JPWO2016151850A1 - Imaging apparatus, signal processing apparatus, and skin diagnosis system - Google Patents

Abstract

Even in a general-purpose imaging device that does not have a mechanical drive mechanism, it is a signal processing device that can obtain a good image with reduced shine components. The imaging apparatus includes an image acquisition unit that acquires an input image signal, a signal processing unit that inputs the input image signal, performs signal processing, and outputs an output image signal, a storage unit that stores the output image signal in a medium, Is provided. The signal processing unit includes a smoothed luminance calculating unit 1 that calculates a smoothed luminance by smoothing an input luminance signal within the first smoothing range around each pixel signal included in the input image signal, and an input luminance The signal and the smoothed brightness are compared. If the input brightness signal is larger than the smoothed brightness, the difference between the input brightness signal and the smoothed brightness is reduced by a first ratio, and the output brightness signal is replaced with the input brightness signal. A luminance signal processing unit 2 to be generated.

Description

  The present invention relates to an imaging device, a signal processing device, and a skin diagnosis system.

  As background art in this technical field, there is JP-A-7-322103 (Patent Document 1). According to the publication, “[Purpose] Obtains a device that generates little heat and is soft on the skin, and is effective for downsizing. [Configuration] A lens 23, an image sensor 25, and a photo shoot facing the shooting window 22 inside the case. A circular fluorescent lamp 42 that illuminates the window side is provided, and a transparent slide pipe 31 is provided, which is provided with a light shielding ring 33. The light shielding ring 33 moves the transparent slide pipe in the axial direction. In response to this, either the path A or the path B of the illumination light is realized. "

JP 7-322103 A

  When photographing human skin, depending on the lighting environment and skin condition, the influence of reflected light on the skin surface appears strongly, and a component having a strong brightness unlike the original color may be generated in the image. This component is called “shine”.

  In Patent Document 1, a transparent slide pipe and a light shielding ring are provided so that a good image with uniform illumination can be obtained. However, a configuration using a dedicated device having a mechanical drive mechanism is used. Therefore, the durability and versatility of the apparatus are not taken into consideration.

  An object of the present invention is to provide a signal processing device that can obtain a good image with reduced shine components even in a general-purpose imaging device that does not have a mechanical drive mechanism.

  The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  An imaging apparatus according to an embodiment includes an image acquisition unit that acquires an input image signal, a signal processing unit that inputs the input image signal, performs signal processing, and outputs an output image signal, and the output image signal as a medium And a storage unit for storing. The signal processing unit smoothes an input luminance signal within a first smoothing range and calculates a smoothed luminance around each pixel signal included in the input image signal, and the input A luminance signal is compared with the smoothed luminance, and if the input luminance signal is larger than the smoothed luminance, the difference between the input luminance signal and the smoothed luminance is reduced by a first ratio, and the input luminance signal And a luminance signal processing unit that generates an output luminance signal.

  A signal processing apparatus according to an embodiment includes: a smoothed luminance calculating unit that calculates a smoothed luminance by smoothing an input luminance signal within a first smoothing range around each pixel signal included in an input image signal; Comparing the input luminance signal with the smoothed luminance, and if the input luminance signal is greater than the smoothed luminance, reducing the difference between the input luminance signal and the smoothed luminance by a first ratio, A luminance signal processing unit that generates an output luminance signal in place of the input luminance signal.

  The skin diagnosis system in one embodiment is a skin diagnosis system using the signal processing device, and includes a mobile terminal device including an image acquisition unit that acquires the input image signal and a communication unit, and the mobile terminal device. An external device that is attached and includes a lens and a light source for performing close-up photography of the skin, an information processing device that includes the signal processing device, and a network to which the portable terminal device and the information processing device are connected. Prepare. The portable terminal device acquires the close-up image of the skin via the external device, and transfers the close-up image of the skin to the information processing device via the network. The information processing device acquires the close-up image of the skin through the network, generates skin diagnosis data from an image obtained by controlling the signal processing device, and sends the result to the mobile terminal device via the network. Forward.

  The effects obtained by typical ones of the inventions disclosed in the present application will be briefly described as follows.

  According to one embodiment, it is possible to provide a signal processing device that can obtain a good image with reduced shine components even in a general-purpose imaging device that does not have a mechanical drive mechanism.

It is a block diagram explaining an example of the whole structure of the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining an example of the internal structure of the smoothing luminance calculation part used with the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining an example of the internal structure of the smoothing range determination part used with the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining the 1st structural example of the smoothing range determination part used with the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining the 2nd structural example of the smoothing range determination part used with the signal processing apparatus in Embodiment 1 of this invention. It is explanatory drawing explaining the relationship between a high-intensity signal ratio and a smoothing range area in the 1st structural example of the smoothing range determination part used with the signal processing apparatus of Embodiment 1 of this invention. It is a block diagram explaining the 1st structural example of the determination area | region detection part used with the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining an example of the internal structure of the gravity center position calculation part contained in the 1st structural example of the determination area | region detection part used with the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining an example of the internal structure of the skin area | region detection part used with the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining an example of an internal structure of the skin area | region detection part which added the function which excludes the pixel which is a low-intensity signal in the skin diagnosis system in Embodiment 3 of this invention, and detected. It is a block diagram explaining the 2nd structural example of the determination area | region detection part used with the signal processing apparatus in Embodiment 1 of this invention. 5 is a flowchart illustrating an example of internal processing in the signal processing apparatus according to Embodiment 1 of the present invention. In the signal processing apparatus in Embodiment 1 of this invention, it is explanatory drawing explaining an example of the input-output characteristic of the luminance signal before correction | amendment, and the luminance signal after correction | amendment. In the signal processing apparatus in Embodiment 1 of this invention, it is explanatory drawing explaining an example of the luminance distribution with respect to the position of the luminance signal before correction | amendment, and the smoothing luminance. In the signal processing apparatus in Embodiment 1 of this invention, it is explanatory drawing explaining an example of the luminance distribution with respect to the position of the luminance signal after correction | amendment in addition to the luminance signal before correction | amendment and smoothed luminance. It is explanatory drawing explaining an example of operation | movement of a smoothing calculating part in the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining an example of an internal structure of the smoothing range determination part which added the function which determines an exposure defect from an input image used with the skin diagnostic system in Embodiment 3 of this invention. In the signal processing apparatus in Embodiment 1 of this invention, it is explanatory drawing explaining an example of HSV color space. It is explanatory drawing explaining the definition example of the skin color used for the detection of the skin area | region used with the signal processing apparatus in Embodiment 1 of this invention. It is a block diagram explaining an example of the whole structure of the imaging device in Embodiment 2 of this invention. In Embodiment 2 of this invention, it is a block diagram explaining an example of the whole structure of the imaging device carrying an exposure control part and an external input control part. In Embodiment 2 of this invention, it is explanatory drawing explaining an example of the image acquired without performing any correction | amendment with respect to the to-be-photographed object containing a shine component. In Embodiment 2 of this invention, it is explanatory drawing explaining an example of the image acquired by performing only exposure control with respect to the to-be-photographed object containing a shine component. In Embodiment 2 of this invention, it is explanatory drawing explaining an example of the image acquired using the imaging device of Embodiment 2 with respect to the to-be-photographed object containing a shine component. 7 is a flowchart illustrating an example of internal processing for performing feedback control in the imaging apparatus according to Embodiment 2 of the present invention. It is a block diagram explaining an example of the whole structure of the skin diagnostic system in Embodiment 3 of this invention. It is sectional drawing explaining an example of the internal structure of the external device used with the skin diagnostic system in Embodiment 3 of this invention. It is a block diagram explaining an example of the internal structure of the cloud server used with the skin diagnostic system in Embodiment 3 of this invention. It is explanatory drawing explaining an example of the skin image acquired using the smart phone in the skin diagnostic system in Embodiment 3 of this invention. It is explanatory drawing explaining an example of the skin image output from the signal processing part used with the skin diagnostic system in Embodiment 3 of this invention.

