. The display device is used as an electronic viewfinder.

  The internal storage unit 80 includes a recording medium such as a semiconductor memory and its control circuit. The internal storage unit 80 is connected to the digital signal processing unit 60 and the image display unit 70.

  The external I / F 90 has a configuration capable of connecting an external storage device. The external I / F 90 is connected to the digital signal processing unit 60.

  Next, a schematic operation of the imaging apparatus will be described with reference to FIG.

  The optical system 10 is directed to a shooting area including the subject O (a target area acquired as the image GI in FIG. 10). As a result, the optical system 10 forms an optical image of the imaging region including the subject O on the solid-state imaging device 20.

  The user inputs flash emission information, which is information for instructing flash emission, to the digital signal processing unit 60 via the input unit 5 at the time of shooting at low illumination. The digital signal processing unit 60 causes the flash 50 to emit light according to the flash emission information. As a result, the flash 50 illuminates the shooting area including the subject O. The solid-state imaging device 20 forms an optical image of an imaging region including the subject O illuminated by the flash. At this time, an optical image of the imaging region may be formed with red eyes in the subject O.

  The solid-state imaging device 20 generates an electrical signal corresponding to the formed optical image, and converts the optical image (input image) into an electrical signal (analog signal). The solid-state imaging device 20 supplies the converted analog signal to the analog signal processing unit 30.

  The analog signal processing unit 30 performs processes such as clamping and gain on the waveform of the analog signal. The analog signal processing unit 30 supplies the processed analog signal to the A / D conversion unit 40.

  The A / D converter 40 converts an analog signal into a digital signal. The A / D converter 40 supplies the converted digital signal to the digital signal processor 60.

  The digital signal processing unit 60 performs various corrections on the digital signal and generates image data for displaying an output image. Here, when generating the image data, the digital signal processing unit 60 accesses the internal storage unit 80 and stores part or all of the image data in the internal storage unit 80. The digital signal processing unit 60 supplies the image data to the image display unit 70, the internal storage unit 80, and the external I / F 90.

  The image display unit 70 receives image data from the digital signal processing unit 60 and the internal storage unit 80, and sequentially displays output images corresponding to the image data. Thereby, the image display unit 70 causes the display device to function as an electronic viewfinder.

  An external storage device is connected to the external I / F 90. Thereby, the image data supplied to the external I / F 90 is finally stored in the external storage device.

  A detailed configuration of the digital signal processing unit will be described with reference to FIG.

  The digital signal processing unit 60 mainly includes a color conversion unit 61, a color processing unit 62, an LPF (Low Pass Filter) 63, a luminance processing unit 64, and a BPF (Band Pass Filter) 65. The digital signal processing unit 60 includes an outline emphasizing unit 66, a red-eye detecting unit 67, a face detecting unit 68, a determining unit 71, and a red-eye correcting unit 69. These are the configurations of the portion of the digital signal processing unit 60 that performs image data generation and red-eye correction.

  The color conversion unit 61 is connected to the A / D conversion unit 40 (see FIG. 1), the color processing unit 62, the LPF 63, and the BPF 65. The color processing unit 62 is connected to the color conversion unit 61, the red eye detection unit 67, and the determination unit 71. The luminance processing unit 64 is connected to the LPF 63, the red eye detection unit 67, and the determination unit 71. The contour enhancement unit 66 is connected to the BPF 65, the face detection unit 68, and the determination unit 71. The determination unit 71 is connected to the color processing unit 62, the luminance processing unit 64, the contour enhancement unit 66, the red eye detection unit 67, the face detection unit 68, and the red eye correction unit 69. The red-eye correction unit 69 is connected to the image display unit 70, the internal storage unit 80, and the external I / F 90.

  Next, the detailed operation of the digital signal processing unit will be described with reference to FIG.

  A digital signal indicating RGB of the optical image (input image) is input from the A / D converter 40 to the color converter 61. The color conversion unit 61 generates and separates the luminance signal Y and the color difference signals Cr and Cb from the RGB of the input image. The color conversion unit 61 supplies the color difference signals Cr, Cb and the like to the color processing unit 62 and supplies the luminance signal Y and the like to the LPF 63 and the BPF 65.

  The color processing unit 62 performs white balance adjustment, color processing, gamma processing, and the like on the color difference signals Cr and Cb. Here, the color processing is, for example, processing for reducing the intensity of high luminance and low luminance colors. The color processing unit 62 supplies the color signal subjected to these processes to the red-eye detection unit 67 and the determination unit 71.

  The luminance signal Y and the like are input from the color conversion unit 61 to the LPF 63. The LPF 63 passes a low frequency component of the input signal such as the luminance signal Y. The LPF 63 supplies a low frequency component such as the luminance signal Y to the luminance processing unit 64.

  The luminance processing unit 64 performs filter processing, luminance correction, gamma processing, and the like on low frequency components such as the luminance signal Y. The luminance processing unit 64 supplies the luminance signal subjected to these processes to the red-eye detection unit 67 and the determination unit 71.

  A luminance signal Y and the like are input from the color conversion unit 61 to the BPF 65. The BPF 65 passes a high frequency component of the input signal such as the luminance signal Y. The BPF 65 supplies a high-frequency component (contour signal) such as a luminance signal Y to the contour emphasizing unit 66.

  The contour emphasizing unit 66 performs contour emphasis processing on high-frequency components (contour signals) such as the luminance signal Y. The contour enhancement unit 66 supplies the contour signal subjected to these processes to the face detection unit 68 and the determination unit 71.

  The red-eye detection unit 67 receives flash emission information from the flash 50 via the input unit 5. The red-eye detection unit 67 receives a color signal from the color processing unit 62 and receives a luminance signal from the luminance processing unit 64. The red-eye detection unit 67 detects a red-eye correction candidate region via a color signal and a luminance signal. Here, the red-eye correction candidate area is an area that is a candidate for red-eye correction of the subject O. At the same time, the red-eye detection unit 67 generates and outputs red-eye detection reliability based on the red-eye detection result. The red-eye detection reliability indicates the redness of an area detected from the subject O as a red-eye correction candidate area. The red-eye detection reliability is, for example, a numerical value indicating the reliability of the detection result of the red-eye correction candidate area of the subject O. When the red-eye detection reliability is a large value, the reliability is high. Indicates that the reliability is low.

