JP2016051302A - Image processor, imaging device, image processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate an image in which appropriate image processing is subjected to an image area representing skin of a subject.SOLUTION: An image processor 100 includes an acquisition part 101 for acquiring a target image, a detection part 102 for detecting a reference area being an image area representing a specific part of the subject from the target image, a prediction part 103 for acquiring a prediction value of the size of an image area representing skin of the subject included in the target image on the basis of the size of the reference area, an actual measurement part 104 for acquiring the actual measurement value of the size of a correction object area being an image area having a pixel value being the same as or similar to the image area representing the skin of the subject included in the target image, a determination part 105 for determining correction intensity by comparing the actual measurement value with the prediction value, and a processing part 106 for generating a processed image subjected to image processing with the correction intensity in the correction object area in the target image.SELECTED DRAWING: Figure 3

Description

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

  A technique is known in which image processing is performed on an image region representing the skin of a subject person included in an image to generate an image with improved impression of the subject person. For example, Patent Document 1 discloses an imaging device that performs a skin beautification process on a target image.

  When performing skin beautification using the imaging apparatus of Patent Document 1, skin beautification processing is executed for all image regions that have a color and brightness similar to the skin of the subject person included in the target image. For this reason, when the target image includes an image of the subject person and an image of a background (for example, flooring) having a color or lightness approximate to the skin of the subject person, not only the image of the subject person but also the background image Skin care is also applied to the image. As a result, by applying the skin beautification process to the entire target image, an unfavorable image with a white and soft impression is generated.

  On the other hand, the image processing apparatus of Patent Document 2 determines the correction strength of various image processing including skin beautification processing by taking into account the ratio of the face image area of the subject person to the entire target image, and thereby the background image. An excessive image processing is applied to the image area to prevent the resolution of the entire image from being lowered more than necessary.

JP 2007-280291 A JP 2010-049312 A

  The image processing apparatus of Patent Document 2 prevents the generation of an unfavorable image by executing image processing with lower correction strength as the ratio of the face image area of the subject person to the entire target image is smaller. However, for example, even when the face image area of the subject person occupies a large proportion of the entire target image, if the background image has a color or brightness that approximates the skin of the subject person, It is necessary to prevent generation of an unfavorable image by suppressing the correction strength of the skin beautifying process. Even when the face image area of the subject person occupies only a small percentage of the entire target image, it is sufficient if the background image does not have a color or lightness that approximates the skin of the subject person. It is desirable to execute image processing with the correction intensity. For this reason, it has been required to appropriately select the correction strength depending on whether the background image has a color or brightness that approximates the skin of the subject person.

  The present invention has been made in view of the above problems, and an object of the present invention is to generate an image obtained by performing appropriate image processing on an image region representing the skin of a subject.

In order to achieve the above object, an image processing apparatus according to the present invention provides:
An acquisition unit for acquiring a target image;
A detection unit that detects a reference region, which is an image region representing a specific part of the subject, from the target image;
A prediction unit that obtains a predicted value of the size of the image region representing the skin of the subject included in the target image based on the size of the reference region;
An actual measurement unit that acquires an actual measurement value of the size of the correction target area, which is an image area having the same or similar pixel value as the image area representing the skin of the subject, included in the target image;
A determination unit that compares the measured value with the predicted value to determine a correction strength;
A processing unit that generates a processed image obtained by performing image processing with the correction intensity on the correction target region of the target image;
Comprising
It is characterized by that.

  According to the present invention, it is possible to generate an image obtained by performing appropriate image processing on an image region representing the skin of a subject.

1 is a block diagram illustrating a configuration of an imaging apparatus according to a first embodiment. It is a figure which shows an example of a target image. 3 is a block diagram illustrating a functional configuration of an imaging apparatus and an image processing apparatus according to the first embodiment. FIG. 4 is a flowchart for explaining skin beautification processing executed by the image processing apparatus according to the first embodiment. It is a figure which shows an example of a correction | amendment intensity | strength table. It is a figure which shows the other example of a correction | amendment intensity | strength table. It is a figure which shows the other example of a correction | amendment intensity | strength table. It is a figure which shows the other example of a target image. 10 is a flowchart for explaining skin beautification processing executed by the image processing apparatus according to the second embodiment.

(Embodiment 1)
Hereinafter, an image processing apparatus, an imaging apparatus, and an image processing method according to the present invention will be described in detail using the imaging apparatus 1 and the image processing apparatus 100 as examples with reference to the drawings. In the drawings, the same or equivalent parts are denoted by the same reference numerals.

  As illustrated in FIG. 1, the imaging apparatus 1 includes an imaging unit 10, a data processing unit 20, and a user interface unit 30.

  The imaging unit 10 includes an optical lens 11 and an image sensor 12. The imaging unit 10 generates a target image TG illustrated in FIG. 2 by imaging a subject person in accordance with a user operation received by the operation unit 32 described later. The target image TG is an image to be subjected to skin beautification processing, and is an image including an image of a subject person as shown in FIG.

  The optical lens 11 includes a lens that collects light emitted from a subject, and a peripheral circuit for adjusting imaging setting parameters such as focus, exposure, and white balance.

  The image sensor 12 includes, for example, a CCD (Charge Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor), and the like. The image sensor 12 acquires an optical image of a subject formed by the optical lens 11 condensing light, and the voltage information of the acquired optical image is converted into digital image data by an analog / digital converter (not shown). Convert to Then, the obtained digital image data is stored in the external storage unit 23 described later as a target image TG.

  The data processing unit 20 includes a main storage unit 21, an output unit 22, an external storage unit 23, a CPU (Central Processing Unit) 24, and an image processing apparatus 100.

  The main storage unit 21 includes, for example, a RAM (Random Access Memory). The main storage unit 21 functions as a work memory for the CPU 24, and temporarily stores various image data and programs including processed images generated by image processing.

  The output unit 22 reads out image data stored in the main storage unit 21 or the external storage unit 23, and RGB (R (Red, red), G (Green, green), B (Blue, blue) corresponding to the image data. )) A signal is generated and output to the display unit 31 described later. Further, the output unit 22 converts the generated RGB signal into an HSV (H (Hue), hue), S (Saturation, saturation), lightness (Value)) signal, and supplies the signal to the CPU 24 and the image processing apparatus 100.

  The external storage unit 23 includes a non-volatile memory (for example, a flash memory or a hard disk), and stores various programs including a control program necessary for controlling the entire imaging apparatus 1 and various fixed data. The external storage unit 23 supplies the stored program and data to the CPU 24 and an image processing apparatus 100 described later, and includes a target image TG generated by the imaging unit 10, a processed image supplied from the image processing apparatus 100, and the like. Various image data are stored.

  The CPU 24 controls the entire imaging apparatus 1 by executing a control program stored in the external storage unit 23 and executes various programs stored in the external storage unit 23.

  The image processing apparatus 100 generates a processed image by performing a skin beautification process on the target image TG including the image of the subject person. The image processing apparatus 100 is controlled by the CPU 24, and functions as the acquisition unit 101, the detection unit 102, the prediction unit 103, the actual measurement unit 104, the determination unit 105, and the processing unit 106 illustrated in FIG.

  Note that the image processing apparatus 100 has a trimming function, an edge detection function, and the like, as in a normal image processing apparatus, but in the following, the target image TG including the image of the subject person, which is characteristic of the present embodiment, has a beautiful skin. The description will focus on the function of processing.