  In the following embodiments, when it is necessary for the sake of convenience, the description will be divided into a plurality of sections or embodiments. However, unless otherwise specified, they are not irrelevant and one is the other. There are some or all of the modifications, details, supplementary explanations, and the like. Also, in the following embodiments, when referring to the number of elements (including the number, numerical value, quantity, range, etc.), particularly when clearly indicated and when clearly limited to a specific number in principle, etc. Except, it is not limited to the specific number, and may be more or less than the specific number.

Further, in the following embodiments, the constituent elements (including element steps and the like) are not necessarily indispensable unless otherwise specified and apparently indispensable in principle. Needless to say. Similarly, in the following embodiments, when referring to the shape and positional relationship of components and the like, the shape is substantially the same unless otherwise specified and the case where it is not clearly apparent in principle. And the like are included. The same applies to the above numerical values and ranges.
[Outline of the embodiment]

  First, an outline of the embodiment will be described. In the outline of the present embodiment, as an example, the description will be given with the reference numerals and the like of the corresponding components of the respective embodiments in parentheses.

  An imaging apparatus according to an embodiment includes an image acquisition unit (101) that acquires an input image signal, a signal processing unit (102) that inputs the input image signal, performs signal processing, and outputs an output image signal; And a storage unit (103) for storing the output image signal in a medium. The signal processing unit includes a smoothed luminance calculating unit (1) that smoothes an input luminance signal within a first smoothing range and calculates a smoothed luminance around each pixel signal included in the input image signal. Comparing the input luminance signal with the smoothed luminance, and if the input luminance signal is greater than the smoothed luminance, reducing the difference between the input luminance signal and the smoothed luminance by a first ratio, And a luminance signal processing unit (2) that generates an output luminance signal in place of the input luminance signal.

  A signal processing apparatus according to an embodiment smoothes an input luminance signal within a first smoothing range and calculates a smoothed luminance around each pixel signal included in the input image signal ( 1) is compared with the input luminance signal and the smoothed luminance. If the input luminance signal is larger than the smoothed luminance, the difference between the input luminance signal and the smoothed luminance is reduced by a first ratio. And a luminance signal processing unit (2) that generates an output luminance signal in place of the input luminance signal.

  The skin diagnosis system in one embodiment is a skin diagnosis system using the signal processing device, and includes a portable terminal device (40) including an image acquisition unit and a communication unit that acquire the input image signal, and the portable An external device (30) attached to a terminal device and including a lens and a light source for close-up of the skin, an information processing device (60) including the signal processing device, the portable terminal device, and the information processing device Are connected to the network (50). The portable terminal device acquires the close-up image of the skin via the external device, and transfers the close-up image of the skin to the information processing device via the network. The information processing device acquires the close-up image of the skin through the network, generates skin diagnosis data from an image obtained by controlling the signal processing device, and sends the result to the mobile terminal device via the network. Forward.

Hereinafter, each embodiment based on the outline | summary of embodiment mentioned above is described in detail based on drawing. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.
[Embodiment 1]

The signal processing apparatus according to Embodiment 1 will be described with reference to FIGS. FIG. 1 is a diagram for explaining the overall configuration of the signal processing apparatus according to the first embodiment. 2 to 18 are diagrams for describing the internal configuration and internal operation of the signal processing device according to Embodiment 1. FIG.
<Signal processing device>

  The configuration and operation of the signal processing apparatus according to the first embodiment will be described with reference to FIG. FIG. 1 is a configuration diagram illustrating an example of the overall configuration of the signal processing apparatus according to the first embodiment.

  As shown in FIG. 1, the signal processing apparatus according to Embodiment 1 includes a smoothed luminance calculation unit 1 and a luminance signal processing unit 2. The luminance signal processing unit 2 includes a luminance / color difference signal separation unit 21, a comparison unit 22, a subtraction unit 23, a gain multiplication unit 24, an addition unit 25, a luminance signal switching unit 26, and a luminance / color difference signal integration unit. 27.

  The smoothed luminance calculating unit 1 receives an input image signal, and outputs a smoothed luminance by smoothing the input luminance signal within a predetermined smoothing range around each pixel signal included in the input image signal. It is.

  In the luminance signal processing unit 2, the luminance / color difference signal separation unit 21 receives an input image signal, separates the input image signal into an input luminance signal and a color difference signal, and outputs an input luminance signal and a color difference signal. It is. The comparison unit 22 receives the input luminance signal from the luminance / color difference signal separation unit 21 and the smoothed luminance from the smoothed luminance calculation unit 1, compares the input luminance signal with the smoothed luminance, and compares the comparison result. It is a means to output.

  The subtracting unit 23 receives the input luminance signal from the luminance / color difference signal separating unit 21 and the smoothed luminance from the smoothed luminance calculating unit 1, and outputs a difference between the input luminance signal and the smoothed luminance by subtraction. Means. The gain multiplication unit 24 is means for receiving the difference between the input luminance signal from the subtraction unit 23 and the smoothed luminance, reducing the difference at a predetermined rate, and outputting the difference reduced at the predetermined rate.

  The addition unit 25 receives the difference reduced at a predetermined rate from the gain multiplication unit 24 and the smoothed luminance from the smoothed luminance calculation unit 1, and calculates the difference reduced at a predetermined rate to the smoothed luminance. A means for adding and outputting a first luminance signal. The luminance signal switching unit 26 receives the first luminance signal from the adding unit 25 and the second luminance signal having the same value as the input luminance signal from the luminance / color difference signal separating unit 21. A means for selecting one of the first luminance signal and the second luminance signal based on the comparison result and outputting the selected signal as an output luminance signal;

  The luminance / color difference signal integration unit 27 receives the output luminance signal from the luminance signal switching unit 26 and the color difference signal from the luminance / color difference signal separation unit 21, and integrates the output luminance signal and the color difference signal to output an output image. It is a means for outputting as a signal.

  The operation of the signal processing apparatus configured as described above is as follows. First, an input image signal is input to the smoothed luminance calculation unit 1. The smoothed luminance calculation unit 1 smoothes an input luminance signal within a predetermined smoothing range and outputs a smoothed luminance around each pixel signal included in the input image signal. The smoothed luminance is then input to the luminance signal processing unit 2 together with the input image signal. The input image signal input to the luminance signal processing unit 2 is first separated into an input luminance signal and a color difference signal by the luminance / color difference signal separation unit 21. Next, the comparison unit 22 compares the input luminance signal with the smoothed luminance.

  For example, when the input luminance signal is larger than the smoothed luminance, the difference between the input luminance signal and the smoothed luminance output from the subtracting unit 23 is reduced by a predetermined ratio in the gain multiplying unit 24, and the adding unit 25 The first luminance signal added back to the smoothed luminance is selected by the luminance signal switching unit 26. On the other hand, when the input luminance signal is not larger than the smoothed luminance, the luminance signal switching unit 26 selects the second luminance signal having the same value as the input luminance signal as the output luminance signal. The output luminance signal and the color difference signal are integrated by the luminance / color difference signal integration unit 27, and an output image signal is output.

  Detailed operation of each component will be described later (FIGS. 2 to 18).

In the signal processing apparatus according to the first embodiment, the luminance signal processing unit 2 is a signal processing means for the purpose of removing only the shine component of the input image signal. The luminance signal processing unit 2 sets a pixel signal having a relatively high luminance with respect to the peripheral pixels in the input image signal as a correction target, so that the luminance signal for the purpose of removing shine without depending on the absolute value of the luminance. A good image can be obtained even for an input image signal that can be processed and is obtained in a non-uniform light environment and has uneven brightness. The smoothed luminance calculation unit 1 is a means for determining the intensity of signal processing performed by the luminance signal processing unit 2 and the range of target pixels, and can perform shine removal suitable for the subject and shooting conditions.
<Smoothing luminance calculation unit>

  Next, the detailed configuration and operation of each component of the signal processing apparatus according to the first embodiment shown in FIG. 1 will be described. First, in the luminance signal processing unit 2, the smoothed luminance used as a reference for determining whether or not each pixel signal of the input image signal has a relatively high luminance with respect to the peripheral pixels is used as the input image signal. The configuration and operation of the smoothed luminance calculation unit 1 that is calculated for each pixel signal will be described.

