JP2006252327A - Image evaluation device, image evaluation method and image evaluation program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily select an image suitable for print without actually printing images by evaluating taken images. <P>SOLUTION: Each image In stored in a storage medium is read in S1. A face extraction part 112a extracts a face area Rn from each image In in S2. An evaluation value calculation part 112d calculates an evaluation value fn showing the sharpness of the face area Rn in S3. Whether all images stored in the storage medium 40 were read in S1 or not is determined, and if an image not read is present, the process is returned to S1 to read the image. When all images were read, the process proceeds to S5. In S5, the evaluation value fn is displayed on a display part 110 with each image In. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus, an image processing method, and an image processing program, and more particularly to an apparatus, method, and program for evaluating an image suitable for printing.

Conventionally, various techniques for supporting selection of an image suitable for printing have been devised. For example, according to Patent Document 1, at least one of image data and image generation history information is analyzed to determine an image quality parameter value related to the image quality based on the image data, and image data is output based on the image quality parameter value. It is determined whether to select the output target.
JP 2004-112596 A

  The technique of Patent Document 1 does not take into consideration how the subject is reflected, and therefore, the subject is not always captured in a state suitable for printing. Although there is a method for making the determination with the naked eye, there are cases where the number of pixels and the resolution of the display for displaying the image are low and it is difficult to determine. Even if the resolution of the display is high, if a large number of images are displayed in a reduced state, it is difficult to discriminate, and it may eventually be necessary to create a print and view the print. SUMMARY An advantage of some aspects of the invention is that it is possible to easily select a photographed image suitable for printing without actually printing the image by evaluating the photographed image. .

  In order to solve the above-described problems, an image evaluation apparatus according to the present invention includes a face extraction unit that extracts a face region including a human face portion from an image, and an evaluation value calculation unit that calculates an evaluation value representing the sharpness of the face region. And a display unit that displays the evaluation value together with the image.

  According to the present invention, the evaluation value representing the sharpness of the face area is calculated, and the evaluation value is displayed together with the image. For this reason, the sharpness of the face area can be easily understood, and it becomes easy to select an image suitable for printing out of images having a person as a subject.

  An image evaluation apparatus according to the present invention includes a face extraction unit that extracts a face region including a human face from an image, an eye region detection unit that detects an eye region where an eye exists from the face region, An eye open / close detection unit that determines whether the eyes are open or closed, and an evaluation value that calculates an evaluation value based on the total number of face areas in which the eyes in the eye area are determined to be open or closed A calculation unit; and a display unit that displays an evaluation value together with the image.

  According to the present invention, it is possible to know from the evaluation value how many eyes the photographed person has opened, and it becomes easy to select an image suitable for printing.

  The evaluation value calculation unit may calculate the evaluation value by dividing the total number of face areas determined to have eyes open in the eye area by the total number of face areas.

  The evaluation value calculation unit may calculate a value obtained by dividing the total number of face regions determined to have closed eyes in the eye region by the total number of face regions as the evaluation value.

  An image evaluation apparatus according to the present invention includes a face extraction unit that extracts a face region including a human face from an image, a coordinate calculation unit that calculates center coordinates between the face regions and a center coordinate of the image, and a face region. An evaluation value calculation unit that calculates an evaluation value based on the central coordinates between the image and the center coordinate of the image, and a display unit that displays the evaluation value together with the image.

  According to the present invention, the balance of the arrangement of the face area can be known from the evaluation value, and it becomes easy to select an image suitable for printing.

  The evaluation value calculation unit may calculate the distance between the center coordinates between the face regions and the center coordinates of the image as the evaluation value.

  The image evaluation method according to the present invention includes a step of extracting a face area including a human face portion from an image, a step of calculating an evaluation value representing the sharpness of the face area, and a step of displaying the evaluation value together with the image. Including.

  The image evaluation method according to the present invention includes a step of extracting a face area including a human face portion from an image, a step of detecting an eye area where an eye exists from the face area, and an eye in the eye area is opened. A step of determining whether the eye in the eye region is open, a step of calculating an evaluation value based on the total number of face regions in which the eyes are determined to be open or closed, and displaying the evaluation value together with the image Steps.

  The image evaluation method according to the present invention includes a step of extracting a face area including a human face portion from an image, a step of calculating center coordinates between the face areas and a center coordinate of the image, and center coordinates between the face areas. And calculating the evaluation value based on the center coordinates of the image, and displaying the evaluation value together with the image.