  The acquisition unit 101 acquires the target image TG generated by the imaging unit 10 and stored in the external storage unit 23.

  The detection unit 102 detects a reference area from the target image TG and acquires a feature amount of the detected reference area. The reference region is an image region representing a predetermined part of the whole body of the subject person, and the prediction unit 103 described later acquires a predicted value of the size of the image representing the skin of the subject person included in the target image TG. Used as a reference. In the present embodiment, the detection unit 102 detects a face image area (face area) of the subject person as a reference area from the target image TG. Since the technique for detecting a face image area from an image is well known in the technical field, a detailed description is omitted.

  In the present embodiment, the detection unit 102 acquires the size of the face image region and the pixel value of the face image region as the feature amount of the detected face image region (reference region).

  Specifically, the detection unit 102 acquires the face frame size (FrameSize) as the size of the detected face image area. Here, as shown in FIG. 2, the face frame size is an area of a rectangular area circumscribing the detected face image area (for example, the number of pixels in the vertical direction × the number of pixels in the horizontal direction). In FIG. 2, since the face frame size is not actually displayed in the target image TG, it is indicated by a one-dot chain line.

  The detection unit 102 acquires the hue, saturation, and brightness of the pixels that constitute the detected face image area as the pixel values of the face image area. The detected face image area is composed of a plurality of pixels having different pixel values. The detection unit 102 acquires the pixel value of each pixel constituting the face image area, and acquires the mode value of the summed pixel values as the pixel value of the face image area.

  The prediction unit 103 acquires a predicted value (SkinBaseFinish) of the size of an image area representing the skin of the subject person included in the target image TG.

  Specifically, the prediction unit 103 acquires the SkinBase stored in the external storage unit 23 and includes it in the target image TG based on the SkinBase and the size of the face image area detected by the detection unit 102. A predicted value (SkinBaseFinish) of the size of the image area representing the skin of the subject person is acquired. SkinBase is reference data for acquiring SkinBaseFinish.

  The external storage unit 23 has the size (area or ratio) occupied by the face image area of the person model and the size occupied by the image area representing the skin of the person model in the whole body image of the person model (the subject person as a model). Model data including, is acquired from an external device and stored in advance as SkinBase. SkinBase analyzes a plurality of whole body images of a subject person as a human model, and sums the size of the face image area of the subject person and the size of the image area representing the skin of the subject person included in each whole body image, etc. It may be prepared in advance by this method. In this embodiment, SkinBase is created in advance using a person with a relatively high degree of skin exposure as a person model. The process of obtaining the predicted value of the size of the image area representing the skin of the subject person included in the target image TG executed by the prediction unit 103 will be described later in detail with reference to the flowchart of FIG.

  The actual measurement unit 104 calculates an actual measurement value of the skin color amount of the target image TG. Here, the skin color amount of the target image TG is the size of the skin color region included in the target image TG. Here, the skin color region has a pixel value (hue, saturation, brightness) in an HSV color space (a color space composed of three components of hue (Hue), saturation (Saturation), and brightness (Value)). This is an image area composed of pixels included in a predetermined area including pixel values of the skin image of the subject person. The skin color region is a correction target region that is a target of skin beautification processing executed by the processing unit 106 described later.

  Specifically, the actual measurement unit 104 creates a skin color map of the target image TG, and acquires an integrated value of the skin color map as an actual measurement value of the skin color amount of the target image TG. Here, the skin color map has a predetermined pixel value (hue, saturation, brightness) including the pixel value of the face image area acquired by the detection unit 102 in the HSV color space among the pixels constituting the target image TG. It is an image composed only of pixels included in a region. The predetermined area is, for example, a predetermined rectangular parallelepiped area centered on the pixel value of the face image area. The predetermined rectangular parallelepiped region may be set in advance by an arbitrary method. Specifically, the skin image of a plurality of subject persons is analyzed, and the pixel value of each pixel constituting the skin image and the pixel value of the skin image that is the mode value of the summed pixel values of each pixel The rectangular parallelepiped region can be set as appropriate by obtaining the correlation between.

  The determination unit 105 includes an actual measurement value of the skin color amount of the target image TG acquired by the actual measurement unit 104, a predicted value of the size of the image area representing the skin of the subject person included in the target image TG acquired by the prediction unit 103, Are compared, and the correction strength of the skin beautifying process executed by the processing unit 106 to be described later is determined.

  Specifically, the determination unit 105 determines whether or not the actual measurement value acquired by the actual measurement unit 104 exceeds the prediction value acquired by the prediction unit 103 by exceeding a predetermined threshold (for example, 1000 pixels × pixel). Determine. The predetermined threshold value is the difference between the measured value of the skin color amount of the target image TG and the predicted value of the size of the image area representing the skin of the subject person included in the target image TG, and the image generated by the skin beautification process. What is necessary is just to obtain | require the correlation with preference by arbitrary methods, such as experiment, and to set suitably. When the determination unit 105 determines that the actual measurement value does not exceed the predicted value by exceeding a predetermined threshold value, the determination unit 105 selects the first correction strength as the strength of the skin beautifying process. On the other hand, when the determination unit 105 determines that the actual measurement value exceeds the predicted value by exceeding a predetermined threshold, the determination unit 105 selects the second correction strength that is weaker than the first correction strength as the strength of the skin beautifying process.

  The processing unit 106 generates a processed image by performing skin beautification processing using the above-described skin color region included in the target image TG as a correction target region with the intensity determined by the determining unit 105.

  Here, the skin beautification process is a process for generating a preferable image by performing filter processing, exposure correction, saturation adjustment, and the like on the target image TG. Since the skin beautifying process is a technique well known in the technical field, a detailed description thereof is omitted. The processing unit 106 adjusts the strength of the skin beautifying process by adjusting the strength of each process such as filter processing, exposure correction, and saturation adjustment included in the skin beautifying process. For example, it is assumed that the skin beautifying process with an intensity of 100% is set in advance to include exposure correction for changing the exposure by +1.0 EV, and the determining unit 105 determines that the intensity of the beautifying process is 50%. In this case, the processing unit 106 performs exposure correction for changing the exposure by (+1.0 EV) × (50/100) = + 0.5 EV as part of the skin beautifying process with 50% intensity. Similarly, the processing unit 106 determines each process included in the skin beautification process based on the intended correction strength and the intensity of each process included in the skin softening process having a preset strength of 100%. By executing each with strength, the skin beautification processing with the desired strength is executed.

  Returning to FIG. 1, the user interface unit 30 includes a display unit 31, an operation unit 32, and an external interface 33.

  The display unit 31 includes, for example, an LCD (Liquid Crystal Display), a CRT (Cathode Ray Tube), an organic EL (Electro Luminescence) display, and the like, and is imaged by the imaging unit 10 based on RGB signals supplied from the output unit 22. Various images including the target image TG stored in the external storage unit 23 and the processed image generated by the image processing apparatus 100 performing skin beautification on the target image TG are displayed.

  The operation unit 32 receives an operation instruction from the user. The operation unit 32 includes various operation buttons such as a power switch of the image pickup apparatus 1, a shutter button, and buttons for selecting various functions of the image pickup apparatus 1. When receiving an operation instruction from the user, the operation unit 32 supplies the received instruction information to the imaging unit 10, the CPU 24 of the data processing unit 20, and the like.