  FIG. 2 is a configuration diagram illustrating an example of an internal configuration of the smoothed luminance calculation unit 1. As shown in FIG. 2, the smoothed luminance calculating unit 1 includes a smoothed range determining unit 3, an input luminance signal extracting unit 11, and a smoothing calculating unit 12.

  The input image signal input to the smoothed luminance calculating unit 1 is input to the smoothed range determining unit 3 and the input luminance signal extracting unit 11. The smoothing range determining unit 3 outputs a smoothing range for smoothing the luminance signal in the peripheral pixels of each pixel of the input image signal. The input luminance signal extraction unit 11 extracts and outputs an input luminance signal for smoothing from the input image signal. The smoothing calculation unit 12 smoothes the input luminance signal output from the input luminance signal extraction unit 11 of each pixel included in the smoothing range output from the smoothing range determination unit 3 and outputs the smoothed luminance. .

  FIG. 15 is an explanatory diagram illustrating an example of the operation of the smoothing calculation unit 12. F151 indicates an example of an image including a shine component, F152 indicates a high luminance area visually recognized as a shine component, F153 indicates one pixel (target pixel) to be processed, and F154 indicates an example of a smoothing range.

  In the example illustrated in FIG. 15, the target pixel F153 is inside the high luminance region F152, and thus has a relatively high luminance as compared with the peripheral pixels of the high luminance region F152. Therefore, when the smoothing range F154 is defined so as to include not only the high luminance region but also many peripheral pixels around the pixel of interest F153, the smoothing luminance calculated from the pixels included in the smoothing range F154 is the same as that of the pixel of interest F153. The value is smaller than the luminance signal.

By adopting the configuration as described above, the luminance signal processing unit 2 performs smoothing as a reference for determining whether or not each pixel signal of the input image signal has a relatively high luminance with respect to surrounding pixels. It is possible to realize a signal processing device that calculates the normalized luminance for each pixel signal of the input image signal.
<Smoothing range determination unit>

  Next, the configuration and operation of the smoothing range determination unit 3 included in the smoothing luminance calculation unit 1 shown in FIG. 2 for determining a smoothing range suitable for the degree of the shine component from the input image signal will be described.

  FIG. 3 is a configuration diagram illustrating an example of an internal configuration of the smoothing range determination unit 3. As shown in FIG. 3, the smoothing range determination unit 3 includes a determination region detection unit 4 and a smoothing range determination unit 31. The determination area detection unit 4 is not necessarily installed.

  The input image signal input to the smoothing range determination unit 3 is input to the determination region detection unit 4 and the smoothing range determination unit 31. The determination area detection unit 4 outputs a determination area used for determining a smoothing range from the input image signal. The smoothing range determination unit 31 outputs a smoothing range to be input to the smoothing calculation unit 12 based on the information of the input image signal within the determination region output from the determination region detection unit 4. When the determination area detection unit 4 is not installed, an area for all pixels included in the input image signal is input to the smoothing range determination unit 31 as a determination area. In this case, the smoothing range can be determined based on the shine component of the entire input image signal without being limited to a specific subject or shooting conditions.

  FIG. 4 is a configuration diagram illustrating a first configuration example of the smoothing range determination unit 31. FIG. 6 is an explanatory diagram for explaining the relationship between the high luminance signal ratio and the smoothing range area in the first configuration example of the smoothing range determining unit 31. As shown in FIG. 4, the first configuration example of the smoothing range determination unit 31 includes an input luminance signal extraction unit 311, a high luminance signal ratio calculation unit 312, and a smoothing range output unit 313. Here, the input luminance signal extraction unit 311 is equivalent to the input luminance signal extraction unit 11 shown in FIG.

  The input image signal input to the smoothing range determination unit 31 is input to the input luminance signal extraction unit 311. The input luminance signal extraction unit 311 extracts and outputs an input luminance signal for calculating the ratio of the high luminance signal from the input image signal. On the other hand, the determination region output from the determination region detection unit 4 and input to the smoothing range determination unit 31 is input to the high luminance signal ratio calculation unit 312. The high luminance signal ratio calculation unit 312 uses the input luminance signals for all the pixels included in the determination region, and calculates the ratio of the high luminance signal based on a predetermined threshold related to luminance. Here, the ratio of the high luminance signal corresponds to the degree of the shine component in the entire determination region.

  The high luminance signal ratio output from the high luminance signal ratio calculation unit 312 is input to the smoothing range output unit 313. As shown in FIG. 6, the smoothing range output unit 313 outputs a larger smoothing range area as the high luminance signal ratio increases. As a result, the luminance signal can be smoothed in a range suitable for the degree of the shine component in the entire determination region. The smoothing range output from the smoothing range output unit 313 is output from the smoothing range determination unit 31.

  FIG. 5 is a configuration diagram illustrating a second configuration example of the smoothing range determination unit 31. As shown in FIG. 5, the second configuration example of the smoothing range determination unit 31 includes an input luminance signal extraction unit 311, a high luminance signal ratio determination unit 314, and a smoothing range switching unit 315.

  The second configuration example of the smoothing range determination unit 31 illustrated in FIG. 5 is a configuration in which the first configuration example illustrated in FIG. 4 is simplified. The high luminance signal ratio determination unit 314 determines, based on a plurality of predetermined threshold values related to luminance and ratio, from the input luminance signal for all pixels included in the input determination region, based on a binary value. And output the determination result. The determination result is input to the smoothing range switching unit 315.

  For example, in the smoothing range switching unit 315, if the first smoothing range is larger than the second smoothing range, if the determination result is true, that is, if it is determined that the degree of the shine component is large, A first smoothing range is selected. On the other hand, when the determination result is false, that is, when it is determined that the degree of the shine component is small, the second smoothing range is selected. Accordingly, it is possible to select one type from two types of smoothing ranges based on the degree of the shine component in the entire determination region and smooth the luminance signal. The smoothing range selected by the smoothing range switching unit 315 is output from the smoothing range determining unit 31.

  The configuration example of the smoothing range determination unit 31 illustrated in FIGS. 4 and 5 is an example of means for determining the smoothing range, and is not limited thereto.

By adopting the configuration as described above, it is possible to realize a signal processing device that determines a smoothing range suitable for the degree of the shine component from the input image signal.
<Determination area detection unit>

  Subsequently, a determination region that is included in the smoothing range determination unit 3 illustrated in FIG. 3 and outputs a determination region suitable for the subject and the shooting conditions as a determination region used to determine the smoothing range from the input image signal. The configuration and operation of the detection unit 4 will be described.

  FIG. 7 is a configuration diagram illustrating a first configuration example of the determination region detection unit 4. As shown in FIG. 7, the first configuration example of the determination region detection unit 4 includes a skin region detection unit 5, a centroid position calculation unit 41, and a rectangular region output unit 42.

  The input image signal input to the determination region detection unit 4 is input to the skin region detection unit 5 and the gravity center position calculation unit 41. The skin area detection unit 5 detects and outputs a skin area included in the input image signal. The centroid position calculation unit 41 calculates and outputs the centroid position for the skin area in the input image signal. The rectangular area output unit 42 receives the position of the center of gravity, defines and outputs a rectangular area having a predetermined length as one side centered on the position of the center of gravity. The first configuration example of the determination area detection unit 4 outputs a rectangular area as a determination area.

  FIG. 8 is a configuration diagram illustrating an example of an internal configuration of the gravity center position calculation unit 41 included in the first configuration example of the determination region detection unit 4. As shown in FIG. 8, the centroid position calculation unit 41 includes an input luminance signal extraction unit 411, a binary image generation unit 412, and a binary image centroid position calculation unit 413. Here, the input luminance signal extraction unit 411 is equivalent to the input luminance signal extraction unit 311 shown in FIG.

The input image signal input to the barycentric position calculation unit 41 is input to the input luminance signal extraction unit 411. The input luminance signal extraction unit 411 extracts and outputs an input luminance signal for generating a binary image from the input image signal. The input luminance signal for generating the binary image is input to the binary image generation unit 412 together with the skin region input to the gravity center position calculation unit 41. Binary image generating unit 412, when the input luminance signal is included in the skin region is white, if not included outputs the binary luminance signal Y _b as the black. Binary image centroid position calculation unit 413 inputs the binary luminance signal Y _b, calculates the barycentric position p _c on the basis of the number 1. Here, p is the position of each pixel, p _c is the center of gravity position, Y _b represents a binary luminance signal.