  The image evaluation program according to the present invention includes a step of extracting a face area including a human face portion from an image, a step of calculating an evaluation value representing the sharpness of the face area, and a step of displaying the evaluation value together with the image. Let the computer run.

  The image evaluation program according to the present invention includes a step of extracting a face area including a human face portion from an image, a step of detecting an eye area where an eye exists from the face area, and an eye in the eye area is opened. A step of determining whether the eye in the eye region is open, a step of calculating an evaluation value based on the total number of face regions in which the eyes are determined to be open or closed, and displaying the evaluation value together with the image And causing the computer to execute the steps.

  The image evaluation program according to the present invention includes a step of extracting a face area including a human face portion from an image, a step of calculating center coordinates between the face areas and a center coordinate of the image, and center coordinates between the face areas. And a step of calculating an evaluation value based on the center coordinates of the image and a step of displaying the evaluation value together with the image.

  According to the present invention, the evaluation value representing the sharpness of the face area is calculated, and the evaluation value is displayed together with the image. For this reason, the sharpness of the face area can be easily understood, and it becomes easy to select an image suitable for printing out of images having a person as a subject.

  Further, according to the present invention, it is possible to know from the evaluation value how many eyes the photographed person has opened, and it becomes easy to select an image suitable for printing.

  Further, according to the present invention, the balance of the arrangement of the face area can be known from the evaluation value, and it becomes easy to select an image suitable for printing.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

<First Embodiment>
FIG. 1 is a functional block diagram of a digital camera 10 according to a preferred embodiment of the present invention. First, in the digital camera 10 shown in FIG. 1, the central processing unit (CPU) 112 is configured to input each button in the digital camera 10 based on input from various buttons and keys of the operation unit 113 including the release switch 22, the mode dial 150, and the like. Supervises and controls the circuit. A program executed by the CPU 112 is stored in the EEPROM 119.

  Now, when the still image shooting mode is set by the mode dial 150, the CPU 112 displays a moving image (through image) on the display unit 110 so that the shooting angle of view can be confirmed. That is, the light that has passed through the imaging lens 14 enters the solid-state imaging device 102a. Photosensors are arranged in a plane on the light receiving surface of the solid-state imaging device 102a, and the subject image formed on the light receiving surface is converted into signal charges of an amount corresponding to the amount of incident light by each photosensor. . The signal charges accumulated in this way are sequentially read out as voltage signals (image signals) corresponding to the signal charges based on pulse signals given from a driver circuit (not shown), and are respectively added to the imaging circuit 102b.

  The imaging circuit 102b includes a gain adjustment circuit, and the obtained image data is sent to the image processing circuit 108 and the high-pass filter (HPF) 125, respectively. The image processing circuit 108 includes an A / D conversion unit 108a, a white balance correction circuit 108b, a gamma correction circuit 108c, a YC processing circuit 108d, a luminance / color difference signal generation circuit, a sharpness correction circuit, a contrast correction circuit, and contour correction for a captured image. The image processing means includes a contour processing unit that performs image processing, a noise reduction processing unit that performs image noise reduction processing, and the like, and processes an image signal in accordance with a command from the CPU 112.

  The image data input to the image processing circuit 108 is converted into a luminance signal (Y signal) and a color difference signal (Cr, Cl signal) and subjected to predetermined processing such as gamma correction, and then stored in the VRAM 132. The

  On the other hand, the image data converted into a digital signal by the A / D conversion unit 108 a of the imaging circuit 108 is extracted only by the G pixel component by the high pass filter (HPF) 125. Then, the G pixel component extracted by the integration processor 126 is integrated and sent to the CPU 112. The CPU 112 calculates an average value of four areas in the vicinity of the center of the subject arbitrarily set with the cross key 81 out of the 64 divided image data for one screen sent from the integration processing unit 126, and auto-focus ( AF) Evaluation value. The AF evaluation value is calculated every elapse of a predetermined period, and is updated and stored in the memory 127 each time it is calculated. The CPU 112 determines the focus of the subject according to the AF evaluation value. The determination method is arbitrary. For example, it is determined whether or not the AF evaluation value has substantially reached the vicinity of the maximum point. If the AF evaluation value has substantially reached the maximum point, the AF evaluation value is determined to be in focus. In this case, it is determined that the subject is not in focus. Alternatively, if the AF evaluation value exceeds a predetermined threshold, it can be determined that the subject is in focus.