  In addition, the display part 31 and the operation part 32 may be comprised by what is called a touch panel arrange | positioned mutually superimposed.

  The external interface 33 is an interface for exchanging data with an external device of the imaging apparatus 1. For example, the external interface 33 converts a processed image generated by the image processing apparatus 100 by performing the skin beautification process on the target image TG into data of USB (Universal Serial Bus) standard, and external equipment via the USB cable. Send and receive data to and from.

  Next, an operation in which the imaging apparatus 1 and the image processing apparatus 100 having the above-described configuration perform the skin beautification process on the target image TG including the image of the subject person will be described with reference to the flowchart of FIG.

  The imaging unit 10 included in the imaging device 1 generates a target image TG illustrated in FIG. 2 by imaging a subject person in advance. The external storage unit 23 acquires and stores the target image TG generated by the imaging unit 10 in advance.

  When the user desires to perform a skin beautification process on the target image TG, the user develops the target image TG in the main storage unit 21 by operating the operation unit 32, and sets one of a plurality of operation modes included in the imaging apparatus 1. “Beautiful skin treatment mode” is selected.

  When the operation unit 32 receives an operation for selecting the “skin-beautification processing mode” by the user, the CPU 24 reads out a program for executing the processing of the flowchart of FIG. The data is expanded in the storage unit 21. The display unit 31 acquires the target image TG illustrated in FIG. 2 from the external storage unit 23 and displays it.

  In such a state, when the user instructs the start of the skin beautifying process using the operation unit 32, the image processing apparatus 100 starts the skin beautifying process illustrated in the flowchart of FIG.

  First, the acquisition unit 101 acquires the target image TG stored in the external storage unit 23 (step S101).

  Next, the detection unit 102 detects a face image area as a reference area from the target image TG acquired in step S101 using any known technique (step S102).

  Next, the detection unit 102 acquires the size of the face image region (reference region) detected in step S102 and the pixel value of the face image region as the detected feature amount of the reference region (step S103). Specifically, the detection unit 102 acquires the area (face frame size, FrameSize) of the rectangular area circumscribing the face image area detected in step S102 as the size of the face image area. In addition, the detection unit 102 totals the pixel values (hue, saturation, and brightness) of each pixel constituting the face image area detected in step S102, and the mode value of the summed pixel values is calculated for this face image area. Is obtained as a pixel value.

  The prediction unit 103 acquires SkinBase (reference data for acquiring SkinBaseFinish) from the external storage unit 23, and includes it in the target image TG based on the SkinBase and the size of the face image area acquired in step S102. A predicted value (SkinBaseFinish) of the size of the image area representing the skin of the subject person is acquired (step S104).

  In the present embodiment, the SkinBase acquired from the external storage unit 23 includes the size (for example, 10000 pixels × pixel) of the face image area of the person model included in the whole body image of the person model (the subject person to be a model). The size of the image area representing the skin of the person model (for example, 120,000 pixels × pixel). Here, for example, it is assumed that the size of the face image area acquired in step S103 and detected in step S102 is 20000 pixels × pixel. In this case, the prediction unit 103 uses the SkinBase of the size of the face image area (20000 pixels × pixel) acquired in step S102 to the size of the image area (120,000 pixels × pixel) representing the skin of the person model indicated by SkinBase. Is multiplied by the ratio (20000 pixels × pixel / 10000 pixels × pixel) to the size (10000 pixels × pixel) of the face image area of the human model, and SkinBaseFinish = (120,000 pixels × pixel) × 2 = 4800000 pixels × Get the pixel.

  Next, the actual measurement unit 104 acquires an actual measurement value of the skin color amount of the target image TG (step S105). The actual measurement unit 104 detects the pixel value (hue, saturation, and lightness) of each pixel that configures the target image TG in step S103 obtained in step S103 in the HSV color space. A skin color map which is an image composed only of pixels included in a predetermined region including the pixel value of the face image region is created, and an integrated value of the skin color map is acquired as an actual measurement value of the skin color amount of the target image TG.

  The determination unit 105 determines that the measured value of the skin color amount of the target image TG acquired in step S105 is larger than the predicted value of the size of the image area representing the skin of the subject person included in the target image TG acquired in step S104. It is determined whether or not it is larger (step S106). If the determining unit 105 determines that the actually measured value is not larger than the predicted value (step S106; No), the determining unit 105 selects the first correction strength as the strength of the skin beautifying process (step S107), and proceeds to step S110.

  If it is determined that the actually measured value is larger than the predicted value (step S106; Yes), the determining unit 105 next determines whether or not the difference between the actually measured value and the predicted value is greater than a predetermined threshold value. (Step S108). If the determination unit 105 determines that the difference between the actually measured value and the predicted value is not greater than the predetermined threshold (Step S108; No), the determination unit 105 selects the first correction strength as the strength of the beautifying process (Step S107). Proceed to step S110.

  If the determination unit 105 determines that the difference between the actually measured value and the predicted value is larger than the predetermined threshold (step S108; Yes), the second correction strength that is weaker than the first correction strength is used as the strength of the skin beautifying process. Select (step S109) and proceed to step S110.

  The processing unit 106 generates a processed image by performing a skin beautification process on the skin color area that is the correction target area in the target image TG with the correction strength selected in Step S107 or Step S109 (Step S110). The generated processed image is displayed on the display unit 31 and stored in the external storage unit 23, and the skin beautification process is terminated.

  As described above, the imaging apparatus 1 according to the present embodiment uses the actual value of the skin color amount of the target image TG as the predicted value of the area of the image area representing the skin of the subject person included in the target image TG. When the value exceeds the range, the correction strength of the image processing (skin-beautifying process) is reduced. For this reason, it is possible to generate a processed image in which appropriate image processing is performed on an image region representing the skin of the subject person while suppressing a reduction in the resolution of the entire target image TG.

  In the present embodiment, the image processing apparatus 100 is provided inside the imaging apparatus 1. The image processing apparatus 1 may be an apparatus independent from the imaging apparatus 1. For example, an information processing apparatus such as a computer can function as the image processing apparatus 100. In this case, the image processing apparatus 100 may acquire the subject image captured by the external imaging apparatus as the target image TG and perform the above-described skin beautification process.

  In the present embodiment, the imaging apparatus 1 acquires an image generated by the imaging unit 10 capturing an image of a subject person and stored in the external storage unit 23 from the external storage unit 23 as a target image TG. The imaging device 1 may acquire the target image TG directly from the imaging unit 10 instead of from the external storage unit 23. The imaging device 1 may acquire and store the target image TG in advance from an external image input device (for example, a digital camera, a memory card, or a network).

  In the present embodiment, the imaging device 1 detects a face image area as a reference area from the target image TG. The imaging apparatus 1 can select any image area other than the face image area as long as it is an image area useful for estimating the size (area or ratio) of the image area representing the skin of the subject person in the entire target image TG. It can be detected as a reference area. For example, an image area representing a part other than the face of the subject person (for example, hands or ears), an image area representing a part of the face of the subject person (for example, eyes or nose), or the like as the reference area from the target image TG. It may be detected.

  In the present embodiment, the imaging apparatus 1 determines the area (vertical size (number of vertical pixels) × horizontal size (number of horizontal pixels)) of a rectangular area circumscribing the detected face image area (reference area) as the face image. Obtained as the size of the area. Not only the rectangular area, but also the area of an arbitrary shape (for example, an ellipse) circumscribed to the detected face image area can be acquired as the size of the face image area.