  The face detection unit 68 receives the contour signal from the contour enhancement unit 66. The face detection unit 68 detects the face of the subject O with respect to the red-eye correction candidate region via a contour signal or the like. At the same time, the face detection unit 68 generates and outputs a face detection reliability based on the face detection result. The face detection reliability indicates the likelihood of a face in an area detected as a face candidate area from the subject O. The face candidate area is an area recognized as the face of the subject O. The face detection reliability is, for example, a numerical value indicating the reliability of the result of detecting the face candidate area of the subject O. When the face detection reliability is a large value, the reliability is high. When the face detection reliability is a small value, Indicates that the reliability is low.

  Note that a known face detection technique is used for the face detection in the present invention. As a known face detection technique, for example, there is a technique using learning represented by a neural network. Alternatively, as a known face detection technique, there is a method in which a part having a characteristic shape such as an eye, a nose, or a mouth is searched from an image area using template matching and is regarded as a face if the similarity is high. In addition, many other methods have been proposed, such as methods that detect image features such as skin color and eye shape and use statistical analysis. In general, multiple methods are used to recognize faces. Is common.

  Specific examples include a face detection method using wavelet transform and image feature amount described in JP-A-2002-251380.

  The determination unit 71 receives information on the red-eye correction candidate region and the red-eye detection reliability from the red-eye detection unit 67, and receives information on the face of the subject O and the face detection reliability from the face detection unit 68. The determination unit 71 receives a color signal from the color processing unit 62, receives a luminance signal from the luminance processing unit 64, and receives an edge enhancement signal from the edge enhancement unit 66. Based on these pieces of information, the determination unit 71 determines the degree of success in detecting the red eye of the subject O for each red-eye correction candidate region. Here, when determining, the determination unit 71 accesses a memory (not shown) and refers to the success degree determination table (see FIG. 7). The determination unit 71 supplies information on the degree of success to the red-eye correction unit 69 in association with identification information (for example, red-eye position information described later) of the red-eye correction candidate region. In addition, the determination unit 71 transfers information on the red-eye correction candidate area, information on the face of the subject O, color signal, luminance signal, and contour signal to the red-eye correction unit 69.

  The red-eye correction unit 69 outputs information indicating the degree of success to the image display unit 70 in association with the identification information of the red-eye correction candidate area.

  In addition, the red-eye correction unit 69 receives a delete command input via the delete button 5c (see FIG. 12) of the input unit 5 in association with identification information of the red-eye correction candidate area. The red-eye correction unit 69 deletes information on the red-eye correction candidate area specified by the deletion command.

  The red-eye correction unit 69 receives a confirmation command input via the confirmation button 5d (see FIG. 12) of the input unit 5. The red-eye correction unit 69 performs red-eye correction on the red-eye correction candidate area according to the confirmation command. Here, red-eye correction is not performed on the red-eye correction candidate area whose information has already been deleted.

  Next, the flow of processing in which the imaging apparatus performs red-eye correction will be described using the flowcharts shown in FIGS. 3 to 9 and the conceptual diagrams of FIGS. The processing shown in FIGS. 3 to 9 is mainly performed by the digital signal processing unit shown in FIG.

  In step S1, the red-eye correction unit 69 determines whether flash photography has been performed. That is, the flash light emission information is input to the digital signal processing unit 60 via the input unit 5 at the time of low-illuminance shooting by the user. The flash emission information is information for instructing flash emission. The digital signal processing unit 60 causes the flash 50 to emit light according to the flash emission information. As a result, the flash 50 illuminates the shooting area including the subject O. The solid-state imaging device 20 forms an optical image of an imaging region including the subject O illuminated by the flash. At this time, an optical image of the imaging region may be formed with red eyes in the subject O.

  When the red-eye correction unit 69 in the digital signal processing unit 60 receives the flash emission information from the flash 50 via the red-eye detection unit 67 and the determination unit 71, the red-eye correction unit 69 determines that flash photography has been performed. When the red-eye correction unit 69 does not receive the flash emission information from the flash 50 via the red-eye detection unit 67 and the determination unit 71, the red-eye correction unit 69 determines that flash photography has not been performed. If the red-eye correction unit 69 determines that flash photography has been performed, the process proceeds to step S2, and if it is determined that flash photography has not been performed, the processing ends.

  In step S2, the digital signal processing unit 60 sets thresholds for red-eye detection reliability and face detection reliability. That is, the red-eye correction unit 69 in the digital signal processing unit 60 reads the red-eye detection threshold value TH1 and the face detection threshold value TH2 stored in the memory (not shown) and passes them to the determination unit 71.

  The determination unit 71 supplies information indicating that the red-eye detection threshold value TH1 has been set to the red-eye detection unit 67 and also supplies information indicating that the face-detection threshold value TH2 has been set to the face detection unit 68.

  In step S3, the red-eye detection unit 67 performs red-eye detection processing in response to receiving information indicating that the red-eye detection threshold TH1 has been set. For example, the red-eye detection unit 67 detects the red-eye correction candidate region and calculates the red-eye detection reliability. When the red-eye detection unit 67 completes the red-eye detection process, the red-eye detection unit 67 supplies information indicating that the red-eye detection process has been completed, a red-eye correction candidate region, and a red-eye detection reliability to the determination unit 71. Details of the red-eye detection process will be described later.

  In step S4, the face detection unit 68 performs the known face detection process described above in response to receiving information indicating that the face detection threshold value TH2 has been set. For example, the face detection unit 68 detects the face candidate area and calculates the face detection reliability. When the face detection process is completed, the face detection unit 68 supplies information indicating that the face detection process is completed, a face correction candidate region, and a face detection reliability to the determination unit 71. The detailed description of the face detection process will be described later.

In step S5, the determination unit 71 performs a success degree determination process in response to receiving both information indicating that the red-eye detection process is completed and information indicating that the face detection process is completed. A detailed description of the success degree determination process will be described later.
Success degree “Low” Success degree “Low” Success degree “Low” Success degree “NG” Success degree “NG” Success degree “NG” In step S 6, the red-eye correction unit 69 performs a red-eye correction process. Details of the red-eye correction process will be described later.

  In this way, the imaging device 1 performs red-eye correction.

  Next, the red-eye detection process will be described using the flowchart shown in FIG. 4 and the conceptual diagrams of FIGS. 10 and 11. 10 and 11 are examples of screens displayed mainly by the image display unit 70 shown in FIG. Note that FIG. 11 is an example of a virtual screen for explaining internal processing, and is not actually displayed by the image display unit 70.