  When the captured image is output to the display unit 110 on the monitor, the YC signal is read from the VRAM 132 and sent to the video encoder 134. The video encoder 134 converts the input YC signal into a predetermined display signal (for example, an NTSC color composite video signal) and outputs the converted signal to the display unit 110. The display unit 110 is driven via the driver 111.

  The YC signal of each frame processed at a predetermined frame rate is alternately written in the A area and B area of the VRAM 132, and the Y area other than the area where the YC signal is written out of the A area and B area of the VRAM 132. The YC signal that has been written is read out from this area. In this way, the YC signal in the VRAM 132 is periodically rewritten, and the video signal generated from the YC signal is supplied to the display unit 110, whereby the image being captured is displayed on the display unit 110 in real time. . The user can check the shooting angle of view from the video (through) displayed on the display unit 110.

  Here, when the release switch 22 is half-pressed, AE and AF processes are started. AE / AF processing is performed, and when the release switch 22 is fully pressed, a recording operation is started. The image data acquired in response to the release switch 22 being fully pressed is converted into a luminance / color difference signal (Y / C signal) by the image processing circuit 108 and subjected to predetermined processing such as gamma correction, and then stored in the VRAM 132. Stored.

  The Y / C signal stored in the VRAM 132 is compressed according to a predetermined format by the compression / decompression processing circuit 144 and then recorded as an Exif (Exchangeable Image File Format) file on the recording medium 40 via the media controller 146. The image is recorded in the data part of the Exif file. The CPU 112 records shooting date / time information and the like in a predetermined tag (such as an image description tag) in the header portion of the Exif file.

  The CPU 112 is also connected to a strobe control circuit 115 that controls the light emission of the strobe 18 made of a xenon tube. When still light shooting is detected as having low illuminance, or the light emission is instructed by the strobe button. In this case or when the forced light emission mode is set, the flash 18 is controlled to emit light for a short time (for example, for a short time of 1 second or more) at a timing synchronized with the timing signal generated by the timing generator 120.

  When the moving image shooting mode is set by the mode dial 150, the moving image recording operation starts in conjunction with the full pressing operation of the release switch 22, and when the release switch 22 is pressed again, the moving image recording operation is stopped. The recording operation may be performed while the release switch 22 is continuously pressed, and the recording may be stopped by releasing the pressing. The moving image data is recorded on the recording medium 40 in, for example, a motion JPEG (MPEG) format.

  When the playback mode is selected by the mode dial 150, the compressed data of the last image file (last recorded file) recorded on the recording medium 40 is read. When the file related to the last recording is a still image file, the read image compressed data is decompressed into an uncompressed YC signal via the compression / decompression processing circuit 144 and stored in the VRAM 132. The YC signal stored in the VRAM 132 is added to the video encoder 134. The video encoder 134 creates an NTSC RGB color composite video signal from the input YC signal and outputs it to the display unit 110. Thereby, the frame image of the last frame recorded on the recording medium 40 is displayed on the display unit 110.

  Thereafter, when the right key of the cross key 81 is pressed, the frame is advanced in the forward direction, and when the left key of the cross key 81 is pressed, the frame is advanced in the reverse direction. Then, the frame-positioned image file at the frame position is read from the recording medium 40, and the frame image is reproduced on the display unit 110 in the same manner as described above. If the last frame image is displayed and the frame is advanced in the forward direction, the first frame image file recorded on the recording medium 40 is read and the first frame image is displayed. Played back to the unit 110.

  The display unit 110 corresponds to an external display device connected to an LCD 114, a finder, a video output terminal, or the like built in the camera 10. The CPU 112 includes an OSD signal generation circuit, and the OSD signal generation circuit generates a signal for displaying characters such as a shutter speed, an aperture value, the number of shootable images, a shooting date / time, a warning message, and a symbol such as an icon. A signal output from the OSD signal generation circuit is mixed with an image signal as necessary and supplied to the LCD 114. As a result, a composite image in which characters, icons, and the like are combined with the through image and the playback image is displayed.