  When the skin area is close to a circle, the barycentric position output by the binary image barycentric position calculating unit 413 is located near the center of the skin area. Therefore, when the determination region detection unit 4 is configured as shown in FIG. 7, it is possible to determine the degree of shine by widely using a region near the center away from the edge of the skin region.

  FIG. 11 is a configuration diagram illustrating a second configuration example of the determination region detection unit 4. As shown in FIG. 11, the determination area detection unit 4 includes a skin area detection unit 5, a face recognition unit 6, and a common region output AND operation unit 43.

  The input image signal input to the determination region detection unit 4 is input to the skin region detection unit 5 and the face recognition unit 6. The skin area detection unit 5 is means for detecting a skin area included in the input image signal, and outputs true if the input pixel is a skin area, and false otherwise. On the other hand, the face recognition unit 6 is a means for detecting a face area from the input image signal. The face recognition section 6 is a means for outputting true if the input pixel is a face area, and outputting false otherwise. It may be configured by technology. The skin region output by the skin region detection unit 5 and the face region output by the face recognition unit 6 are input to the common region output logical product operation unit 43, and a common region based on the logical product of both is output as a determination region.

  When an image including a face is input as a subject and the focus is on removal of the shine component generated on the face, the determination region detection unit 4 is configured as shown in FIG. Signal processing can be realized.

By adopting the configuration as described above, it is possible to realize a signal processing device that outputs a determination region suitable for a subject and photographing conditions as a determination region used for determining a smoothing range from an input image signal.
<Skin area detection unit>

  Next, the configuration and operation of the skin region detection unit 5 that is included in the determination region detection unit 4 illustrated in FIGS. 7 to 11 and detects a skin region from the input image signal will be described.

  FIG. 9 is a configuration diagram illustrating an example of an internal configuration of the skin region detection unit 5. FIG. 17 is an explanatory diagram for explaining an example of the HSV color space. FIG. 18 is an explanatory diagram for explaining a definition example of the skin color used for detecting the skin region. As shown in FIG. 9, the skin region detection unit 5 includes a color signal extraction unit 51, a color signal R smoothing filter 52, a color signal G smoothing filter 53, and a color signal B smoothing filter 54, A hue calculation unit 55 and a skin color determination unit 56 are included.

  The input image signal input to the skin area detection unit 5 is input to the color signal extraction unit 51. The color signal extraction unit 51 extracts three types of color signals R, color signals G, and color signals B from the input image signals that have been input, and outputs them. The three kinds of color signals are obtained by using the color signal R by the color signal R smoothing filter 52, the color signal G by the color signal G smoothing filter 53, and the color signal B by the color signal B smoothing filter 54. Is smoothed in a predetermined range with respect to. The color signal R, color signal G, and color signal B output from the various smoothing filters 52 to 54 are input to the hue calculation unit 55. The hue calculation unit 55 calculates the hue signal H based on Equation 2.

  The hue signal H is a kind of signal constituting the HSV color space shown in FIG. The HSV color space is a kind of color space similar to the RGB color space, and a hue signal (H) representing the type of color, a saturation signal (S) representing the vividness of the color, and a lightness signal (V) representing the brightness. It is composed of The hue signal output from the hue calculation unit 55 is input to the skin color determination unit 56. As shown in FIG. 18, the skin color determination unit 56 defines the hue signal included between the predetermined first threshold value F181 and the predetermined second threshold value F182 related to the hue signal as the skin color hue F183 and is input. The skin area is output by comparing the hue signals of the respective pixels. The values of the predetermined first threshold value F181 and the predetermined second threshold value F182 may be adjusted so as to be suitable for the skin color as the subject. The skin area output by the skin color determination unit 56 is output from the skin area detection unit 5.

  By adopting the configuration as described above, it is possible to realize a signal processing device that detects a skin region from an input image signal.

As described above, by combining the smoothing range determination unit 3, the determination region detection unit 4, and other signal processing units, and by appropriately using the skin region detection unit 5, the face recognition unit 6, and the like, the luminance signal processing unit 2 (FIG. 1), it is possible to realize a signal processing apparatus that calculates the smoothed luminance for performing signal processing suitable for the subject and shooting conditions for each pixel signal of the input image signal.
<Luminance signal processor>

  Subsequently, regarding the configuration and operation of the luminance signal processing unit 2 that performs the luminance signal processing that removes only the shine component of the input image signal, with the pixel signal having a relatively high luminance relative to the surrounding pixels in the input image signal being corrected. explain.

  The luminance signal processing unit 2 shown in FIG. 1 includes a luminance / color difference signal separation unit 21, a comparison unit 22, a subtraction unit 23, a gain multiplication unit 24, an addition unit 25, a luminance signal switching unit 26, And a color difference signal integration unit 27.

  FIG. 13 is an explanatory diagram illustrating an example of input / output characteristics of the luminance signal before correction and the luminance signal after correction. FIG. 14A is an explanatory diagram illustrating an example of a luminance distribution with respect to the position of the luminance signal before correction and the smoothed luminance. FIG. 14B is an explanatory diagram illustrating an example of a luminance distribution with respect to the position of the corrected luminance signal in addition to the luminance signal before correction and the smoothed luminance. The position of each pixel is actually represented in two dimensions, but is simplified to one dimension in FIGS. 14A and 14B.

  The input image signal input to the luminance signal processing unit 2 is input to the luminance / color difference signal separation unit 21 and outputs an uncorrected luminance signal for performing luminance signal processing and an output image signal in the luminance / color difference signal integration unit 27. Therefore, it is separated into color difference signals to be used. On the other hand, the luminance signal processing unit 2 receives the smoothed luminance output from the smoothed luminance calculating unit 1. The luminance signal before correction and the smoothed luminance are input to the comparison unit 22. As a determination result, the comparison unit 22 outputs true when the pre-correction luminance signal is greater than the smoothed luminance, and outputs false otherwise. The determination result and the output result are input to the luminance signal switching unit 26. The luminance signal switching unit 26 selects and outputs the first luminance signal as the corrected luminance signal if the determination result is true and the second luminance signal if the determination result is false. Here, the second luminance signal is a luminance signal having the same value as that of the luminance signal before correction without performing any processing on the luminance signal before correction.

  Next, a method for generating the first luminance signal will be described. The pre-correction luminance signal is input to the subtracting unit 23. The smoothing luminance is also input to the subtraction unit 23, and the smoothing luminance is subtracted from the pre-correction luminance signal, and the difference is output. This difference is input to the gain multiplier 24, multiplied by a predetermined gain, and output. In the luminance signal processing for the purpose of removing shine, the gain is less than 1 because the processing is to suppress the high luminance component. As shown in FIG. 13, this gain may be a predetermined uniform value or a value that varies depending on the magnitude of the difference. Moreover, it is good also as a structure which can change a value from the outside.

  The output value of the gain multiplication unit 24 is input to the addition unit 25, added back to the smoothed luminance, and becomes the first luminance signal. The corrected luminance signal output when the luminance signal switching unit 26 selects either the first luminance signal or the second luminance signal based on the determination result of the comparison unit 22 is luminance. / The color difference signal integration unit 27 is input. The luminance / color difference signal integration unit 27 receives the color difference signal output from the luminance / color difference signal separation unit 21 together, and generates and outputs an output image signal by integrating both signals. The luminance signal processing unit 2 outputs an output image signal.

  When the luminance signal before correction has a luminance distribution as shown in FIG. 14A, the luminance distribution of the smoothed luminance calculated for each pixel is between the maximum value and the minimum value of the luminance distribution in the luminance signal before correction. Transition to sew. Then, by performing non-linear luminance processing based on the smoothed luminance, as shown in FIG. 14B, a pixel signal having a relatively high luminance with respect to surrounding pixels can be set as a correction target. Therefore, it is possible to perform luminance signal processing for the purpose of removing shine without depending on the absolute value of luminance, and it is good even for an input image signal that is acquired in a non-uniform light environment and has uneven luminance. An image can be obtained.