  In step S21, the color difference signal is input to the color processing unit 62, and the luminance signal is input to the LPF 63 and the BPF 65. That is, a digital signal indicating RGB of the optical image (input image) is input from the A / D conversion unit 40 to the color conversion unit 61 of the digital signal processing unit 60. The color conversion unit 61 generates and separates the luminance signal Y and the color difference signals Cr and Cb from the RGB of the input image. The color conversion unit 61 supplies the color difference signals Cr, Cb and the like to the color processing unit 62 and supplies the luminance signal Y and the like to the LPF 63 and the BPF 65.

  The color processing unit 62 performs white balance adjustment, color processing, gamma processing, and the like on the color difference signals Cr and Cb. Here, the color processing is, for example, processing for reducing the intensity of high luminance and low luminance colors. The color processing unit 62 supplies the color signal subjected to these processes to the red-eye detection unit 67 and the determination unit 71.

  The luminance signal Y and the like are input from the color conversion unit 61 to the LPF 63. The LPF 63 passes a low frequency component of the input signal such as the luminance signal Y. The LPF 63 supplies a low frequency component such as the luminance signal Y to the luminance processing unit 64.

  The luminance processing unit 64 performs filter processing, luminance correction, gamma processing, and the like on low frequency components such as the luminance signal Y. The luminance processing unit 64 supplies the luminance signal subjected to these processes to the red-eye detection unit 67 and the determination unit 71.

  In step S22, the red eye detector 67 performs red eye detection on the input signal. That is, the red-eye detection unit 67 receives a color signal from the color processing unit 62 and receives a luminance signal from the luminance processing unit 64. The red-eye detection unit 67 detects a red-eye correction candidate region via a color signal and a luminance signal. Here, the red-eye correction candidate area is an area that is a candidate for red-eye correction of the subject O.

  For example, as shown in FIG. 10, consider a case where subjects O1 to O3 exist in the shooting area indicated by the image GI. Here, the subjects O1 and O2 face the front, while the subject O3 faces the side. At this time, the red-eye detection unit 67 detects red-eye correction candidate regions R1, R2, R3, R4, and R5 as shown in FIG.

  In step S23, the red eye detection unit 67 generates and outputs a red eye detection reliability. The red-eye detection reliability indicates the redness of an area detected from the subject O as a red-eye correction candidate area. The red-eye detection reliability is, for example, a numerical value indicating the reliability of the detection result of the red-eye correction candidate area of the subject O, and indicates that the reliability is high when the value is large, and is a small value. In some cases, the reliability is low.

Specifically, the red eye detection unit 67 analyzes hue, saturation, brightness, luminance, and the like. Then, the red-eye detection unit 67 calculates whether the corresponding pixel (area) is red (parameter RP1) or whether the peripheral pixel (area) of the corresponding pixel (area) is skin color (parameter RP2). The red-eye detection unit 67 calculates whether the corresponding pixel (area) has high luminance (parameter RP3) or whether the corresponding pixel (area) has a round shape (parameter RP4). Then, the red-eye detection unit 67 sets the red-eye detection reliability RR to RR = K1 × RP1 + K2 × RP2 + K3 × RP3 + K4 × RP4 Equation 1
It calculates by the formula of Here, K1, K2, K3, and K4 are constants obtained in advance through experiments or the like.

  In the present embodiment, the elements constituting the red-eye detection reliability RR are defined as described above. However, the elements are not limited to the above-described elements as long as they are effective elements for estimating the reliability of red-eye detection. .

  For example, the red-eye detection unit 67 performs the calculation of Equation 1 and the red-eye detection reliability RR1, RR2, RR3, RR4, RR5 for the red-eye correction candidate regions R1, R2, R3, R4, R5 (see FIG. 11). Is generated. For example, when an input signal as shown in FIG. 11 is obtained, the red-eye detection reliability RR1, RR2, RR3, RR4, RR5 and the red-eye detection threshold TH1 are in a relationship as shown in Equations 2 to 6, respectively. It will be.

RR1 ≧ TH1 Formula 2
RR2 ≧ TH1 Formula 3
RR3 <TH1 ... Formula 4
RR4 ≧ TH1 Formula 5
RR5 <TH1 ... Formula 6
In step S24, the red eye detection unit 67 outputs the red eye detection reliability. That is, the red-eye detection unit 67 in the digital signal processing unit 60 outputs information on the red-eye correction candidate regions (for example, R1 to R5) and red-eye detection reliability (for example, RR1 to RR5) to the determination unit 71.

  Next, the face detection process will be described mainly using the flowchart shown in FIG. 5 and the conceptual diagrams shown in FIGS. 10 and 11 are examples of screens displayed mainly by the image display unit 70 shown in FIG. Note that FIG. 11 is an example of a virtual screen for explaining internal processing, and is not actually displayed by the image display unit 70.

  In step S31, the face detection unit 68 detects a face area by using the known face detection method described above for the input signal.

  For example, as shown in FIG. 10, consider a case where subjects O1 to O3 exist in the shooting area indicated by the image GI. Here, the subjects O1 and O2 face the front, while the subject O3 faces the side. At this time, the face detection unit 68 detects face candidate regions F1, F2, and F3 as shown in FIG. The face candidate areas F1, F2, and F3 are areas recognized as the faces of the subjects O1, O2, and O3, respectively.

  In step S32, the face detection unit 68 calculates and outputs a face detection reliability based on the face detection result in step S31. The face detection reliability indicates the likelihood of a face in an area detected as a face candidate area from the subject O. The face detection reliability is, for example, a numerical value indicating the reliability of the result of detecting the face candidate area of the subject O. When the face detection reliability is a large value, the reliability is high. When the face detection reliability is a small value, Indicates that the reliability is low.

Specifically, the face detection unit 68 analyzes a luminance pattern and the like. Then, the face detection unit 68 calculates whether eyes are detected (parameter FP1) or nose is detected (parameter FP2). Further, the face detection unit 68 calculates whether the mouth is detected (parameter FP3) or the ear is detected (parameter FP4). Then, the face detection unit 68 sets the face detection reliability FR to FR = K5 × FP1 + K6 × FP2 + K7 × FP3 + K8 × FP4 Equation 7
It calculates by the formula of Here, K5, K6, K7, and K8 are constants obtained in advance by experiments or the like.