  The recording medium 40 is means for holding image data obtained by shooting, and for example, a memory card called smart media is used. The form of the recording medium is not limited to this, and it may be a PC card, a compact flash (registered trademark), a magnetic disk, an optical disk, a magneto-optical disk, a memory stick, etc., electronic, magnetic, optical, or a combination thereof A readable / writable medium can be used in accordance with the above-described method. A configuration in which a plurality of media can be mounted regardless of different types or the same type of recording media may be adopted. The means for storing the image file is not limited to a removable medium that can be attached to and detached from the camera body, but may be a recording medium (built-in memory) built in the camera.

  The CPU 112 is a control circuit that comprehensively controls each circuit of the camera 10, and includes a release switch 22, a cross key 81, a power switch 82, a mode dial 150, an information position designation key 83 or a not-shown strobe button, zoom key, menu / execution button. Based on signals received from the operation unit 113 including others, the operation of the corresponding circuit is controlled, and display control, strobe light emission control, autofocus (AF) control, automatic exposure (AE) control, and the like are performed on the display unit 110. .

  Each time the power supply of the camera 10 is turned on by the power switch 82, power is supplied to each circuit of the digital camera 10 from the main power supply 164 formed of a battery mounted inside the camera body.

  Image data from the image processing circuit 108 is also sent to the photometric processing unit 157. The photometric processing unit 157 is an example of a unit that measures the amount of external light. The photometric processing unit 157 calculates a photometric value (EV value) of subject brightness based on each input image data and the charge accumulation time of the solid-state imaging device 102a, that is, the shutter time of the electronic shutter. Can be configured. Thereby, the drive of the solid-state image sensor 102a is adjusted. The aperture value may be changed with the shutter time of the electronic shutter. When changing the aperture value, the photometric value corresponding to the subject brightness is calculated with the aperture value taken into account. As described above, the photometry processing unit 157 detects the brightness of the subject (subject brightness) by TTL (Through The Lens) photometry using the solid-state imaging device 102a as a light receiving sensor. The photometric processing unit 157 may be a photometric sensor composed of a phototransistor, and is not particularly limited as long as it is a means capable of photometry.

  FIG. 2 shows a program executed by the CPU 112 in a block diagram. The CPU 112 executes a face extraction unit 112a and an evaluation value calculation unit 112d, which are programs. The face extraction unit 112a and the evaluation value calculation unit 112d are stored in the EEPROM 119, and are read by the CPU 112 to the memory 127 as appropriate and executed.

  The face extraction unit 112a extracts a face area that is an area including the face portion of a person from the captured image recorded on the recording medium 40. As a face area extraction method, for example, the technique disclosed in Japanese Patent Laid-Open No. 9-101579 “Face Area Extraction Method and Copying Condition Determination Method” by the present applicant can be applied. This technology determines whether or not the hue of each pixel of the captured image is included in the skin color range and divides it into a skin color region and a non-skin color region, and detects edges in the image to detect each of the images in the image. The location is classified as an edge portion or a non-edge portion. Then, an area composed of pixels classified in the skin color area and classified as a non-edge part and surrounded by pixels determined to be the edge part is extracted as a face candidate area, and the extracted face candidate area becomes a human face. A corresponding region is determined, and a face region is extracted based on the determination result. In addition, a face region can be extracted by a method described in Japanese Patent Laid-Open No. 2003-209683 and Japanese Patent Laid-Open No. 2002-199221.

  The evaluation value calculation unit 112d calculates an evaluation value f for the extracted face area.

  The specific evaluation value f is calculated as follows, for example. The evaluation value calculation unit 112d divides the input luminance signal of the face area into an odd field and an even field, and obtains a delta histogram in each field. That is, first, a blur image is created by applying an averaging filter having a predetermined coefficient to each field of the luminance signal.

  Next, for a certain pixel in each field, a difference value from the surrounding eight pixels is made into a histogram. This is performed for all the pixels in the field to obtain a delta histogram.

  Then, the luminance difference levels 0 to 255 of the delta histogram are cumulatively added to obtain a cumulative delta histogram. This is similarly performed for the input luminance signal. The absolute value of the difference between the cumulative delta histogram of the input luminance signal and the cumulative histogram of the blurred image is averaged at the luminance difference level, and this is used as the evaluation value f (for example, “Analysis and Evaluation of Digital Color Image (University of Tokyo Press) ).

  The calculated f is stored in the memory 127. The evaluation value f is a feature amount representing the sharpness of the face area. The larger f is, the more high-frequency components of the image are, and the contour of the face area is clearer.