By adopting the configuration as described above, a signal processing apparatus that performs luminance signal processing for correcting a pixel signal having a relatively high luminance with respect to surrounding pixels in an input image signal and removing only a shine component of the input image signal. Is feasible.
<Internal processing of signal processing device>

  Finally, an example of the internal processing flow of the signal processing apparatus according to the first embodiment shown in FIG. 1 will be described. FIG. 12 is a flowchart for explaining an example of internal processing in the signal processing apparatus shown in FIG.

  First, in step S1, the smoothing range determination unit 3 in the smoothing luminance calculation unit 1 determines a smoothing region from input image signals of all pixels. Next, the process proceeds to step S2. From step S2 onward, processing is performed in units of one pixel. In the following description, it is assumed that the luminance signal before correction is luminance Y and the luminance signal after correction is luminance Y ′. Since the luminance signal processing is performed pixel by pixel by raster scan, the target pixel is set in step S2. Raster scanning is a scanning method that starts from the origin and moves to the next row when scanning of one horizontal row is completed.

  At the start of luminance signal processing, the origin is set as the target pixel. Next, in step S3, the smoothing calculation unit 12 calculates the smoothing luminance around the target pixel based on the smoothing region determined in step S1, based on the luminance Y of the target pixel.

  In step S4, the comparison unit 22 in the luminance signal processing unit 2 compares the luminance Y of the pixel of interest with the smoothed luminance. If the luminance Y of the pixel of interest is greater than the smoothed luminance, the subtracting unit in step S5. 23, the gain multiplication unit 24 and the addition unit 25 calculate the luminance Y ′ by reducing the difference between the luminance Y of the target pixel and the smoothed luminance at a predetermined ratio and adding it back to the smoothed luminance. On the other hand, when the luminance Y of the target pixel is not larger than the smoothed luminance in step S4, the luminance Y ′ is set to the same value as the luminance Y without being processed as shown in step S6.

  The above steps S2 to S6 are repeated until the processing is completed for the input image signals of all the pixels in step S7. After it is determined in step S7 that all the pixels have been processed, in step S8, the luminance / color difference signal integration unit 27 replaces the luminance Y in the input image signal before correction with the luminance Y ′, and converts the output image signal into the output image signal. Output and use as corrected image.

  By taking the processing flow as described above, it is possible to realize a signal processing method for performing the luminance signal processing of the first embodiment in units of pixels.

As described above, in the first embodiment, it is possible to realize a signal processing device that performs luminance signal processing for the purpose of removing shine, which is suitable for the subject and shooting conditions, for the acquired input image signal.
<Effect of Embodiment 1>

  According to the signal processing apparatus in the first embodiment described above, a good image with reduced shine components can be obtained. Specifically, a good image with reduced shine components can be obtained by performing luminance signal processing for the purpose of removing the shine, which is suitable for the subject and shooting conditions, on the acquired input image signal. More details are as follows.

  (1) The signal processing apparatus includes a smoothed luminance calculation unit 1 and a luminance signal processing unit 2. As a result, the smoothed luminance calculation unit 1 can calculate the smoothed luminance by smoothing the input luminance signal within a predetermined smoothing range around each pixel signal included in the input image signal. Further, the luminance signal processing unit 2 compares the input luminance signal with the smoothed luminance, and when the input luminance signal is larger than the smoothed luminance, the difference between the input luminance signal and the smoothed luminance is reduced by a predetermined ratio, An output luminance signal can be generated by replacing the input luminance signal.

  (2) The smoothed luminance calculating unit 1 includes a smoothed range determining unit 3 that determines a smoothed range from the input image signal. Thereby, the smoothed luminance calculating unit 1 can calculate the smoothed luminance by smoothing the input luminance signal within the smoothed range output from the smoothed range determining unit 3 in the input image signal.

  (3) The smoothing range determination unit 3 includes a determination region detection unit 4 that detects a determination region from the input image signal. Thereby, the smoothing range determination unit 3 calculates the number of pixels having a luminance signal higher than a predetermined threshold of the determination region detected by the determination region detection unit 4 and the ratio to the total number of pixels in the input image signal, The larger the ratio, the larger the smoothing range.

  (4) The determination area detection unit 4 can output all the pixels included in the input image signal as a determination area.

  (5) The determination region detection unit 4 includes a skin region detection unit 5 that detects a skin region from the input image signal. As a result, the determination region detection unit 4 calculates the centroid position of the skin region detected by the skin region detection unit 5 in the input image signal, and determines a rectangular region having a predetermined length on one side centered on the centroid position. It can be output as a region.

  (6) The determination region detection unit 4 includes a skin region detection unit 5 that detects a skin region from an input image signal, and a face recognition unit 6 that outputs a face region from the input image signal. Thereby, the determination region detection unit 4 can output a region where the skin region detected by the skin region detection unit 5 and the face region output from the face recognition unit 6 overlap as a determination region.

(7) According to the first embodiment, the acquired input image signal is subjected to the luminance signal processing for the purpose of removing the shine, which is suitable for the subject and the shooting conditions, so that a good image in which the shine component is reduced. It is possible to realize a signal processing device capable of obtaining the above.
[Embodiment 2]

An imaging device according to Embodiment 2 will be described with reference to FIGS. FIG. 19 is a diagram for explaining the overall configuration of the imaging apparatus according to the second embodiment. 20 to 22 are diagrams for describing an internal configuration and an internal operation of the imaging apparatus according to the second embodiment.
<Imaging device>

  The configuration and operation of the imaging apparatus according to Embodiment 2 will be described with reference to FIG. FIG. 19 is a configuration diagram illustrating an example of the overall configuration of the imaging apparatus according to the second embodiment.

  The imaging apparatus 10 according to the second embodiment includes an image acquisition unit 101, a signal processing unit 102, a storage unit 103, and a control unit 104, as shown in FIG. The second embodiment is an imaging device equipped with the signal processing device described in the first embodiment, which performs luminance signal processing for the purpose of removing shine suitable for the subject and shooting conditions for the acquired input image signal. is there. By mounting the signal processing device, for example, an excessive decrease in luminance due to exposure control can be suppressed, and an image with good visibility can be obtained.

  The image acquisition unit 101 includes a lens and a shutter having predetermined performance, acquires an input image, and outputs an input image signal. The signal processing unit 102 is equipped with the signal processing apparatus described in the first embodiment as shown in FIG. 1, and receives the input image signal output from the image acquisition unit 101 and outputs the output image signal. . The output image signal is input to the storage unit 103 and stored in a predetermined storage medium. These image acquisition unit 101, signal processing unit 102, and storage unit 103 are controlled by the control unit 104.

  FIG. 20 is a configuration diagram illustrating an example of the overall configuration of the imaging apparatus 10 in which the exposure control unit 1041 and the external input control unit 1042 are mounted as a configuration example of the control unit 104. The external input control unit 1042 is intended to transmit an external input such as a key operation to the components in the imaging apparatus 10. For example, an instruction to save or delete image data is sent to the storage unit 103 to check whether or not strobe light is emitted. The representing signal is transmitted to the image acquisition unit 101 and the exposure control unit 1041.

  In addition to these, a function for operating the signal processing unit 102 is installed in the external input control unit 1042. For example, the gain multiplier 24 shown in FIG. 1, the high luminance signal ratio calculator 312 shown in FIG. 4, the high luminance signal ratio determiner 314 shown in FIG. 5, the skin color determiner 56 shown in FIG. When the user manually adjusts the predetermined threshold value, it is possible to realize a shine removal process that is suitable for a subject having a special condition or a shooting environment.

  The exposure control unit 1041 can input an output image signal output from the signal processing unit 102. Further, a signal indicating use of strobe light emission is input from the external input control unit 1042. In general, the exposure control is aimed at obtaining an imaging result image in which the brightness is balanced by decreasing the exposure amount when the subject is bright and increasing the exposure amount when the subject is dark. For this reason, when an image containing a large amount of shine components is acquired, the exposure control operates so as to lower the luminance of the entire image, so that the luminance decreases excessively and the visibility of the area excluding the shine components may decrease. It was. In the second embodiment, by combining the signal processing unit 102 that removes shine and the exposure control unit 1041, an excessive decrease in luminance due to exposure control can be suppressed, and an image with high visibility can be obtained.