  In the present embodiment, the elements constituting the red-eye detection reliability RR are defined as described above. However, the elements are not limited to the above-described elements as long as they are effective elements for estimating the reliability of red-eye detection. .

  For example, the face detection unit 68 performs the calculation of Equation 7, and generates face detection reliability FR1, FR2, FR3 for the face candidate regions F1, F2, F3. For example, in the case shown in FIG. 11, the face detection reliability FR1, FR2, FR3 and the face detection threshold TH2 are in a relationship as shown in Expression 8 to Expression 10, respectively.

FR1 ≧ TH2 Expression 8
FR2 ≧ TH2 Expression 9
FR3 ≧ TH2 Expression 10
In step S33, the face detection reliability is output. That is, the face detection unit 68 of the digital signal processing unit 60 includes information on the face of the subject O (for example, information on the face candidate areas F1, F2, and F3) and face detection reliability (for example, face detection reliability FR1, FR2, and so on). FR3) to the determination unit 71.

  Next, the success degree determination process will be described with reference to a flowchart shown in FIG. 6 and conceptual diagrams of FIGS. 7, 10, and 11. FIG. FIG. 7 shows a success degree reference table that the determination unit 71 shown in FIG. 2 accesses and references a memory (not shown). 10 and 11 are examples of screens displayed mainly by the image display unit 70 shown in FIG. Note that FIG. 11 is an example of a virtual screen for explaining internal processing, and is not actually displayed by the image display unit 70.

  In step S51, the determination unit 71 selects a red-eye correction candidate area. That is, the determination unit 71 in the digital signal processing unit 60 receives information on the red-eye correction candidate area (including red-eye position information) and the red-eye detection reliability from the red-eye detection unit 67. The red-eye position information is information indicating the position of the red-eye correction candidate area in the image. The determination unit 71 selects a red-eye correction candidate region (red-eye position information) to be determined from a plurality of red-eye correction candidate regions. Note that the determination unit 71 specifies a red-eye correction candidate area that has already been determined with reference to a memory (not shown), and additionally stores the red-eye correction candidate area to be determined in the memory.

  In step S52, the determination unit 71 determines whether the red-eye correction candidate region is included in the face candidate region. That is, the determination unit 71 receives information on face correction candidate regions (including face position information) and face detection reliability from the face detection unit 68. The face position information is information indicating the position of the face candidate area in the image.

  The determination unit 71 determines the inclusion relationship between the red-eye correction candidate area and the face candidate area based on the red-eye position information and the face position information. If the determination unit 71 determines that the red-eye position information is included in any one or more of the face position information, the red-eye correction candidate area corresponding to the red-eye position information is included in the face candidate area. Judge. If the determination unit 71 determines that the red-eye position information is not included in any of the one or more pieces of face position information, the red-eye correction candidate area corresponding to the red-eye position information is not included in the face candidate area. Judge.

  For example, consider a case where red-eye correction candidate regions R1 to R5 and face candidate regions F1 to F3 are detected for the subjects O1 to O3 shown in FIG. 10, as shown in FIG. In this case, the determination unit 71 determines that the red-eye correction candidate regions R1 to R3 are included in the face candidate regions F1 to F3, respectively. The determination unit 71 determines that the red-eye correction candidate regions R4 and R5 are not included in any of the face candidate regions F1 to F3.

  If the determination unit 71 determines that the red-eye correction candidate area is included in the face candidate area, the process proceeds to step S54. If the determination unit 71 determines that the red-eye correction candidate area is not included in the face candidate area, the process proceeds to step S53.

  In step S53, the determination unit 71 determines that the face detection reliability is “low”. That is, since the red-eye correction candidate area is not included in the face candidate area, a virtual face candidate area including the red-eye correction candidate area is assumed, and it is determined that the face candidate area has a low likelihood of being a face.

  For example, in the case illustrated in FIG. 11, the determination unit 71 assumes virtual face candidate regions F4 and F5 including red-eye correction candidate regions R4 and R5. Then, the determination unit 71 determines the face detection reliability of the face candidate areas F4 and F5 to be “low”.

  In step S54, the determination unit 71 determines whether or not the face detection reliability is greater than or equal to a threshold value. That is, the determination unit 71 of the digital signal processing unit 60 receives the face information of the subject O and the face detection reliability from the face detection unit 68. The determination unit 71 compares the face detection reliability FR indicated by the face detection reliability with the face detection threshold TH2. Then, when determining that FR ≧ TH2, the determination unit 71 determines that the face detection reliability is “high”. The determination unit 71 determines that the face detection reliability is “low” when determining that FR <TH2.

  For example, in the case illustrated in FIG. 11, the determination unit 71 determines that the face detection reliability is “high” for the face candidate regions F1, F2, and F3, as illustrated in Equations 8 to 10.

  Here, the reason for using the face detection reliability in addition to the red-eye detection reliability in order to determine the success degree will be described.

  Originally, eyes exist only in the face. Therefore, based on the assumption that the eyes exist in the face, even if it is a red eye correction candidate area, if the reliability as the face candidate area is low, it is determined that it is not an eye, and the red eye correction candidate area Exclude. Thereby, the precision of determination can be improved more.

  That is, since it is determined and determined whether or not the red-eye detection is successful in consideration of the face detection reliability, it is easier to distinguish the red-eye correction candidate area that should be corrected for red-eye and the red-eye correction candidate area that should not be corrected for red-eye. Become.

  In the present embodiment, only one threshold value is set. However, by setting a plurality of threshold values, the ranking of facial appearance may be further refined. By doing so, it is possible to prioritize more finely when a plurality of face candidate regions are detected.

  In step S55, the determination unit 71 determines whether or not the red-eye detection reliability is equal to or higher than a threshold value. That is, the determination unit 71 in the digital signal processing unit 60 compares the red-eye detection reliability RR1 indicated by the red-eye detection reliability with the red-eye detection threshold TH1. When determining that RR1 ≧ TH1, the determination unit 71 determines that the red-eye detection reliability is “high”. The determination unit 71 determines that the red-eye detection reliability is “low” when determining that RR1 <TH1.

  For example, in the case illustrated in FIG. 11, the determination unit 71 determines that the red-eye detection reliability is “high” for the red-eye correction candidate regions R1, R2, and R4 as illustrated in Equations 2, 2, and 5. . As shown in Equations 4 and 6, the determination unit 71 determines that the red-eye detection reliability is “low” for the red-eye correction candidate regions R 3 and R 5.