  The specific calculation method of the evaluation value f is not limited to the above, and various sharpness indexes obtained using MTF (modulation transfer function) can be used as the evaluation value f (for example, “New Image Analysis Handbook (published by the University of Tokyo) Meeting))).

  The flow of image evaluation processing executed by the CPU 112 will be described below with reference to the flowchart of FIG. In this process, an evaluation value f is calculated and displayed for each image recorded on the recording medium 40, thereby giving a guide as to which image is most suitable for printing among images recorded with people.

  In S 1, each image In recorded in the recording medium 40 is read out, decompressed into an uncompressed YC signal via the compression / decompression processing circuit 144, and stored in the VRAM 132.

  In S2, the face extraction unit 112a extracts a face area Rn from each image In. n is the frame number of each image. For simplification of explanation, it is assumed that there is one face area recorded in each image In, and that one face area Rn is extracted from each image In by the face extraction unit 112a. If no face area can be extracted from the read image, a predetermined message indicating that fact may be displayed on the display unit 110.

  In S3, the evaluation value calculation unit 112d calculates an evaluation value fn representing the sharpness of the face region Rn.

  In S4, it is determined whether or not all the images recorded on the recording medium 40 in S1 have been read. If there is an image that has not been read, the process returns to S1 to read the image. If all the images have been read, the process proceeds to S5.

  In S5, the evaluation value fn is displayed on the display unit 110 together with each image In. FIG. 4 shows how the evaluation values f1 to f3 are displayed together with the images I1 to I3. When all the images In are displayed in one screen, the display size of the image In may be thinned out appropriately according to the display size of the display unit 110. The images In may be sequentially arranged and displayed according to the magnitude relationship of the evaluation values fn.

  Alternatively, only one image In may be displayed on one screen of the display unit 110, the evaluation value fn may be displayed in accordance with the image In, and the display of the evaluation value fn may be switched according to the frame advance of the image In. The user refers to the image displayed on the display unit 110 and the evaluation value, and even if the images are similar, the user can determine which face has the clearest outline and is suitable for printing without printing. If an image inappropriate for printing is deleted from the recording medium 40, the capacity of the recording medium 40 can be saved.

Second Embodiment
The evaluation value f can take various forms other than the value representing the sharpness of the face area. FIG. 5 is a block diagram showing a program executed by the CPU 112 according to the second embodiment. The CPU 112 executes a face extraction unit 112a, an eye region detection unit 112e, an eye open / close detection unit 112f, and an evaluation value calculation unit 112d, which are programs. The face extraction unit 112a extracts a face area from each image In recorded in the recording medium 40, as in the first embodiment. However, the number of persons recorded in each image In is one or more, and the face extraction unit 112a may extract a plurality of face regions from one image.

  The eye area detection unit 112e detects an eye area where an eye exists from the face area extracted from each image In by the face extraction unit 112a. The eye region can be detected from the relative positional relationship of the approximate face center position obtained as a result of face detection. The eye opening / closing detection unit 112f determines whether the eyes in the eye region are open or closed. The method for determining whether to open or close the eyes is not particularly limited. For example, the eye opening / closing detection unit 112f detects the number of black pixels in the vertical direction in two eye regions corresponding to both eyes. Then, based on the number of vertical black pixels in the two eye areas, it is determined whether or not both eyes are open.

  Based on the determination result of the eye opening / closing detection unit 112f, the evaluation value calculation unit 112d totals the total number M1 (n) of face regions determined to have eyes open for each image In. Then, the evaluation value fn is calculated by the following formula.

fn = M1 (n) / M0 (n)
Here, M0 (n) is the total number of face regions extracted from the image In. fn indicates the ratio of persons whose eyes are open among the persons recorded in the image In.

Fn = [M0 (n) -M1 (n) / M0 (n)]
As a result, the percentage of persons who are meditating their eyes may be calculated.

  FIG. 6 is a flowchart of image evaluation processing according to the second embodiment. The CPU 112 according to the second embodiment also performs image evaluation processing in the same flow as in the first embodiment.

  That is, in S 10, each image In recorded in the recording medium 40 is read out, decompressed into an uncompressed YC signal via the compression / decompression processing circuit 144, and stored in the VRAM 132.

  In S11, the face extraction unit 112a extracts a face area from each image In. n is the frame number of each image.