  When the strobe light emission is not used, the exposure control unit 1041 performs exposure from the output image signal output from the signal processing unit 102 while the imaging device 10 is directed toward the subject and the image acquisition unit 101 continuously acquires images. A control amount is calculated and the image acquisition unit 101 is controlled. Accordingly, it is possible to appropriately calculate the exposure control amount from the image signal after the removal of the shine, and to realize feedback control based on the degree of the shine component of the subject.

  On the other hand, when strobe light emission is used, feedback control cannot be performed, but it is expected that the shine component will be removed by the signal processing unit 102, and the exposure control amount of the exposure control unit 1041 can be set brighter. It becomes. Alternatively, even in the case of using strobe light emission, in order to realize feedback control, the strobe light is emitted twice in a short time, and the image signal acquired at the first strobe light emission is used for feedback control, and the exposure control unit 1041 It is also possible to take means for calculating the exposure control amount and storing the image acquired upon the second flash emission.

  FIG. 21A is an explanatory diagram illustrating an example of an image acquired without performing any correction on a subject including a shine component. FIG. 21B is an explanatory diagram illustrating an example of an image acquired by performing only exposure control on a subject including a shine component. FIG. 21C is an explanatory diagram illustrating an example of an image acquired using the imaging apparatus according to the second embodiment for a subject including a shine component.

  As shown in FIG. 21A, the subject is an image of a human F211 at a short distance from the imaging device and a tree F213 at a long distance from the imaging device at the same time using strobe light emission at night. On the face of the human F211 that is present, the shine component F212 is generated. When only subject exposure control is performed, as shown in FIG. 21B, the shine component F215 after exposure control is reduced compared to the shine component F212, but the human F214 and tree F216 after exposure control are also Each becomes darker than the human F211 and the tree F213 shown in FIG. 21A, and the visibility decreases.

On the other hand, when the imaging device of the second embodiment is used, as shown in FIG. 21C, the shine component F218 after using the imaging device of the second embodiment is compared with the shine component F212 shown in FIG. 21A. In addition to being reduced, the human F217 and the tree F219 after using the imaging apparatus according to the second embodiment are not deteriorated in brightness as compared with the human F211 and the tree F213 shown in FIG. An image can be obtained.
<Internal processing of imaging device>

  Finally, an example of the internal processing flow of the imaging apparatus according to the second embodiment shown in FIG. 20 will be described. FIG. 22 is a flowchart for explaining an example of internal processing for performing feedback control in the imaging apparatus shown in FIG.

  First, in step S221, the external input control unit 1042 shown in FIG. This external input is a shutter operation or setting of the image acquisition unit 101 or the signal processing unit 102 by the user. Next, in step S222, various settings of the image acquisition unit 101 and the signal processing unit 102 are read.

  In step S223, the image acquisition unit 101 is operated while reflecting various settings, and an input image signal is acquired. In step S224, this input image signal is input to the signal processing unit 102, performs the signal processing described in the first embodiment as shown in FIG. 1, and outputs an output image signal. This output image signal is input to the exposure control unit 1041 in step S225, and the predetermined setting of the image acquisition unit 101 is changed from the output image signal after the signal processing in order to perform exposure control.

  The above steps S221 to S225 are repeated until it is confirmed in step S226 that the shutter operation by pressing the shutter is confirmed. If the shutter operation is confirmed, in step S227, the storage unit 103 stores the output image signal in a predetermined storage medium.

  By taking the processing flow as described above, the imaging method of the second embodiment can be realized.

As described above, in the second embodiment, it is possible to realize an imaging apparatus that performs luminance signal processing for the purpose of removing shine, which is suitable for the subject and shooting conditions, for the acquired input image signal.
<Effect of Embodiment 2>

  According to the imaging apparatus in the second embodiment described above, the effects (1) to (7) according to the first embodiment can be obtained, and a general-purpose apparatus that does not have a mechanical drive mechanism. Also in the imaging apparatus, a good image with reduced shine component can be obtained by performing luminance signal processing for the purpose of removing the shine, which is suitable for the subject and shooting conditions, on the acquired input image signal. More details are as follows.

  (11) The imaging device 10 includes an image acquisition unit 101, a signal processing unit 102, and a storage unit 103. Thereby, the image acquisition part 101 can acquire an input image signal. Further, the signal processing unit 102 can input an input image signal, perform signal processing, and output an output image signal. The storage unit 103 can store the output image signal in a predetermined medium.

  (12) The imaging apparatus 10 includes an exposure control unit 1041. Accordingly, the exposure control unit 1041 can calculate the exposure control amount from the output image signal output from the signal processing unit 102 and can control the image acquisition unit 101.

  (13) The imaging apparatus 10 includes an external input control unit 1042. Accordingly, the external input control unit 1042 can operate the image acquisition unit 101, the signal processing unit 102, the storage unit 103, and the exposure control unit 1041.

(14) According to the second embodiment, even in a general-purpose imaging device 10 that does not have a mechanical drive mechanism, the purpose is to remove shine on the acquired input image signal that is suitable for the subject and shooting conditions. By performing the luminance signal processing as follows, it is possible to realize the imaging device 10 that can obtain a good image with reduced shine components.
[Embodiment 3]

A skin diagnosis system according to Embodiment 3 will be described with reference to FIGS. 10, 16, and 23 to 26. FIG. 23 is a diagram for explaining the overall configuration of the skin diagnostic system according to the third embodiment. FIGS. 10, 16, and 24 to 26 are diagrams for describing the internal configuration and internal operation of the skin diagnostic system according to Embodiment 3. FIG.
<Skin diagnosis system>

  The configuration and operation of the skin diagnosis system according to Embodiment 3 will be described with reference to FIG. FIG. 23 is a configuration diagram illustrating an example of the overall configuration of the skin diagnostic system according to the third embodiment.

  As shown in FIG. 23, the skin diagnosis system according to the third embodiment includes a smartphone 40, an external device 30, a network 50, and a cloud server 60. The skin diagnosis performed in the present third embodiment is characterized in that “texture” of the skin 20 is included in the diagnosis target. That is, the third embodiment performs high-accuracy skin diagnosis while leaving a texture component with low luminance and removing only a high-brightness shine component with respect to the input skin image to obtain a good image. It is a feature.

  In addition, since the third embodiment is an application that uses a general-purpose portable terminal device such as the smartphone 40 that includes an image acquisition unit and a communication unit, the user can select a model of the portable terminal device that the user possesses. Another feature is that the application can be used without dependence. In the third embodiment, the cloud server 60 is used as an information processing apparatus including the signal processing apparatus described in the first embodiment.

  In the skin diagnosis system according to the third embodiment, external device 30 is attached to smartphone 40, external device 30 is brought into close contact with skin 20, and a skin image is acquired by smartphone 40. The acquired skin image is transmitted to the cloud server 60 via the network 50. The cloud server 60 performs skin diagnosis on the input skin image based on a predetermined algorithm, and outputs a skin diagnosis result. The output skin diagnosis result is output from the cloud server 60 and sent to the smartphone 40 via the network 50.

  If the exposure condition of the input skin image is bad, the cloud server 60 outputs an exposure failure determination based on a predetermined algorithm and is sent from the cloud server 60 to the smartphone 40 via the network 50.

  FIG. 24 is a cross-sectional view illustrating an example of the internal configuration of the external device 30. In the third embodiment, as shown in FIG. 24, the external device 30 includes a housing F255, a ring-shaped component F256 for skin contact, an eyepiece F251, an LEDF252, a lens F253, and an objective hole F254. It is composed of In the external device 30, the housing F255 has an eyepiece hole F251 and an objective hole F254, and a ring-shaped component F256 is attached to the objective hole F254. Inside the housing F255, a lens F253 having a predetermined performance and an LEDF252 which is a light source are provided.