  Note that the imaging apparatus 1 may acquire a feature amount other than the area indicating the size of the detected face image region (reference region). For example, the vertical size (number of pixels in the vertical direction) or horizontal size of the rectangular area circumscribing the face image area, the major axis (number of pixels in the major axis direction) or minor axis of the ellipse area circumscribing the face image area, etc. Can be obtained as The imaging device 1 can also acquire the ratio of the detected face image area to the entire target image TG as a feature amount indicating the size of the face image area.

  Note that when a plurality of face image areas are detected from the target image TG, the imaging device 1 totals the sizes (areas or ratios) of the face image areas, and the mode value, maximum value, minimum value, The average value, the weighted average value, the median value, and the like set so that the weight increases as the distance from the center of the target image TG becomes closer can be acquired as the size of the detected face image region.

  In the present embodiment, the imaging device 1 acquires the hue, saturation, and brightness of each pixel that constitutes the detected face image area as the pixel value of the face image area. The imaging device 1 can acquire arbitrary pixel values without being limited to hue, saturation, and brightness. For example, the imaging device 1 may acquire the RGB signal value of each pixel constituting the detected face image area as the pixel value of this face image area.

  In the present embodiment, the imaging apparatus 1 totals the pixel values (hue, saturation, and brightness) of each pixel constituting the detected face image area, and detects the mode value of the totaled pixel values. Obtained as the pixel value of the region. The imaging device 1 can acquire a statistical value other than the mode value as the pixel value of the face image area. For example, a weighted average that is set so that the weight increases as it approaches a predetermined value that is set by any method such as analyzing the maximum value, minimum value, average value, and skin images of a plurality of subject persons. A value, a median value, or the like may be acquired as a pixel value of the detected face image area.

  In addition, when the imaging apparatus 1 detects a plurality of face image areas from the target image TG, the pixel values of the detected face image areas are totalized by summing up the pixel values of the constituent pixels for all the face image areas. Can be acquired. Further, after the pixel values of the constituent pixels are totaled for each face image area and the pixel values of each face image area are individually determined, the ratio of the largest (smallest) face image area or the entire target image TG It is also possible to obtain the pixel value of the face image area having the largest (smaller) as the pixel value of the detected face image area.

  In the present embodiment, the imaging apparatus 1 acquires the predicted value of the area of the image area that represents the skin of the subject person included in the target image TG. The imaging device 1 may acquire a predicted value of the ratio of the image area representing the skin of the subject person included in the target image TG to the entire target image TG.

  In this embodiment, SkinBase is the size (area or ratio) of the face image area of the person model and the image area representing the skin of the person model included in the whole body image of the person model (the subject person who is the model). Model data including size. Arbitrary model data that serves as a reference for predicting the size of the image area representing the skin of the subject person included in the target image TG can be adopted as the SkinBase.

  For example, in the whole body image of a person model, the size (area or ratio) occupied by an image area representing a portion (for example, ear, eye, hand) other than the face of the person model and the image area representing the skin of the person model are Model data including the occupied size (area or ratio) may be used as the SkinBase.

  Alternatively, SkinBase may be a function representing the correlation between the size of the face image area included in the whole body image of the person model and the size of the image area representing the skin. For example, empirically, the size of the image area representing the skin included in the whole body image of an arbitrary subject person is about 12 times the size of the face image area of the subject person included in the whole body image. Are known. Accordingly, as the SkinBase, 12 which is a typical ratio of the size of the image area representing the skin to the size of the face image area is stored in the external storage unit 23, and the size of the face image area detected from the target image TG is set. By multiplying this ratio, it is possible to acquire a predicted value of the size of the image area representing the skin of the subject person included in the target image TG.

  In the present embodiment, SkinBase is created in advance using a subject person with a relatively high degree of skin exposure as a person model. The imaging apparatus 1 can store a plurality of types of SkinBases created in advance based on a plurality of person models having different physiques and skin exposure levels. The imaging apparatus 1 uses any one of a plurality of types of SkinBases to acquire SkinBaseFinish (a predicted value of the size (area or ratio) of an image region representing the skin of the subject person included in the target image TG). . The selection of SkinBase may be performed by the user operating the operation unit 32, or the imaging device 1 may detect the feature amount (for example, pixel value) of the detected face image area acquired by the detection unit 102. May be based on. For example, if it is determined that the skin color of the subject person is dark based on the pixel value of the face image area of the subject person, the skin of the subject person may be tanned because he likes clothes with high skin exposure Since it is high, it is good also as selecting SkinBase created based on a person model with high skin exposure.

  When the imaging apparatus 1 detects a plurality of face image areas from the target image TG, the size (area) of the image area representing the skin of the subject person included in the target image TG is considered in consideration of the number of detected face image areas. (Or percentage) can be obtained. When a plurality of face image areas are detected, the target image TG includes a plurality of subject persons. The more the subject person included in the target image TG, the larger the size (area or ratio) of the image area representing the skin of the subject person included in the target image TG. Specifically, the imaging device 1 acquires a predicted value of the size of an image area representing the skin per person to be included in the target image TG, and then detects the detected face image as the predicted value per person. The predicted value (SkinBaseFinish) of the size of the image area representing the skin of the subject person included in the target image TG may be acquired by multiplying the number of areas (that is, the number of subject persons).

  Note that the imaging apparatus 1 can estimate the number of subject persons based on an image representing a part other than the face of the subject person detected from the target image TG, and can also obtain a predicted value using the estimated number of persons. . For example, when the imaging apparatus 1 detects an image area representing the eyes or ears of the subject person from the target image TG, the imaging apparatus 1 estimates the number of the subject person by dividing the number of detected image areas by 2, and this estimation The predicted value can be acquired using the number of people.

  In the present embodiment, the skin color amount of the target image TG is the area of the skin color region included in the target image TG. The skin color amount may be the ratio of the skin color area included in the target image TG to the entire target image TG.

  In the present embodiment, the skin color area has a pixel value (hue, saturation, brightness) as an HSV color space (a color space composed of three components of hue (Hue), saturation (Saturation), and brightness (Value)). The image area is composed of pixels included in a predetermined rectangular parallelepiped area centered on the pixel value of the skin image of the subject person. In the skin color area, pixel values other than hue, saturation, and lightness (for example, RGB pixel values) are pixel values of the skin image of the subject person in a color space other than the HSV color space (for example, RGB color space). It may be a pixel included in a predetermined region at the center.

  Note that the predetermined area set in the color space is not limited to a rectangular parallelepiped area centered on the pixel value of the skin image of the subject person, and has an arbitrary shape such as a spherical shape or a quadrangular prism shape, for example. It may be a region. Further, the pixel value of the skin image of the subject person only needs to be included in the predetermined area, and may be located at a place other than the center.

  In the present embodiment, the imaging apparatus 1 acquires the pixel values (hue, saturation, and brightness) of the face image area detected by the detection unit 102, and the pixel values in the target image TG are in the HSV color space. Create an image (skin color map) composed only of pixels included in a predetermined area including the pixel value of the acquired face image area, and integrate the skin color map to integrate the size of the skin color area that is the correction target area ( Actual value of skin color amount was acquired. The imaging apparatus 1 can also acquire an actual measurement value of the size (skin color amount) of the correction target region by this other method. For example, the imaging apparatus 1 only includes pixels in the target image TG in which pixel values (for example, hue value, saturation value, and brightness value) are included in a predetermined region in a color space (for example, HSV color space). An integrated value of this image may be acquired as an actual measured value of the skin color amount. The predetermined region may be set as appropriate so as to represent the skin color of a typical subject person by an arbitrary method such as analyzing skin images of a plurality of subject persons.