  In S12, the eye region detection unit 112e detects an eye region from the face region extracted from the image In, and the eye opening / closing detection unit 112f determines whether or not the eyes in the detected eye region are open.

  In S13, the evaluation value calculation unit 112d calculates the evaluation value fn = M1 (n) / M0 (n) based on the determination result of the eye opening / closing detection unit 112f.

  In S14, it is determined whether or not all the images recorded on the recording medium 40 in S10 have been read. If there is an image that has not been read, the process returns to S10 to read the image. If all the images have been read, the process proceeds to S15.

  In S15, the evaluation value fn is displayed on the display unit 110 together with each image In. The display format of the evaluation value fn is the same as that of the first embodiment (see FIG. 4).

  The user can refer to the image displayed on the display unit 110 and the evaluation value to determine an image that has the highest percentage of people who are not blinded and is suitable for printing without printing. If an image inappropriate for printing is deleted from the recording medium 40, the capacity of the recording medium 40 can be saved.

<Third Embodiment>
When a plurality of persons are recorded in the image, the evaluation value fn may be calculated from the position of the face area extracted by the face extraction unit 112a.

  FIG. 7 is a block diagram showing a program executed by the CPU 112 according to the third embodiment. The CPU 112 executes a face extraction unit 112a, a coordinate calculation unit 112g, and an evaluation value calculation unit 112d, which are programs. The face extraction unit 112a extracts a face area from each image In recorded in the recording medium 40, as in the first embodiment. However, the number of persons recorded in each image In is one or more, and the face extraction unit 112a may extract a plurality of face regions from one image.

The coordinate calculation unit 112g calculates center coordinates G (n) between the face areas from the positions of the face areas. As an example of the center coordinates between the face areas, the face extraction unit 112a uses the center coordinates a1 (n), a2 (n), ..., al (n), ... between the extracted face areas Rl (n). The coordinates of the center of gravity of the closed region formed by linking are calculated as the center coordinates. For example, as shown in FIG. 8, three face regions R1 (n), R2 (n), and R3 (n) are extracted from an image In, and their center coordinates are set as a1 (n), a2 (n), a3 (n)
G (n) = [a1 (n) + a2 (n) + a3 (n)] / 3.

  As another example of the center coordinates between the face areas, the coordinate calculation unit 112g determines a rectangle having the smallest area among rectangles including all the face areas, and uses the center of gravity of the rectangle as the center coordinates between the face areas. calculate.

  The coordinate calculation unit 112g calculates the center coordinates Cn of the image In. The center coordinates Cn of the image In can be calculated from the number of vertical and horizontal pixels of the image In.

Then, the evaluation value calculation unit 112d calculates the distance between the center coordinate Cn of the image In and the center coordinate G (n) between the face regions as the evaluation value fn of the image In. That is,
fn = | Cn- G (n) |
It is. The smaller the value of fn, the more the face area is recorded in a balanced manner.

  FIG. 9 is a flowchart of the image evaluation process according to the third embodiment.

  That is, in S 20, each image In recorded on the recording medium 40 is read out, decompressed into an uncompressed YC signal via the compression / decompression processing circuit 144, and stored in the VRAM 132.

  In S21, the face extraction unit 112a extracts a face region Rl (n) from each image In. n is a frame number of each image, and l is a serial number assigned to the face area.

  In S22, the coordinate calculation unit 112g calculates the center coordinate G (n) between the face regions Rl (n) and the center coordinate Cn of the image In.

  In S23, the evaluation value calculation unit 112d calculates the distance between the center coordinates Cn and the center of gravity G (n) of the image In as the evaluation value fn of the image In.

  In S24, it is determined whether or not all the images recorded on the recording medium 40 in S20 have been read. If there is an image that has not been read, the process returns to S20 to read the image. If all the images have been read, the process proceeds to S25.

  In S25, the evaluation value fn is displayed on the display unit 110 together with each image In. The display format of the evaluation value fn is the same as that of the first embodiment (see FIG. 4).

  The user can refer to the image displayed on the display unit 110 and the evaluation value to determine an image having a good face area arrangement balance and suitable for printing without printing. If an image inappropriate for printing is deleted from the recording medium 40, the capacity of the recording medium 40 can be saved.