  In the third embodiment, it is assumed that the interior of the housing F255 is black and the ring-shaped component F256 for skin contact is gray. At the time of shooting, the ring-shaped component F256 is brought into close contact with the skin 20, and the LEDF 252 is turned on. Since the camera of the smartphone 40 is in contact with the eyepiece hole F251, by operating the camera of the smartphone 40, a skin image is acquired to the smartphone 40 via the objective hole F254, the lens F253, and the eyepiece hole F251.

  FIG. 26A is an explanatory diagram illustrating an example of a skin image acquired using the smartphone 40. The skin image acquired using the smartphone 40 shows the inside of the external device 30 around the skin F261 to be diagnosed. In the third embodiment, the black casing F255 and the gray ring-shaped component F256 is shown. Also, the skin reflection image F264 may appear thinly in the housing F255.

  FIG. 25 is a configuration diagram illustrating an example of the internal configuration of the cloud server 60. As shown in FIG. 25, the cloud server 60 includes a signal processing unit 601 and a skin diagnosis unit 602. The skin image input to the cloud server 60 via the network 50 is input to the signal processing unit 601. The signal processing unit 601 performs correction, and the output skin image is input to the skin diagnosis unit 602. The skin diagnosis unit 602 performs skin diagnosis on the input skin image based on a predetermined algorithm, and outputs a skin diagnosis result. The output skin diagnosis result is output from the cloud server 60. If the exposure condition of the skin image input to the signal processing unit 601 is bad, the signal processing unit 601 outputs an exposure failure determination and is output from the cloud server 60.

  A signal processing unit 601 shown in FIG. 25 has a function of determining exposure failure from an input image and a low luminance signal in detection of a skin area, as shown in FIG. A function for excluding a certain pixel is added. The function of determining an exposure failure is for determining whether or not the input skin image is an exposure condition that is not suitable for skin diagnosis in the skin diagnosis unit 602. Specifically, in the skin diagnosis, skin texture is determined. When the input image is too bright so that the texture cannot be visually recognized, it is determined that the exposure is defective.

  The function of excluding pixels that are low luminance signals is for accurately detecting the skin area from the input skin image. Specifically, as shown in FIG. 26A, the skin that appears thinly in the housing F255. The reflected image F264 is excluded. In addition, as the determination region detection unit 4 included in the smoothing range determination unit 3 illustrated in FIG. 3, in the third embodiment, the first configuration example described in the first embodiment illustrated in FIG. 7 is used. . This is because the diagnosis target skin F261 included in the image acquired in the third embodiment has a circular shape surrounded by the ring-shaped component F256 as shown in FIG. This is because a wide area near the center away from the edge of the skin area can be used as the determination area.

  First, a signal processing device that determines an exposure failure when the input image is too bright so that the skin texture cannot be visually recognized will be described. The determination of exposure failure is output from the smoothing range determination unit 31 shown in FIG. 3 in the smoothing range determination unit 3 shown in FIG. 2 in the smoothed luminance calculation unit 1 shown in FIG.

  FIG. 16 is a configuration diagram illustrating an example of an internal configuration of the smoothing range determination unit 31 to which a function of determining an exposure failure from an input image is added. As shown in FIG. 16, the smoothing range determination unit 31 according to the third embodiment has an input luminance signal extraction unit 311, a high luminance signal ratio determination unit 314, a smoothing range switching unit 315, and an average luminance determination unit 316. And an exposure failure determination logical product operation unit 317.

  The high luminance signal ratio determination unit 314 performs the determination based on a plurality of predetermined threshold values related to luminance and ratio from input luminance signals for all pixels included in the input determination region by the method described in the first embodiment. The extent of the shine component in the region is determined by binary, and true is output when the shine component is large, and false is output otherwise. On the other hand, the average luminance determination unit 316 calculates the average luminance of the input luminance signal for all the pixels included in the input determination region, and determines the brightness level of the determination region based on a predetermined threshold related to the average luminance. Judgment is made based on the value, and true is output when the whole is bright, and false is output otherwise. The output results of the high luminance signal ratio determination unit 314 and the average luminance determination unit 316 are input to the exposure failure determination logical product operation unit 317, and the logical product of both is output as the exposure failure determination. That is, the condition for determining that the exposure is defective is that the degree of the shine component in the determination region is large and the whole is bright.

  By adopting the configuration as described above, it is possible to realize a signal processing device that determines an exposure failure when the input image is too bright so that the skin texture cannot be visually recognized.

  Subsequently, in order to exclude the reflected image F264 of the skin that is thinly reflected in the housing F255 as shown in FIG. 26A, a signal processing device that excludes pixels that are low luminance signals will be described. The exclusion of pixels that are low luminance signals is performed in the skin region detection unit 5 shown in FIG. 9, which is the skin region detection unit 5 in the determination region detection unit 4 shown in FIG.

  FIG. 10 is a configuration diagram illustrating an example of an internal configuration of the skin region detection unit 5 to which a function of excluding pixels that are low luminance signals is added when detecting a skin region. As shown in FIG. 10, the skin region detection unit 5 in the present embodiment includes a color signal R smoothing filter 52, a color signal G smoothing filter 53, a color signal B smoothing filter 54, and a hue calculation. A unit 55, a skin color determination unit 56, a color / luminance signal extraction unit 57, a low luminance signal exclusion determination unit 58, and a low luminance signal exclusion AND operation unit 59.

  The color / luminance signal extraction unit 57 extracts and outputs the skin color detection color signal R, color signal G, and color signal B from the input image signal that is input, and at the same time excludes pixels that are low luminance signals. The luminance signal is extracted and output. The skin color determination for detecting the skin area is output from the skin color determination unit 56 according to the method described in the first embodiment. If the input pixel is a skin color, true is output, and otherwise, false is output. On the other hand, the low luminance signal exclusion determination unit 58 outputs true if the input luminance signal is larger than the predetermined threshold based on a predetermined threshold relating to luminance from the input luminance signal of each pixel, and outputs false otherwise. . The output results of the skin color determination unit 56 and the low luminance signal exclusion determination unit 58 are input to the low luminance signal exclusion logical product operation unit 59, and the logical product of both is output as a skin region. That is, the condition detected as the skin region is that the pixel has an input luminance signal that is determined to be a skin color in the skin color determination and that is larger than a predetermined threshold related to luminance.

  By adopting the configuration as described above, it is possible to realize a signal processing apparatus that excludes pixels that are low-intensity signals in order to exclude the reflected image F264 of the skin that appears thinly in the housing F255 as shown in FIG. 26A. It is.

  FIG. 26B is an explanatory diagram illustrating an example of a skin image output from the signal processing unit 601 illustrated in FIG. The skin image before correction shown in FIG. 26A includes a texture portion F263 and a shine component F262. The skin image after correction output from the signal processing unit 601 shown in FIG. 26B does not change the appearance of the texture portion F263, but the shine component F262 in the skin image before correction shown in FIG. 26A is the skin image after correction. Thus, a good image can be obtained.

As described above, in the third embodiment, the luminance signal processing for the purpose of removing the shine component that removes only the shine component while leaving the texture component of the input skin image is performed, and the skin that includes the skin texture as the diagnosis target It is possible to realize a skin diagnosis system capable of performing diagnosis.
<Effect of Embodiment 3>

  According to the skin diagnosis system in the third embodiment described above, the effects (1) to (7) according to the first embodiment can be obtained, and the glossy component of the input skin image is left while being left behind. Luminance signal processing for the purpose of removing shine that removes only components can be performed, and skin diagnosis including skin texture as a diagnosis target can be performed. More details are as follows.

  (21) The skin diagnosis system includes a smartphone 40, an external device 30, a cloud server 60, and a network 50. Accordingly, the smartphone 40 can acquire a close-up image of the skin via the external device 30 and transfer the close-up image of the skin to the cloud server 60 via the network 50. The cloud server 60 acquires a close-up image of the skin through the network 50, generates skin diagnosis data from an image obtained by controlling the signal processing unit 601, and transfers the result to the smartphone 40 via the network 50. be able to.