  In the present embodiment, only one threshold value is set. However, by setting a plurality of threshold values, the rank classification of redness may be further refined. By doing so, it is possible to prioritize more finely when a plurality of red-eye candidate regions are detected.

  In step S56, the determination unit 71 determines the degree of success. That is, the determination unit 71 determines the degree (success level) of successful red-eye detection based on the determination result regarding the red-eye detection reliability and the determination result regarding the face detection reliability. Here, when determining, the determination unit 71 accesses a memory (not shown) and refers to the success degree determination table (see FIG. 7).

  As illustrated in FIG. 7, when the determination unit 71 determines that the red-eye detection reliability is “high” and the face detection reliability is “high”, the determination unit 71 determines that the success degree is “high”. When the determination unit 71 determines that the red-eye detection reliability is “low” and the face detection reliability is “high”, the determination unit 71 determines that the success degree is “medium”. When the determination unit 71 determines that the red-eye detection reliability is “high” and the face detection reliability is “low”, the determination unit 71 determines that the success degree is “low”. When the determination unit 71 determines that the red-eye detection reliability is “low” and the face detection reliability is “low”, the determination unit 71 determines the success level as “NG”.

  For example, in the case illustrated in FIG. 11, the determination unit 71 determines that the red-eye detection reliability of the red-eye correction candidate regions R1 and R2 is “high”, and the faces of the face candidate regions F1 and F2 including the red-eye correction candidate regions R1 and R2 The detection reliability is determined as “high”. Accordingly, the determination unit 71 determines the success degree as “high” for the red-eye correction candidate region R1.

  The determination unit 71 determines that the red-eye detection reliability of the red-eye correction candidate region R3 is “low”, and determines the face detection reliability of the face candidate region F3 including the red-eye correction candidate region R3 is “high”. Accordingly, the determination unit 71 determines the success degree as “medium” for the red-eye correction candidate region R3.

  The determination unit 71 determines the red eye detection reliability of the red eye correction candidate region R4 as “high”, and determines the face detection reliability of the virtual face candidate region F4 including the red eye correction candidate region R4 as “low”. Yes. Thereby, the determination unit 71 determines the success degree as “small” for the red-eye correction candidate region R4.

  The determination unit 71 determines that the red-eye detection reliability of the red-eye correction candidate region R5 is “low”, and determines the face detection reliability of the virtual face candidate region F5 including the red-eye correction candidate region R5 is “low”. Yes. Accordingly, the determination unit 71 determines the success degree as “NG” for the red-eye correction candidate region R5.

In step S57, the determination unit 71 determines whether determination has been performed for all red-eye correction candidate regions. If it is determined that the determination has been made, the process proceeds to step S12. If it is determined that the determination has not been made, the process proceeds to step S4.
In addition, in step S5 and step S8, by making the rank division finer, the rank division of the degree of success in red-eye detection can be performed in more detail.

  As described above, the success degree determination process ranks the degree of success of red-eye detection for all red-eye correction candidate areas as “high”, “medium”, “low”, and “NG”. That is, the determination unit 71 of the digital signal processing unit 60 determines the degree of successful red-eye detection.

  Next, the red-eye correction process will be described with reference to the flowcharts shown in FIGS. 8 and 9 and the conceptual diagrams of FIGS. 12 to 15 are examples of rear views of the imaging apparatus.

  In step S <b> 41, the image display unit 70 displays an area determined as the success level “NG” (hereinafter referred to as “NG area”). That is, the input unit 5 of the digital signal processing unit 60 receives a display command for the NG area by detecting that the user has pressed the selection button 5b, for example. The input unit 5 supplies an NG area display command to the digital signal processing unit 60. The display command for the NG area is supplied to the red-eye correction unit 69 via the red-eye detection unit 67 and the determination unit 71 in the digital signal processing unit 60. The red-eye correction unit 69 selects an NG area to be displayed from one or more NG areas based on an NG area display command. The red-eye correction unit 69 supplies the information on the selected NG area together with the image information to the image display unit 70. The image display unit 70 displays deletion candidates on the display device based on the image information and the information on the selected NG area. Here, the deletion candidate is a red-eye correction candidate area that is a candidate to be deleted from the red-eye correction target in a plurality of red-eye correction candidate areas.

  For example, in FIG. 12, the image display unit 70 displays a frame RF5 indicating that the red-eye correction candidate region R5 is an NG region together with the image GI as information indicating a deletion candidate.

  In the present embodiment, the image GI is displayed in accordance with the size of the image display unit 70, but it is also possible to display an enlarged image by operating the selection button 5b.

  Thereby, it is possible to grasp that the red-eye correction candidate region R5 is a deletion candidate, and it is possible to prompt the user to determine whether or not the red-eye correction candidate region R5 should be deleted from the target for red-eye correction. That is, it is possible to prompt the user to determine whether or not the red-eye correction candidate region R5 should be deleted, and to prompt the user to input a deletion command when it is determined that it should be deleted.

  In step S42, the determination unit 71 determines whether or not deletion is designated. That is, the input unit 5 of the digital signal processing unit 60 receives a deletion command by detecting that the user has pressed the deletion button 5c, for example. The input unit 5 supplies a deletion command to the digital signal processing unit 60. The deletion command is supplied to the determination unit 71 via the red-eye detection unit 67 in the digital signal processing unit 60. The determination unit 71 determines that the deletion is specified when the deletion command is received, and determines that the deletion is not specified when the deletion command is not received. If the determination unit 71 determines that deletion is specified, the process proceeds to step S43. If the determination unit 71 determines that deletion is not specified, the determination unit 71 proceeds to step S44.

  In step S43, the red-eye correction unit 69 deletes the information of the red-eye correction candidate area designated for deletion. That is, the determination unit 71 supplies a deletion command to the red-eye correction unit 69. Based on the delete command, the red-eye correction unit 69 refers to a memory (not shown) and deletes information on the selected red-eye correction candidate area. It is assumed that information on all red-eye correction candidate areas is stored in the memory.

  Then, the red-eye correction unit 69 supplies the deleted NG area information to the image display unit 70 together with the image information. The image display unit 70 displays the image GI shown in FIG. 13 on the display device instead of the image GI shown in FIG. 12 based on the image information and the deleted NG area information.