  Note that the evaluation value calculation unit 112d may separately calculate the evaluation values disclosed in the first to third embodiments, or may calculate some or all of these combinations simultaneously. In this case, other programs executed by the CPU 112 are all or part of the face extraction unit 112a, the eye region detection unit 112e, the eye open / close detection unit 112f, and the coordinate calculation unit 112g according to the evaluation value to be calculated. .

<Fourth embodiment>
The CPU 112 that executes all or part of the face extraction unit 112a, the evaluation value calculation unit 112d, the eye region detection unit 112e, the eye opening / closing detection unit 112f, and the coordinate calculation unit 112g may not be included in the digital camera. Various image processing apparatuses may be provided.

  FIG. 10 is a front view of the storefront printing apparatus 70 capable of executing the above program. In this storefront printing apparatus 70, a customer inserts a recording medium 40 (not shown) such as smart media into an insertion slot 89 and inserts money, and the digital image stored in the recording medium 40 is printed on a printing medium such as TA paper. It is a machine to do. The printing device 70 is a storefront information terminal installed at a storefront of various stores such as a photo shop, a bookstore, and a supermarket, and is normally used alone by a customer who purchases a print without the assistance of a store clerk.

  This storefront printing apparatus 70 is provided with a liquid crystal type display unit 110 having a touch panel 98 stacked on the front surface. A customer operates the touch panel 98 while looking at the display unit 110 to determine the print size, the number of prints, and the like. Enter the required information. One or more printers 90 (not shown) are installed inside the storefront printing apparatus 70, and each printer 90 prints a digital image. The storefront printing apparatus 70 of this embodiment is provided with a maximum of four printers 80. An outlet 83 through which the printed printing medium is discharged is provided on the front outside of the storefront printing apparatus 70, and the coin slot 88 and the coin machine 88 return coins to the coin machine 88 to be described later. A return port 86 is provided.

  FIG. 11 is a block diagram showing a hardware configuration of the storefront printing apparatus 70.

  When the customer inserts the recording medium 40 into the media insertion port 89, the hard disk controller 74 loads the control software recorded on the hard disk (HDD) 75 into the program memory 71. Image data is recorded on the recording medium 40 in a compressed state. When the CPU 112 instructs the media reader controller 82 to read out the image data, the required image data is acquired from the recording medium 40. The image processing unit 81 decompresses the image data and restores the original image data. Further, the image processing unit 81 performs necessary image processing and stores the image data in the work memory 76.

  Necessary information such as the print size and the number of prints is converted into an electrical signal corresponding to the position when the customer presses a desired position on the touch panel 98 and is input to the CPU 112 via the touch panel interface unit 97. The input information is displayed on the display unit 110. The display unit 110 is operated by the display control unit 78. The coin machine 88 converts the amount displayed on the display unit 110 into an electrical signal based on the information input from the touch panel 98 and inputs it to the CPU 112 via the coin machine I / F unit 87. The CPU 112 confirms whether or not the customer has entered a necessary amount. When the necessary amount is input, the print data processing unit 91 creates print job data from the image data stored in the work memory 76, and the print job data is sent to each printer 90 via the printer I / F unit 84. send. The CPU 112 performs a series of these controls.

  Each printer 90 employs a printing system called a TA system, and uses TA paper having a characteristic that color is developed when heat is applied and fixing is performed when light of a predetermined wavelength is irradiated.

  When the printing is completed, a receipt describing the amount inserted into the coin machine 88 and the amount of change returned is output from the receipt printer 92. The contents to be printed by the receipt printer 92 are instructed by the CPU 112 via the receipt printer I / F unit 93.

  The face extraction unit 112a, the eye region detection unit 112e, the eye open / close detection unit 112f, the coordinate calculation unit 112g, and the evaluation value calculation unit 112d, which are programs executed by the CPU 112, are recorded in the HDD 75 and loaded into the program memory 71. Executed.

  In the storefront printing apparatus 70 having such a configuration, when an image brought in by the customer using the recording medium 40 is printed, it can be known from the evaluation value displayed on the display unit 110 which image is most suitable for printing. it can.

<Fifth Embodiment>
The face extraction unit 112a, the evaluation value calculation unit 112d, the eye region detection unit 112e, the eye open / close detection unit 112f, and the coordinate calculation unit 112g in the first to fourth embodiments are programs executed by the general-purpose CPU 112. It is also possible to adopt a configuration in which a part or all of the program is implemented by a semiconductor integrated circuit (ASIC, Application Specific Integrated Circuit) dedicated to image processing. This is preferable because the load on the CPU 112 is reduced.