  (22) The external device 30 has predetermined performance provided in the housing F255, the housing F255 having the eyepiece hole F251 and the objective hole F254, the ring-shaped component F256 attached to the objective hole F254, and the housing F255. It includes a lens F253 and a light source LED F252. Thus, with the smartphone 40 in contact with the eyepiece hole F251, the ring-shaped component F256 is brought into close contact with the skin, the LED F252 is turned on, and a close-up image of the skin is acquired by the image acquisition unit of the smartphone 40 via the lens F253. be able to.

  (23) The signal processing unit 601 included in the cloud server 60 can determine from the close-up image of the skin whether the exposure conditions are not suitable for skin diagnosis.

  (24) The signal processing unit 601 included in the cloud server 60 can exclude pixels that are luminance signals lower than a predetermined threshold from the close-up image of the skin.

  (25) According to the third embodiment, the skin signal diagnosis includes skin texture as a diagnosis target by performing luminance signal processing for the purpose of removing only the shine component while leaving the texture component of the input skin image. It is possible to realize a skin diagnosis system capable of performing the above.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say. For example, the above-described embodiment has been described in detail for easy understanding of the present invention, and is not necessarily limited to one having all the configurations described. Further, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. . Further, it is possible to add, delete, and replace other configurations for a part of the configuration of each embodiment.

  In addition, each of the above-described configurations may be configured such that some or all of them are configured by hardware, or are implemented by executing a program by a processor. Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

DESCRIPTION OF SYMBOLS 1 ... Smoothed luminance calculation part, 11 ... Input luminance signal extraction part, 12 ... Smoothing calculating part 2, ... Luminance signal processing part, 21 ... Luminance / color difference signal separation part, 22 ... Comparison part, 23 ... Subtraction part, 24 ... Gain multiplying unit 25 ... Adding unit 26 ... Luminance signal switching unit 27 ... Luminance / chrominance signal integrating unit 3 ... Smoothing range determining unit 31 ... Smoothing range determining unit 311 ... Input luminance signal extracting unit 312 ... High luminance signal ratio calculation unit, 313 ... smoothing range output unit, 314 ... high luminance signal ratio determination unit, 315 ... smoothing range switching unit, 316 ... average luminance determination unit, 317 ... logical product calculation unit for exposure failure determination 4 ... determination area detection part, 41 ... centroid position calculation part, 411 ... input luminance signal extraction part, 412 ... binary image generation part, 413 ... binary image centroid position calculation part, 42 ... rectangular area output part, 43 ... common area Output AND operation block DESCRIPTION OF SYMBOLS 5 ... Skin area | region detection part, 51 ... Color signal extraction part, 52 ... Smoothing filter for color signals R, 53 ... Smoothing filter for color signals G, 54 ... Smoothing filter for color signals B, 55 ... Hue calculation part, 56 ... Skin color determination unit, 57 ... Color / luminance signal extraction unit, 58 ... Low luminance signal exclusion determination unit, 59 ... Low luminance signal exclusion AND unit 6 ... Face recognition unit 10 ... Imaging device, 101 ... Image acquisition unit DESCRIPTION OF SYMBOLS 102 ... Signal processing part 103 ... Memory | storage part 104 ... Control part 1041 ... Exposure control part 1042 ... External input control part 20 ... Skin 30 ... Externally attached device 40 ... Smartphone 50 ... Network 60 ... Cloud server, 601 ... Signal processing unit, 602 ... skin diagnosis unit

Claims (15)

An image acquisition unit for acquiring an input image signal;
A signal processing unit that inputs the input image signal, performs signal processing, and outputs an output image signal;
A storage unit for storing the output image signal in a medium;
With
The signal processing unit
A smoothed luminance calculating unit that calculates a smoothed luminance by smoothing an input luminance signal within a first smoothing range around each pixel signal included in the input image signal;
The input luminance signal is compared with the smoothed luminance, and if the input luminance signal is larger than the smoothed luminance, the difference between the input luminance signal and the smoothed luminance is reduced by a first ratio, and the input luminance signal is reduced. A luminance signal processing unit that generates an output luminance signal by replacing the luminance signal;
An imaging apparatus. The imaging device according to claim 1,
The smoothed luminance calculating unit includes a smoothed range determining unit that determines a smoothed range from the input image signal, and the input image signal includes the smoothed range within the smoothed range output from the smoothed range determining unit. An imaging apparatus that smoothes an input luminance signal and calculates the smoothed luminance. The imaging device according to claim 2,
The smoothing range determination unit includes a determination region detection unit that detects a determination region from the input image signal, and is higher than a first threshold value of the determination region detected by the determination region detection unit in the input image signal. An imaging apparatus that calculates the number of pixels having a luminance signal and a ratio to the total number of pixels, and the larger the ratio, the larger the smoothing range. The imaging device according to claim 3.
The imaging apparatus, wherein the determination area detection unit outputs all pixels included in the input image signal as the determination area. The imaging device according to claim 3.
The determination region detection unit includes a skin region detection unit that detects a skin region from the input image signal, calculates a gravity center position of the skin region detected by the skin region detection unit in the input image signal, and An imaging apparatus that outputs a rectangular region having a center of gravity at the center and a side having a first length as the determination region. The imaging device according to claim 3.
The determination region detection unit includes a skin region detection unit that detects a skin region from the input image signal and a face recognition unit that outputs a face region from the input image signal, and is detected by the skin region detection unit. An imaging apparatus that outputs, as the determination area, an area in which a skin area overlapped with a face area output from the face recognition unit. The imaging device according to claim 1,
The imaging apparatus further includes an exposure control unit that calculates an exposure control amount from an output image signal output from the signal processing unit and controls the image acquisition unit. The imaging apparatus according to claim 7,
The imaging apparatus further includes an external input control unit that operates the image acquisition unit, the signal processing unit, the storage unit, and the exposure control unit. A smoothed luminance calculating unit that calculates the smoothed luminance by smoothing the input luminance signal within the first smoothing range around each pixel signal included in the input image signal;
The input luminance signal is compared with the smoothed luminance, and if the input luminance signal is larger than the smoothed luminance, the difference between the input luminance signal and the smoothed luminance is reduced by a first ratio, and the input luminance signal is reduced. A luminance signal processing unit that generates an output luminance signal by replacing the luminance signal;
A signal processing apparatus comprising: The signal processing device according to claim 9,
The smoothed luminance calculating unit includes a smoothed range determining unit that determines a smoothed range from the input image signal, and the input image signal includes the smoothed range within the smoothed range output from the smoothed range determining unit. A signal processing device that smoothes an input luminance signal and calculates the smoothed luminance. The signal processing device according to claim 10,
The smoothing range determination unit includes a determination region detection unit that detects a determination region from the input image signal, and is higher than a first threshold value of the determination region detected by the determination region detection unit in the input image signal. A signal processing apparatus that calculates the number of pixels having a luminance signal and a ratio with respect to the total number of pixels, and the larger the ratio, the larger the smoothing range. The signal processing apparatus according to claim 11,
The determination region detection unit is a signal processing device that outputs all pixels included in the input image signal as the determination region. The signal processing apparatus according to claim 11,
The determination region detection unit includes a skin region detection unit that detects a skin region from the input image signal, calculates a gravity center position of the skin region detected by the skin region detection unit in the input image signal, and A signal processing device that outputs a rectangular region having a center of gravity at the center and a side having a first length as the determination region. The signal processing apparatus according to claim 11,
The determination region detection unit includes a skin region detection unit that detects a skin region from the input image signal and a face recognition unit that outputs a face region from the input image signal, and is detected by the skin region detection unit. A signal processing device that outputs, as the determination region, a region in which the skin region overlapped with the face region output from the face recognition unit. A skin diagnosis system using the signal processing device according to any one of claims 9 to 14,
A portable terminal device including an image acquisition unit for acquiring the input image signal and a communication unit;
An external device attached to the portable terminal device and including a lens and a light source for close-up of the skin;
An information processing device including the signal processing device;
A network to which the portable terminal device and the information processing device are connected;
With
The portable terminal device acquires the close-up image of the skin via the external device, and transfers the close-up image of the skin to the information processing device via the network;
The information processing device acquires the close-up image of the skin through the network, generates skin diagnosis data from an image obtained by controlling the signal processing device, and sends the result to the mobile terminal device via the network. Transfer skin diagnosis system.