  In the present embodiment, the imaging apparatus 1 predicts the area of an image region that represents the skin of a subject person that is included in the target image TG and the actual measurement value of the skin color region that is the correction target region included in the target image TG. When the value does not exceed the predetermined area or more, the first correction intensity is selected as the correction intensity for the skin beautifying process, and when the actual measurement value exceeds the predicted value beyond the predetermined area, the first correction intensity is weaker. The second correction strength was selected as the correction strength for skin beautification. A predicted value of the size (area or ratio) of the image area representing the skin of the subject person included in the target image TG and an actual measurement value of the size (area or ratio) of the skin color area included in the target image TG that is the correction target area. Any comparison result between and can be adopted as a condition for determining the correction strength. For example, the correction strength may be determined based on whether or not the actual measurement value exceeds the predicted value by a predetermined ratio or more.

  Note that the imaging apparatus 1 can also determine the correction strength based on conditions that change according to the attributes of the target image TG. For example, based on whether or not the actual measurement value of the skin color amount of the target image TG exceeds the predicted value of the size (area or ratio) of the image area representing the skin of the subject person included in the target image TG above a threshold value. When determining the correction strength, this threshold value may be set as a function of the ratio of the face image area to the entire target image TG, and the threshold value may be set higher as the ratio increases. When the face image area, that is, the image area that the user originally desires to perform skin beautification occupies a sufficiently large ratio in the entire target image TG, the skin color area included in the target image TG is used as the correction target area and the strong correction strength Even if the skin beautification process is performed at, most of the effect of the skin beautification process extends to the face image area that is the original purpose. For this reason, the possibility that an undesired processed image is generated is low. Therefore, image processing can be performed with an appropriate correction strength by using the threshold as a function of the ratio of the face image area. For example, the threshold value may be set to be lower as the brightness (lightness) of the target image TG is higher. This is because if the target image TG having high brightness is subjected to the skin-beautifying process or the whitening process using the skin color area as the correction target area, an extremely whitish and undesirable image is likely to be generated. Note that the threshold value determination method may be appropriately set by an arbitrary method such as obtaining a correlation between the attribute of the target image TG, the threshold value, and the preference of the generated image through experiments.

  Note that the imaging apparatus 1 can select one of the two or more correction intensities as the image processing correction intensity. Specifically, the imaging apparatus 1 may determine the correction intensity with reference to the correction intensity table TB1 illustrated in FIG. The correction strength table TB1 is stored in, for example, the external storage unit 23, and stores the correction strength and the condition for determining the correction strength (comparison result between the actual measurement value and the predicted value) in association with each other. For example, when the actual measurement value of the skin color amount of the target image TG exceeds the predicted value of the size (area or ratio) of the image area representing the skin of the subject person included in the target image TG by 10% to 15% The apparatus 1 acquires the correction intensity (90%) of the skin beautifying process corresponding to the comparison result from the correction intensity table TB1 as the correction intensity of the skin beautifying process.

  Note that the imaging apparatus 1 can perform various image processes on the target image TG in addition to the skin beautification process described above. For example, the imaging apparatus 1 performs a whitening process for whitening the skin color of the subject person, a wrinkle / stain removal process for removing wrinkles or spots on the skin of the subject person, a redness reduction process for reducing eye redness, and a skin color. Blood color improvement treatment that improves whiteness, whitening treatment that whitens yellowed teeth, eye opening treatment that opens eyes wide, cheek lift-up treatment that makes fat cheeks and cheeks that are stretched slender, and removes red eye condition when flashing Image processing such as flash red-eye removal processing can be executed. Since these image processes are well-known techniques in the technical field, a detailed description thereof will be omitted.

  When performing multiple types of image processing on the target image TG, the imaging apparatus 1 determines the correction strength of each image processing with reference to the correction strength table TB2 illustrated in FIG. 6 and the correction strength table TB3 illustrated in FIG. Can do. The correction strength tables TB2 and TB3 are stored in the external storage unit 23, and both store the correction strength and the condition for determining the correction strength (comparison result of the actual measurement value and the predicted value) in association with each other. ing. In the correction intensity table TB2, as shown in FIG. 6, different conditions are set for two types of image processing (skin-beautifying process and whitening process). For this reason, the imaging apparatus 1 that determines the correction strength of the image processing with reference to the correction strength table TB2 can determine the correction strength for each image processing. For example, in the correction intensity table TB2, as shown in FIG. 6, the measured value of the skin color amount of the target image TG is the size (area or ratio) of the image area representing the skin of the subject person included in the target image TG. The correction strength of the skin beautifying process is the same (100%) when the predicted value is 0% to 5% above and 6% to 9% above, but the correction strength of the whitening process is different (100% and 90%). %). Since the correction strength for each image processing is determined individually, the correction strength can be determined finely for each image processing, and a preferable image can be generated by performing image processing with an appropriate correction strength. On the other hand, in the correction intensity table TB3, conditions common to the two types of image processing are set as shown in FIG. For this reason, the imaging device 1 that determines the correction strength of the image processing with reference to the correction strength table TB3 can collectively determine the correction strengths of the plurality of image processing. For example, when the actual measurement value exceeds 6% to 9%, the imaging apparatus 1 that determines the correction intensity with reference to the correction intensity table TB3 of FIG. 7 corrects the correction intensity (95%) of the skin beautification process and the whitening process. The intensity (90%) is acquired from the correction intensity table TB3 in a lump. By collectively determining the correction strengths for a plurality of image processes, the processing load can be reduced. In the above-described correction intensity tables TB1 to TB3, it is possible to set image processing executed under specific conditions and image processing not executed by setting the correction intensity of at least some image processing to 0%. it can.

  In the present embodiment, after determining the correction strength, the imaging apparatus 1 directly performs a skin beautification process on the target image TG with the determined correction strength, thereby obtaining a processed image that has undergone the skin correction process with the determined correction strength. Generated. The imaging apparatus 1 can also generate a processed image that has been subjected to skin treatment with the determined correction strength by a method other than the above-described method that directly performs skin beautification processing with the determined correction strength on the target image TG. For example, the imaging apparatus 1 weights and synthesizes the target image TG and an image obtained by applying a skin beautification process to the target image TG with a predetermined correction strength (for example, 100%) at an appropriate ratio. A processed image that has been subjected to skin beautification processing at the correction intensity of the period may be generated. In this case, the correction strength of the skin beautification process directly applied to the target image TG by the imaging apparatus 1 is always constant (in this case, 100%). The imaging apparatus 1 does not use the correction strength of the skin beautification process applied to the target image TG, but at what ratio the target image and the image obtained by applying the skin beautification process to the target image with a predetermined correction intensity are weighted and combined. Then, it is determined based on the comparison result between the predicted value of the size (area or ratio) of the image region representing the skin of the subject person included in the target image TG and the actual measured value of the skin color amount of the target image TG. Note that the weighted synthesis may be performed on the entire two images, or only a part of the two images (for example, a portion detected as a skin color region).