  For example, in FIG. 13, the image display unit 70 displays the highlighted display of the portion corresponding to the red-eye correction candidate region R <b> 5 (shaded line in the drawing). As a result, it is possible to grasp that the red-eye correction candidate region R5 has been deleted from the red-eye correction target.

  In step S44, the determination unit 71 determines whether all NG regions have been selected. If determining unit 71 determines that all NG regions have been selected, the process proceeds to step S45. If determining unit 71 determines that all NG regions have not been selected, the process proceeds to step S41.

  In step S45, the determination unit 71 determines whether or not confirmation is designated. That is, the input unit 5 of the digital signal processing unit 60 receives a confirmation command by detecting that the user has pressed the confirmation button 5d, for example. The input unit 5 supplies a confirmation command to the digital signal processing unit 60. The confirmation command is supplied to the determination unit 71 via the red-eye detection unit 67 in the digital signal processing unit 60. The determination unit 71 determines that the confirmation is designated when the confirmation command is received, and determines that the confirmation is not designated when the confirmation command is not received. If the determination unit 71 determines that confirmation is designated, the process proceeds to step S46 ((1) shown in FIGS. 8 and 7). If it is determined that confirmation is not designated, the process proceeds to step S50 (FIG. 8). , 7 (2)).

  In step S <b> 46, the image display unit 70 displays an area determined to have a success level of “low” (hereinafter referred to as “low area”). That is, the input unit 5 of the digital signal processing unit 60 accepts a display command for a low region by detecting that the user has pressed the selection button 5b, for example. The input unit 5 supplies a display command for a low region to the digital signal processing unit 60. The display command for the low region is supplied to the red-eye correction unit 69 via the red-eye detection unit 67 and the determination unit 71 in the digital signal processing unit 60. The red-eye correction unit 69 selects a low region to be displayed from one or more low regions based on the low region display command. The red-eye correction unit 69 supplies the selected low region information together with the image information to the image display unit 70. The image display unit 70 displays the deletion candidate on the display device based on the image information and the selected low area information. Here, the deletion candidate is a red-eye correction candidate area that is a candidate to be deleted from the red-eye correction target in a plurality of red-eye correction candidate areas.

  For example, in FIG. 13, the image display unit 70 displays a frame RF4 indicating that the red-eye correction candidate region R4 is a low region together with the image GI as information indicating the deletion candidate. Thereby, it is possible to grasp that the red-eye correction candidate region R4 is a deletion candidate, and it is possible to prompt the user to determine whether or not the red-eye correction candidate region R4 should be deleted from the target for red-eye correction. That is, the user can be prompted to determine whether or not the red-eye correction candidate region R4 should be deleted, and the user can be prompted to input a deletion command when it is determined that the red-eye correction candidate region R4 should be deleted.

  In step S47, the determination unit 71 determines whether or not deletion is designated. That is, the input unit 5 of the digital signal processing unit 60 receives a deletion command by detecting that the user has pressed the deletion button 5c, for example. The input unit 5 supplies a deletion command to the digital signal processing unit 60. The deletion command is supplied to the determination unit 71 via the red-eye detection unit 67 in the digital signal processing unit 60. The determination unit 71 determines that the deletion is specified when the deletion command is received, and determines that the deletion is not specified when the deletion command is not received. If the determination unit 71 determines that deletion is specified, the process proceeds to step S48. If the determination unit 71 determines that deletion is not specified, the determination unit 71 proceeds to step S49.

  In step S48, the red-eye correction unit 69 deletes the information of the red-eye correction candidate area designated for deletion. That is, the determination unit 71 supplies a deletion command to the red-eye correction unit 69. Based on the delete command, the red-eye correction unit 69 refers to a memory (not shown) and deletes information on the selected red-eye correction candidate area. It is assumed that information on all red-eye correction candidate areas is stored in the memory.

  Then, the red-eye correction unit 69 supplies the deleted low area information together with the image information to the image display unit 70. The image display unit 70 displays the image GI shown in FIG. 14 on the display device instead of the image GI shown in FIG. 13 based on the image information and the deleted information on the low region.

  For example, in FIG. 14, the image display unit 70 displays the highlighted display (indicated by the slanted line in the drawing) of the portion corresponding to the red-eye correction candidate region R <b> 4. Thereby, it can be understood that the red-eye correction candidate region R4 has been deleted from the target of red-eye correction.

  In step S49, the determination unit 71 determines whether all the low regions have been selected. If the determination unit 71 determines that all the low regions have been selected, the process proceeds to step S45. If the determination unit 71 determines that all the low regions are not selected, the process proceeds to step S41.

  In this embodiment, it is specified whether or not to delete the NG area and the low area, but it may be specified whether or not the middle area is also deleted. By doing so, it is possible to remove the red-eye region with higher accuracy.

  In step S50, the red-eye correction unit 69 determines a red-eye correction candidate area. That is, the determination unit 71 of the digital signal processing unit 60 supplies a confirmation command to the red-eye correction unit 69. Based on the confirmation command, the red-eye correction unit 69 refers to a memory (not shown) and confirms all red-eye correction candidate regions. Then, the red-eye correction unit 69 supplies information on the determined red-eye correction candidate area together with the image information to the image display unit 70. The image display unit 70 displays the image GI shown in FIG. 15 on the display device instead of the image GI shown in FIG. 14 based on the image information and information on the determined red-eye correction candidate area.

  For example, in FIG. 15, the image display unit 70 displays the highlighted display (hatched in the figure) corresponding to the red-eye correction candidate regions R1 to R3 as another highlighted display (lattice hatching in the figure). Is done. Thereby, it can be grasped that the red-eye correction candidate regions R1 to R3 are determined as the targets for red-eye correction.

  By the processing as described above, the image displayed by the image display unit 70 becomes, for example, an image as shown in FIG. In FIG. 15, the portion indicated by the solid diagonal line indicates red-eye position information. The red-eye position information is information indicating the positions of the red-eye correction candidate regions R1, R2, and R3 when the determination unit 71 determines that red-eye detection is successful in the image GI.

  Then, the red-eye correction unit 69 performs red-eye correction on the determined red-eye correction candidate region.

  For example, in FIG. 15, the red-eye correction unit 69 corrects the red-eye correction candidate areas R1 to R3 excluding the red-eye correction candidate areas R4 and R5 specified by the deletion command in the plurality of red-eye correction candidate areas R1 to R5.