1 is a block diagram of a digital camera according to a preferred embodiment of the present invention. The block diagram which shows the program which CPU which concerns on 1st Embodiment performs The flowchart which shows the flow of the image evaluation process which concerns on 1st Embodiment. The figure which shows the example of a display of the evaluation value displayed with the image The block diagram which shows the program which CPU which concerns on 2nd Embodiment performs The flowchart which shows the flow of the image evaluation process which concerns on 2nd Embodiment. The block diagram which shows the program which CPU which concerns on 3rd Embodiment performs The figure which shows an example of the center coordinate between face areas The flowchart which shows the flow of the image evaluation process which concerns on 3rd Embodiment. The front view of the shop front printing apparatus which concerns on preferable embodiment of this invention 1 is a block diagram of a storefront printing apparatus according to a preferred embodiment of the present invention.

Explanation of symbols

40: recording medium, 110: display unit, 112: CPU, 112a: face extraction unit, 112d: evaluation value calculation unit, 112e: eye area detection unit, 112f: eye open / close detection unit, 112g: coordinate calculation unit

Claims (12)

A face extraction unit that extracts a face area including a human face from an image;
An evaluation value calculation unit for calculating an evaluation value representing the sharpness of the face area;
A display unit for displaying the evaluation value together with the image;
An image evaluation apparatus comprising: A face extraction unit that extracts a face area including a human face from an image;
An eye area detection unit for detecting an eye area in which an eye exists from the face area;
An eye opening / closing detection unit for determining whether the eye in the eye region is open or closed;
An evaluation value calculation unit that calculates an evaluation value based on the total number of face regions in which the eyes in the eye region are determined to be open or closed;
A display unit for displaying the evaluation value together with the image;
An image evaluation apparatus comprising:   The image evaluation apparatus according to claim 2, wherein the evaluation value calculation unit calculates, as an evaluation value, a number obtained by dividing the total number of face regions determined to have eyes in the eye region divided by the total number of face regions.   The image evaluation apparatus according to claim 2, wherein the evaluation value calculation unit calculates, as an evaluation value, a number obtained by dividing the total number of face regions determined to have eyes closed in the eye region by the total number of face regions. A face extraction unit that extracts a face area including a human face from an image;
A coordinate calculation unit that calculates center coordinates between the face regions and center coordinates of the image;
An evaluation value calculation unit that calculates an evaluation value based on the center coordinates between the face regions and the center coordinates of the image;
A display unit for displaying the evaluation value together with the image;
An image evaluation apparatus comprising:   The image evaluation apparatus according to claim 5, wherein the evaluation value calculation unit calculates a distance between a central coordinate between the face regions and a central coordinate of the image as an evaluation value. Extracting a face area including a human face from an image;
Calculating an evaluation value representing the sharpness of the face area;
Displaying the evaluation value together with the image;
An image evaluation method including: Extracting a face area including a human face from an image;
Detecting an eye area where eyes are present from the face area;
Determining whether eyes in the eye area are open or closed;
Calculating an evaluation value based on the total number of face regions in which the eyes in the eye region are determined to be open or closed;
Displaying the evaluation value together with the image;
An image evaluation method including: Extracting a face area including a human face from an image;
Calculating center coordinates between the face regions and center coordinates of the image;
Calculating an evaluation value based on center coordinates between the face regions and center coordinates of the image;
Displaying the evaluation value together with the image;
An image evaluation method including: Extracting a face area including a human face from an image;
Calculating an evaluation value representing the sharpness of the face area;
Displaying the evaluation value together with the image;
An image evaluation program for causing a computer to execute. Extracting a face area including a human face from an image;
Detecting an eye area where eyes are present from the face area;
Determining whether eyes in the eye area are open or closed;
Calculating an evaluation value based on the total number of face regions in which the eyes in the eye region are determined to be open or closed;
Displaying the evaluation value together with the image;
An image evaluation program for causing a computer to execute. Extracting a face area including a human face from an image;
Calculating center coordinates between the face regions and center coordinates of the image;
Calculating an evaluation value based on center coordinates between the face regions and center coordinates of the image;
Displaying the evaluation value together with the image;
An image evaluation program for causing a computer to execute.

Images