(Embodiment 2)
The imaging apparatus 1 and the image processing apparatus 100 according to the first embodiment include images representing the skin of a subject person included in the target image TG based on the detected size of the face image area (reference area) and the person model. The predicted value of the area size was obtained. The predicted value can also be obtained by other methods.

  Hereinafter, based on the detected size and arrangement information of the face image area and the person model, a predicted value of the size of the image area representing the skin of the subject person included in the target image TG is acquired, and skin beautification processing is executed. Functions and operations of the imaging apparatus 1 and the image processing apparatus 100 that perform the operations will be described.

  The imaging apparatus 1 and the image processing apparatus 100 according to the present embodiment have substantially the same functional configuration as the imaging apparatus 1 and the image processing apparatus 100 according to the first embodiment. However, the functions of the detection unit 102 and the prediction unit 103 included in the image processing apparatus 100 according to the present embodiment are partly different from the functions of the detection unit 102 and the prediction unit 103 of the image processing apparatus 100 according to the first embodiment.

  The detection unit 102 detects the face image area from the target image TG as a reference area, and the detected face image area (reference area) size, the pixel value of the face image area, and arrangement information of the face image area Is acquired as a feature amount of the detected reference region. The arrangement information is information representing a position occupied by the detected face image area in the target image TG. In the present embodiment, the arrangement information includes the position of the face image area in the target image TG (for example, position coordinates in orthogonal coordinates with the center of the target image TG as the origin) and the orientation of the face image area (for example, the face image area). The direction of the vector from the nose to the mouth included in the information).

  The prediction unit 103 includes the size of the detected face image region (reference region) acquired by the detection unit 102, SkinBase (reference value for acquiring SkinBaseFinish), and the whole body image of the subject person. Based on the estimated ratio of the portion included in the target image TG, a predicted value (SkinBaseFinish) of the area of the image representing the skin of the subject person included in the target image TG is acquired.

  Hereinafter, the operations of the imaging apparatus 1 and the image processing apparatus 100 according to the present embodiment will be described in detail with a focus on the operations of the detection unit 102 and the prediction unit 103 with reference to the flowchart of FIG. 9.

  The image processing apparatus 100 included in the imaging apparatus 1 according to the present embodiment performs processing that is substantially the same as the processing shown in the flowchart of FIG. 4, and performs skin beautification processing on the target image TG illustrated in FIG. However, the image processing apparatus 100 according to the present embodiment executes the processes of steps S201 to S205 of the flowchart of FIG. 9 instead of the processes of steps S103 and S104 of the flowchart of FIG.

  The detection unit 102 detects a face image region as a reference region from the target image TG illustrated in FIG. 8 by the same process as step S102 in the flowchart of FIG.

  The detection unit 102 acquires the size of the detected face image area (reference area), the pixel value of the face image area, and the arrangement information of the face image area as feature amounts (step S201). Specifically, the detection unit 102 acquires the face frame size as the size of the detected face image area. Here, as shown in FIG. 8, the face frame size (FrameSize) is an area of the rectangular area circumscribing the detected face image area (vertical pixel number × horizontal pixel number). In addition, the detection unit 102 totals the pixel values (in this embodiment, hue, saturation, and brightness) of each pixel that constitutes the detected face image area, and calculates the mode value of the pixel value of this face image area. Obtained as a pixel value. Further, the detection unit 102 determines the position coordinates of the detected face image area in the orthogonal coordinates with the center of the target image TG as the origin, and the direction of the vector from the nose to the mouth included in the face image area. Acquired as arrangement information of the face image area.

  The prediction unit 103 acquires SkinBase ′ based on the SkinBase acquired from the external storage unit 23 and the size of the detected face image area acquired in Step S201 (Step S202).

  In this embodiment, SkinBase is a size of the face image area of the person model in the whole body image of the seven-headed human model (model subject person), the size of the image area representing the skin of the person model, It is data including. The prediction unit 103 adds the size of the detected face image area (for example, 20000 pixels ×) acquired in step S201 to the size (for example, 120,000 pixels × pixel) of the image area representing the skin of the human model indicated by SkinBase. SkinBase ′ = 2 by multiplying the ratio (eg, 20000 pixels × pixel / 10000 pixels × pixel) to the size (eg, 10000 pixels × pixel) of the face image area of the person model shown in the person model * 120,000 pixels * pixel = 24,000 pixels * pixel is acquired.

  As described above when the processing in step S104 in the flowchart of FIG. 4 is described, the image processing apparatus 100 according to the first embodiment uses the SkinBaseFinish (target image) based on the SkinBase and the size of the detected face image area. A predicted value of the size (area or ratio) of the image region representing the skin of the subject person included in the TG was acquired. In contrast, the image processing apparatus 100 according to the present embodiment acquires SkinBase ′ instead of SkinBaseFinish based on SkinBase and the size of the face image area. This is because the image processing apparatus 100 according to the first embodiment performs the skin beautification process on the target image TG including the whole body image of the subject person shown in FIG. 2, whereas the image according to the present embodiment. This is because the processing apparatus 100 performs the skin beautification process on the target image TG including only a part of the whole body image of the subject person shown in FIG. The image processing apparatus 100 according to the present embodiment considers the ratio of the portion included in the target image TG in the whole body image of the subject person, and the size of the image area representing the skin of the subject person included in the target image TG. Get the predicted value of. Therefore, the image processing apparatus 100 according to the present embodiment is first included in the target image TG in the case where all the whole body images of the seven human bodies are included in the target image TG by the process in step S202. The predicted value of the size of the image area representing the skin of the subject person is acquired as SkinBase ′. Then, by performing the processing of steps S203 to S205 in the flowchart of FIG. 9 to be described later, by adjusting SkinBase ′ by the ratio of the portion included in the target image TG in the whole body image of the subject person, SkinBaseFinish (a predicted value of the size of the image area representing the skin of the subject person included in the target image TG) is acquired.

  The prediction unit 103 acquires the head size based on the size of the face image area acquired in step S201 (step S203). As shown in FIG. 8, the head size (HeadSize) is an area of a rectangular region circumscribing the head region (head image region) of the subject person (for example, the number of vertical pixels × the number of horizontal pixels). Specifically, the prediction unit 103 acquires the head size by substituting the face frame size (FrameSize) acquired as the size of the face image area in step S201 into a predetermined function. Here, the predetermined function may be set as appropriate by obtaining the correlation between the face frame size and the head size in advance by an arbitrary method such as an experiment. For example, it is empirically known that the head size can be obtained by multiplying the face frame size by 1.4.

The prediction unit 103 acquires the head and body (HeadAndBody) based on the detected face image area arrangement information acquired in step S201 and the head size acquired in step S203 (step S204). The head and body (HeadAndBody) is the ratio of the size of the portion included in the target image TG to the head size (HeadSize) in the whole body image of the subject person. The relationship between the head and body (HeadAndBody) and the head size (HeadSize) is expressed by the following equation (1).
HeadAndBody = 1 + (BodySize / HeadSize) = (HeadSize + BodySize) / HeadSize (1)
Here, BodySize (body size) is a part of the whole body image of the subject person that is included in the target image TG, excluding the head region (head image region), as shown in FIG. is there. In FIG. 8, the face frame size (FrameSize), the head size (HeadSize), and the body size (BodySize) are not actually displayed in the target image TG, and thus are indicated by alternate long and short dash lines. Specifically, the prediction unit 103 uses the person model in which the subject person is employed in SkinBase based on the arrangement information of the face image area acquired in step S202 (the position and orientation of the face image area in the target image TG). Similarly, assuming that there are seven heads and bodies, the size of the part included in the target image TG in the whole body image of the subject person is estimated. By dividing the estimated size (that is, HeadSize + BodySize) by the head size (HeadSize), the head and body (HeadAndBody) can be obtained as shown in Equation (1).