  As described above, in the image GI taken with the flash, the input indicating whether or not to delete the red eye correction in the order of the high possibility that the red eye correction has failed (in the order of the low degree of the red eye detection success) is used. Can be encouraged. Thereby, an inappropriate red-eye correction candidate area can be deleted from the target of red-eye correction, and it can be avoided to specify whether or not to perform red-eye correction on all red-eye correction candidate areas. For this reason, for example, red-eye correction can be performed appropriately, and the processing burden on the user side can be reduced.

  For example, when the reliability ranking is finely set as in NG1, NG2, NG3 (assuming that the reliability is low in order from 1), it is not deleted in the order of low reliability. You may make it delete from right to left.

  For example, in the case where the red eye correction candidate area is confirmed and deleted from the enlarged reproduced image, if the deletion target is deleted in the order of low reliability with respect to the image in which the deletion target exists discretely, the display target is displayed every time the deletion is performed. (Delete target) may change from right to left in the entire image. In this case, it is difficult to know how many deletion targets remain in the entire image. Therefore, as described above, for example, by deleting from the right to the left in the screen, it is possible to reduce the deletion residue.

  Note that the digital signal processing unit of the imaging apparatus may include an HPF instead of the BPF 65. Even in this case, the high frequency component of the input signal such as the luminance signal Y can be passed and supplied to the contour emphasizing unit 66.

  An imaging apparatus according to a preferred second embodiment of the present invention will be described with reference to FIGS. The description of the same parts as in the first embodiment will be omitted, and the description will focus on the parts that are different from the first embodiment.

  As shown in FIG. 16, the imaging apparatus 100 has the same basic configuration as that of the first embodiment, but includes a digital signal processing unit 160 instead of the digital signal processing unit 60, and inputs instead of the input unit 5. It differs from 1st Embodiment by the point provided with the part 105. FIG. Specifically, the following points are different.

  That is, based on the image (for example, the image GI shown in FIG. 21) and the red-eye position information (for example, the area indicated by the solid line in FIG. 21) displayed by the user on the input unit 105, A command for changing the red-eye detection threshold TH1 is input.

  As shown in FIG. 17, the digital signal processing unit 160 includes a red-eye detection unit 167 instead of the red-eye detection unit 67, a determination unit 171 instead of the determination unit 71, and a red-eye correction unit instead of the red-eye correction unit 69. 169. The red-eye correction unit 169 receives an instruction to change the red-eye detection threshold value TH1 from the input unit 105 via the red-eye detection unit 167 and the determination unit 171. The red-eye correction unit 169 generates the changed red-eye detection threshold TH1a and passes it to the determination unit 171. The determination unit 171 performs determination using the changed red-eye detection threshold TH1a instead of the red-eye detection threshold TH1.

  Further, as shown in FIGS. 18 and 19, the flow of processing in which the imaging apparatus performs red-eye detection differs from the first embodiment in the following points. In FIG. 18 and FIG. 19, the same processes as those shown in FIG.

  In step S131 ((4) shown in FIGS. 18 and 19), it is determined whether or not a red-eye detection threshold change command has been received. That is, the image display unit 70 displays an image together with the red eye position information. Red-eye position information and images are viewed by the user.

  For example, the image displayed by the image display unit 70 is an image as shown in FIG. In FIG. 21, the part indicated by the solid diagonal line indicates red-eye position information. In the case of FIG. 21, red-eye position information is displayed for the red-eye correction candidate regions R1, R2, R3, and R4 when the determination unit 171 determines that red-eye detection is successful. Here, the red-eye correction candidate regions R1, R2, and R3 are appropriately determined. The red-eye correction candidate region R4 is erroneously determined. A user who has viewed that the red-eye correction candidate area R4 is displayed together with the red-eye position information inputs a change command for the red-eye detection threshold TH1 so as to set it strictly.

  In this way, for example, when a region that should not be corrected for red-eye is erroneously determined, a change command is input so that the red-eye detection threshold TH1 is strictly set.

  Alternatively, for example, the image displayed by the image display unit 70 is an image as shown in FIG. In FIG. 22, the part indicated by the solid diagonal line indicates red-eye position information. In the case of FIG. 22, red-eye position information is displayed for the red-eye correction candidate regions R1 and R2 when the determination unit 171 determines that red-eye detection is successful. Here, the red-eye correction candidate regions R1 and R2 are appropriately determined. The red-eye correction candidate region R3 is a region where red-eye should be detected originally, and is erroneously determined. A user who has viewed that the red-eye correction candidate region R4 is displayed together with the red-eye position information inputs a change command for the red-eye detection threshold TH1 so as to set it loosely.

  In this way, for example, when a region to be corrected for red-eye is erroneously determined, a change command is input so as to reset the red-eye detection threshold TH1 to be loose.

  An instruction to change the red-eye detection threshold TH1 is input to the input unit 105 of the digital signal processing unit 160. The red-eye correction unit 169 receives an instruction to change the red-eye detection threshold value TH1 from the input unit 105 via the red-eye detection unit 167 and the determination unit 171. When the red-eye correction unit 169 receives a command to change the red-eye detection threshold TH1, it determines that a command to change the red-eye detection threshold has been received. If the red-eye correction unit 169 has not received a change command for the red-eye detection threshold TH1, it determines that a red-eye detection threshold change command has not been received.

  When it is determined that a red-eye detection threshold change command has been received, the red-eye correction unit 169 generates a changed red-eye detection threshold TH1a and passes it to the determination unit 171 to perform step S1 ((3) shown in FIGS. 19 and 18). To proceed. If it is determined that a red-eye detection threshold change command has not been received, the process ends.

  Similar to the first embodiment, red-eye correction is reduced for a region that should not be corrected for red-eye in a captured image. Therefore, even with such an imaging apparatus 100, for example, red-eye correction is appropriately performed.

  In step S105, the determination unit 171 performs a success degree determination process. Specifically, as shown in FIG. 20, the success degree determination process differs from the first embodiment in the following points.

  In step S155, when the determination unit 171 receives the changed red-eye detection threshold TH1a, the determination unit 171 performs determination using the changed red-eye detection threshold TH1a instead of the red-eye detection threshold TH1. This is different from step S55 in FIG.

  In this way, for example, when an area that should not be corrected for red-eye is erroneously determined, whether or not the red-eye detection has succeeded with the red-eye detection threshold TH1a that has been strictly set instead of the red-eye detection threshold TH1. It will be judged. This further reduces red-eye correction for areas that should not be red-eye corrected in the captured image.

  In addition, for example, when a region to be corrected for red-eye is erroneously determined, it is determined whether or not the red-eye detection has succeeded based on the red-eye detection threshold TH1a that is set loosely instead of the red-eye detection threshold TH1. become. As a result, it is possible to reduce the fact that red-eye correction is not performed on a region to be corrected for red-eye in a captured image.

  Note that what is input to the input unit of the imaging apparatus by the user may be a change instruction for the face detection threshold TH2 instead of a change instruction for the red-eye detection threshold TH1. Alternatively, what is input to the input unit of the imaging apparatus by the user may be a command to change the red-eye detection threshold TH1 and the face detection threshold TH2 instead of a command to change the red-eye detection threshold TH1.

  Next, a program for realizing the functions of the above-described embodiments will be described.

  Software program code for operating various devices so as to realize the functions of the above-described embodiments and realizing the functions of the above-described embodiments for a computer in an apparatus or system connected to the various devices. Supply (program).

  Also, the scope of the present invention includes those in which the various devices described above are operated in accordance with a program (software or hardware) stored in a computer (CPU or MPU) of the system or apparatus. It is.

  In this case, the software program itself realizes the functions of the above-described embodiment.

  Further, the program itself and means for supplying the program code of the program to a computer, for example, a storage medium storing such a program are also included in the scope of the present invention.

  As a storage medium for storing such a program, for example, a floppy (registered trademark) disk, a hard disk, an optical disk, a magneto-optical disk, a CD-ROM, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

  In addition, the functions of the above-described embodiments are not only realized by the computer executing the supplied program.

  For example, when the program realizes the functions of the above-described embodiment in cooperation with an OS (operating system) running on a computer or another application, the program is also included in the scope of the present invention. .

  Further, the supplied program is stored in a memory provided in a function expansion board of the computer or a function expansion unit connected to the computer.

  Then, based on the instructions of the program, the CPU or the like provided in the function expansion board or function expansion unit performs part or all of the actual processing, and when the functions of the above-described embodiments are realized by the processing, Such a program is included in the category of the present invention.

The block diagram of the imaging device which concerns on suitable 1st Embodiment of this invention. The block diagram of the digital signal processing part in 1st Embodiment. 6 is a flowchart illustrating a flow of processing in which the imaging apparatus performs red-eye detection. The flowchart which shows a red eye correction process. The flowchart which shows a face correction process. The flowchart which shows a success degree determination process. The conceptual diagram which shows a success degree determination table. The flowchart which shows a red eye correction process. The flowchart which shows a red eye correction process. An example of the screen displayed by an image display part. An example of the screen displayed by an image display part. An example of the rear view of an imaging device. An example of the rear view of an imaging device. An example of the rear view of an imaging device. An example of the rear view of an imaging device. The block diagram of the imaging device which concerns on suitable 2nd Embodiment of this invention. The block diagram of the digital signal processing part in 2nd Embodiment. 6 is a flowchart illustrating a flow of processing in which the imaging apparatus performs red-eye detection. 6 is a flowchart illustrating a flow of processing in which the imaging apparatus performs red-eye detection. The flowchart which shows a success degree determination process. An example of the screen displayed by an image display part. An example of the screen displayed by an image display part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,100 Imaging device 5,105 Input part 67,167 Red eye detection part 68,168 Face detection part 69,169 Red eye correction part 70 Image display part 71,171 Determination part

Claims (7)

A plurality of red-eye correction candidate areas that are candidates for red-eye correction are detected from the image, and a red-eye detection reliability indicating the redness of the detected red-eye correction candidate area is set in each of the red-eye correction candidate areas. A red-eye detector to be generated,
A determination unit that determines the degree of successful detection for each of the red-eye correction candidate regions based on at least the red-eye detection reliability;
Based on the degree, a display unit that sequentially displays deletion candidates that are red-eye correction candidate regions that are candidates to be deleted from red-eye correction targets in the plurality of red-eye correction candidate regions, together with the image;
An imaging apparatus comprising: The display unit sequentially displays, together with the image, deletion candidates that are red-eye correction candidate regions that are candidates to be deleted from the red-eye correction target in the plurality of red-eye correction candidate regions in order of decreasing degree. The imaging device according to claim 1. An input unit for inputting a deletion instruction that is an instruction for designating deletion of the deletion candidate;
A red-eye correction unit that corrects red-eye correction candidate areas excluding the red-eye correction candidate area specified by the deletion instruction in the plurality of red-eye correction candidate areas;
The imaging apparatus according to claim 1, further comprising: A plurality of face candidate areas that are areas recognized as the face of the subject are detected from the image, and a face detection reliability indicating the face likeness of the detected face candidate area is generated for each face candidate area Further comprising a face detecting unit,
4. The determination unit according to claim 1, wherein the determination unit determines the degree based on the red-eye detection reliability of the red-eye correction candidate region and the face detection reliability of the face candidate region including the red-eye correction region. The imaging device according to item. A plurality of red-eye correction candidate areas that are candidates for red-eye correction are detected from the image, and a red-eye detection reliability indicating the redness of the detected red-eye correction candidate area is set in each of the red-eye correction candidate areas. A red-eye detection step to generate for
A determination step of determining, for each of the red-eye correction candidate regions, a degree of success in detection based on at least the red-eye detection reliability;
A display step of displaying, together with the image, deletion candidates that are red-eye correction candidate areas that are candidates to be deleted from red-eye correction targets in the plurality of red-eye correction candidate areas in order of decreasing degree;
An imaging method comprising: An imaging program for causing an imaging apparatus to perform the imaging method according to claim 5. A plurality of red-eye correction candidate areas that are candidates for red-eye correction are detected from the image, and a red-eye detection reliability indicating the redness of the detected red-eye correction candidate area is set in each of the red-eye correction candidate areas. A red-eye detector to be generated,
A determination unit that determines the degree of successful detection for each of the red-eye correction candidate regions based on at least the red-eye detection reliability;
Based on the degree, a display unit that sequentially displays deletion candidates that are red-eye correction candidate regions that are candidates to be deleted from red-eye correction targets in the plurality of red-eye correction candidate regions, together with the image;
An image processing apparatus comprising:

Images