  The prediction unit 103 assumes that the target image TG includes all the whole body images of the subject person who is seven heads and bodies similar to the SkinBase ′ acquired in Step S202, that is, the person model used in the SkinBase. Based on the predicted value of the size of the image area representing the skin of the subject person included in the target image TG and the head and body (HeadAndBody) acquired in step S204, SkinBaseFinish (the subject person included in the target image TG) (Predicted value of the size of the image region representing the skin) is acquired (step S205).

Specifically, the prediction unit 103 acquires SkinBaseFinish using the following equation (2).
SkinBaseFinish = SkinBase ′ × (HeadAndBody / 7) (2)
Here, HeadAndBody / 7 is the ratio of the head and body (HeadAndBody) acquired in step S204 to the head and body of the person model used in the SkinBase used in step S202 (7 heads and bodies in this embodiment). . For example, assume that 4 is acquired as the head and body in step S204. This means that four images of the subject person's whole body image are included in the target image TG. In this case, in accordance with Equation (2), the prediction unit 103 uses the SkinBase ′ acquired in Step S202 as the head of the person model adopted in the SkinBase used in Step S202 of the head acquired in Step S204 (the book In the embodiment, SkinBaseFinish is obtained by multiplying 4/7, which is a ratio to 7 heads and bodies).

  After acquiring the predicted value in step S205, the image processing apparatus 100 executes processing similar to that in steps S105 to S110 in the flowchart of FIG. 4, and generates a processed image by performing skin beautification processing on the target image TG.

  As described above, the imaging apparatus 1 and the image processing apparatus 100 according to the present embodiment are included in the target image TG in consideration of the ratio of the portion included in the target image TG in the whole body image of the subject person. The predicted value of the size of the image area representing the skin of the subject person is acquired. When the actual measurement value of the skin color area (skin color amount) included in the target image TG, which is the correction target area, exceeds the acquired predicted value beyond a predetermined range, the imaging apparatus 1 according to the first embodiment. Similarly to the image processing apparatus 100, an image suitable for an image region representing the skin of the subject while reducing the correction strength of the image processing (skin-beautifying process) and suppressing a reduction in resolution of the entire target image TG. A processed image that has been processed can be generated.

  In the present embodiment, the imaging apparatus 1 has a head that is a ratio of the size of the portion included in the target image TG in the whole body image of the subject person to the size of the head image of the subject person (head size (HeadSize)). The predicted value of the size of the image area representing the skin of the subject person included in the target image TG was acquired in consideration of the body (HeadSize). The imaging apparatus 1 can acquire a predicted value in consideration of an arbitrary value indicating the proportion of a part included in the target image TG in the whole body image of the subject person, not limited to the head and body. For example, the imaging apparatus 1 considers the ratio of the size (area or ratio) of the portion included in the target image TG in the whole body image of the subject person to the size of the whole body image of the subject person, and calculates the predicted value. You may get it.

  In the present embodiment, the imaging apparatus 1 determines the position coordinates of the face image area and the direction of the vector from the nose to the mouth included in the face image area in orthogonal coordinates with the center of the target image TG as the origin. Acquired as area location information. The imaging apparatus 1 can acquire any other information as arrangement information as long as it is information indicating the positional relationship between the whole body image of the subject person and the frame of the target image TG. For example, the imaging apparatus 1 may acquire, as arrangement information, the position coordinates of the face image area in orthogonal coordinates with one corner of the target image TG as the origin, the orientation of the vector from the mouth to the nose included in the face image area, and the like. Good.

  In this embodiment, the person model (subject person to be a model) employed when creating the SkinBase is seven heads and bodies. A SkinBase can be created in advance using a human model of an arbitrary head and body, not limited to seven heads and bodies, and can be stored in the external storage unit 23 of the image processing apparatus 100.

  Note that the imaging apparatus 1 stores in advance a plurality of types of SkinBases created using a plurality of human models having different heads and bodies, and obtains a SkinBaseFinish using any one of the plurality of types of SkinBases. You can also In this case, the selection of SkinBase may be performed by the user operating the operation unit 32, or the imaging device 1 may perform the selection based on an arbitrary feature amount acquired by the detection unit 102 from the face image region. . For example, the imaging apparatus 1 may estimate the age of the subject person from the feature amount (for example, the number of wrinkles) of the detected face image area, and select SkinBase based on the estimated age. When the subject person is a child, there is a high possibility that the subject person has a smaller head and body than an adult. Therefore, it is preferable to obtain the SkinBaseFinish using a SkinBase created based on a person model having a smaller head and body. By selecting SkinBase based on the estimated age, the size of the image area representing the skin of the subject person included in the target image TG can be predicted more accurately. When the imaging device 1 detects a plurality of face image areas from the target image TG, the age of the subject person estimated from each face image area is totaled, and the mode value, maximum value, minimum value, and average of the estimated age are totaled. Value, median, weighted average value that increases in weight as the estimated age of the face image area closer to the center of the target image TG, estimated age obtained from the face image area having the maximum size (area or ratio), etc. It is also possible to select SkinBase.

  In the above-described embodiment, the image processing apparatus, the imaging apparatus, and the image processing method according to the present invention have been described using the imaging apparatus 1 and the image processing apparatus 100 as examples. The image processing apparatus, the imaging apparatus, and the image processing method according to the present invention can be realized by an arbitrary electronic device such as a computer, a mobile phone, a digital camera, or a PDA (Personal Digital Assistance).

  Specifically, a program for causing a computer, a mobile phone, a digital camera, a PDA, etc. to operate as an imaging apparatus and an image processing apparatus according to the present invention is recorded on a recording medium (for example, a memory card or An image pickup apparatus and an image processing apparatus according to the present invention are realized by storing, distributing, and installing on a CD-ROM (Compact Disc Read-Only Memory), DVD-ROM (Digital Versatile Disc Read-Only Memory), and the like. be able to.

  Alternatively, the above program is stored in a storage device (for example, a disk device) included in a server device on a communication network such as the Internet, and the computer, mobile phone, digital camera, PDA, or the like downloads this program. You may implement | achieve the imaging device and image processing apparatus which concern on this invention.

  Further, when the functions of the imaging apparatus and the image processing apparatus according to the present invention are realized by cooperation or sharing of an operating system (OS) and an application program, only the application program part is stored in a recording medium or a storage device. May be.

  Further, the application program may be superimposed on a carrier wave and distributed via a communication network. For example, an application program may be posted on a bulletin board (BBS: Bulletin Board System) on a communication network, and the application program may be distributed via the network. Then, the imaging apparatus and the image processing apparatus according to the present invention may be realized by installing and starting up the application program in a computer and executing the application program in the same manner as other application programs under the control of the OS.

  As mentioned above, although preferable embodiment of this invention was described, this invention is not limited to the specific embodiment which concerns, This invention includes the invention described in the claim, and its equivalent range It is. The invention described in the scope of claims at the beginning of the present application will be appended.

(Appendix 1)
An acquisition unit for acquiring a target image;
A detection unit that detects a reference region, which is an image region representing a specific part of the subject, from the target image;
A prediction unit that obtains a predicted value of the size of the image region representing the skin of the subject included in the target image based on the size of the reference region;
An actual measurement unit that acquires an actual measurement value of the size of the correction target area, which is an image area having the same or similar pixel value as the image area representing the skin of the subject, included in the target image;
A determination unit that compares the measured value with the predicted value to determine a correction strength;
A processing unit that generates a processed image obtained by performing image processing with the correction intensity on the correction target region of the target image;
Comprising
An image processing apparatus.

(Appendix 2)
The detection unit acquires the number of subjects included in the target image based on the number of the reference regions detected from the target image,
The prediction unit obtains the predicted value in consideration of the number of persons;
The image processing apparatus according to appendix 1, wherein:

(Appendix 3)
The reference area is a face image area of a subject.
The image processing apparatus according to appendix 1 or 2, characterized in that:

(Appendix 4)
Stores model data for obtaining the predicted value, which is created in advance based on the size of the image area representing the specific part of the subject to be the model and the size of the image area representing the skin of the subject to be the model A storage unit
The prediction unit obtains the prediction value in consideration of the model data stored in the storage unit;
The image processing apparatus according to any one of supplementary notes 1 to 3, wherein

(Appendix 5)
The predicting unit estimates a ratio of a portion included in the target image of the whole body image of the subject, and acquires the predicted value in consideration of the estimated ratio;
The image processing apparatus according to any one of appendices 1 to 4, wherein

(Appendix 6)
The prediction unit estimates the ratio in consideration of arrangement information in the target image of the reference region,
The image processing apparatus according to appendix 5, characterized in that:

(Appendix 7)
The prediction unit estimates the ratio in consideration of the age of the subject person estimated based on the reference region,
The image processing apparatus according to appendix 5 or 6, characterized by the above.

(Appendix 8)
The determination unit, when the measured value exceeds the predicted value exceeding a predetermined threshold, compared to the intensity determined when the measured value does not exceed the predicted value exceeding the predetermined threshold A small intensity is determined as the corrected intensity,
The image processing apparatus according to any one of appendices 1 to 7, wherein

(Appendix 9)
An imaging unit that generates a target image by imaging a subject;
The image processing apparatus according to any one of appendices 1 to 8, which acquires a target image generated by the imaging unit;
An imaging apparatus comprising:

(Appendix 10)
An acquisition step of acquiring a target image;
A detection step of detecting a reference region, which is an image region representing a specific part of the subject, from the target image;
A prediction step of obtaining a predicted value of the size of the image area representing the skin of the subject included in the target image based on the size of the reference area;
An actual measurement step of acquiring an actual measurement value of the size of the correction target area, which is an image area having the same or similar pixel value as the image area representing the skin of the subject, included in the target image;
A determination step of determining the correction strength by comparing the measured value and the predicted value;
A processing step of generating a processed image obtained by performing image processing with the correction intensity on the correction target region of the target image;
Comprising
An image processing method.

(Appendix 11)
Computer
An acquisition unit for acquiring a target image;
A detection unit that detects a reference region that is an image region representing a specific part of the subject from the target image;
A prediction unit that obtains a predicted value of the size of the image area representing the skin of the subject included in the target image based on the size of the reference area;
An actual measurement unit that acquires an actual measurement value of the size of the correction target region that is an image region that has the same or similar pixel value as the image region that represents the skin of the subject included in the target image,
A determination unit that compares the measured value with the predicted value to determine a correction strength,
A processing unit that generates a processed image obtained by performing image processing with the correction intensity on the correction target region of the target image;
A program characterized by functioning as

DESCRIPTION OF SYMBOLS 1 ... Imaging device, 10 ... Imaging part, 11 ... Optical lens, 12 ... Image sensor, 20 ... Data processing part, 21 ... Main memory part, 22 ... Output part, 23 ... External storage part, 24 ... CPU, 30 ... User Interface unit 31 ... Display unit 32 ... Operation unit 33 ... External interface 100 ... Image processing apparatus 101 ... Acquisition unit 102 ... Detection unit 103 ... Prediction unit 104 ... Measurement unit 105 ... Decision unit 106 ... Processing unit, TG ... Target image, TB1 to TB3 ... Correction intensity table

Claims (11)

An acquisition unit for acquiring a target image;
A detection unit that detects a reference region, which is an image region representing a specific part of the subject, from the target image;
A prediction unit that obtains a predicted value of the size of the image region representing the skin of the subject included in the target image based on the size of the reference region;
An actual measurement unit that acquires an actual measurement value of the size of the correction target area, which is an image area having the same or similar pixel value as the image area representing the skin of the subject, included in the target image;
A determination unit that compares the measured value with the predicted value to determine a correction strength;
A processing unit that generates a processed image obtained by performing image processing with the correction intensity on the correction target region of the target image;
Comprising
An image processing apparatus. The detection unit acquires the number of subjects included in the target image based on the number of the reference regions detected from the target image,
The prediction unit obtains the predicted value in consideration of the number of persons;
The image processing apparatus according to claim 1. The reference area is a face image area of a subject.
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. Stores model data for obtaining the predicted value, which is created in advance based on the size of the image area representing the specific part of the subject to be the model and the size of the image area representing the skin of the subject to be the model A storage unit
The prediction unit obtains the prediction value in consideration of the model data stored in the storage unit;
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. The predicting unit estimates a ratio of a portion included in the target image of the whole body image of the subject, and acquires the predicted value in consideration of the estimated ratio;
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. The prediction unit estimates the ratio in consideration of arrangement information in the target image of the reference region,
The image processing apparatus according to claim 5. The prediction unit estimates the ratio in consideration of the age of the subject person estimated based on the reference region,
The image processing apparatus according to claim 5, wherein the image processing apparatus is an image processing apparatus. The determination unit, when the measured value exceeds the predicted value exceeding a predetermined threshold, compared to the intensity determined when the measured value does not exceed the predicted value exceeding the predetermined threshold A small intensity is determined as the corrected intensity,
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. An imaging unit that generates a target image by imaging a subject;
The image processing apparatus according to any one of claims 1 to 8, wherein a target image generated by the imaging unit is acquired;
An imaging apparatus comprising: An acquisition step of acquiring a target image;
A detection step of detecting a reference region, which is an image region representing a specific part of the subject, from the target image;
A prediction step of obtaining a predicted value of the size of the image area representing the skin of the subject included in the target image based on the size of the reference area;
An actual measurement step of acquiring an actual measurement value of the size of the correction target area, which is an image area having the same or similar pixel value as the image area representing the skin of the subject, included in the target image;
A determination step of determining the correction strength by comparing the measured value and the predicted value;
A processing step of generating a processed image obtained by performing image processing with the correction intensity on the correction target region of the target image;
Comprising
An image processing method. Computer
An acquisition unit for acquiring a target image;
A detection unit that detects a reference region that is an image region representing a specific part of the subject from the target image;
A prediction unit that obtains a predicted value of the size of the image area representing the skin of the subject included in the target image based on the size of the reference area;
An actual measurement unit that acquires an actual measurement value of the size of the correction target region that is an image region that has the same or similar pixel value as the image region that represents the skin of the subject included in the target image,
A determination unit that compares the measured value with the predicted value to determine a correction strength,
A processing unit that generates a processed image obtained by performing image processing with the correction intensity on the correction target region of the target image;
A program characterized by functioning as

Images