JP3868089B2 - Face determination method and face determination support device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a technique for diagnosing the beauty or aging of a face or identifying a face.
[0002]
BACKGROUND OF THE INVENTION
The inventor of the present application has continued research for many years on factors that determine the impression of a human face, particularly its beauty, from the standpoint of plastic surgery and cosmetic surgery. As a result, it is possible to objectively evaluate the degree of beauty and aging, etc. by analyzing the light and dark conditions that reflect the skin properties and irregularities on the face, that is, the states of the bright and dark areas. We found that it is possible to objectively evaluate facial expressions that can be used for fortune-telling and other purposes, to identify individual faces, to evaluate facial expressions useful for emotion determination and quantification.
[0003]
Analyzing the faces of models and actors, which are generally recognized as beautiful, with respect to light and darkness, they can find a great deal in common. That is, when a face image is represented by a region partitioned by light and dark by applying image processing to the face image, the contour shape of the region is in order. More specifically, there are three types of faces, roughly divided into light and dark areas. As schematically shown in FIG. 5, the shape of the bright region is an inverted triangle type (FIG. 5A), an oval type (FIG. 5B), and a cloverleaf type (FIG. 5). 5 (c)]. And in the face generally called a beauty, the shape of the bright region or dark region of the type in the face is arranged. In other words, for example, in the case of an inverted triangle type, the presence or absence of distortion or disorder in the inverted triangle shape is a major factor that brings a beauty impression. Therefore, by using the face image represented by the bright and dark area, it is possible to objectively determine the impression of the face such as the degree of beauty. The same is true for the degree of aging of the face, that is, the degree of wrinkles and skin sag, and as the degree of aging increases, the shape of the light region or dark region tends to become more distorted. The degree of aging can be determined. Even when the other elements are evaluated, they are common in that the bright region or dark region of the face is an important material.
[0004]
In other words, it can be said that the face of each individual has a state such as a shape of a bright region or a dark region unique to the individual or a disorder thereof. This means that each person's face can be identified based on the shape of the bright region to the dark region or the disorder thereof.
[0005]
In addition, the shape of the bright region to the dark region can be changed according to the change of facial expression depending on the way of focusing on the shape. That is, the facial expression is accompanied by the contraction and relaxation of the facial expression muscle. This change in the facial expression muscle causes a change in the unevenness of the face, that is, a change in the bright region to dark region. Therefore, by focusing on this, even an ambiguous expression such as a facial expression can be quantified and objectively evaluated.
[0006]
[Problems to be solved by the invention]
The present invention has been made with the intention of making effective use of the above-mentioned knowledge, and in particular, the above-mentioned knowledge, for example, diagnosis of facial impression and skin condition, and advice on problems and skin care. To make decisions, to make makeup, to give advice on how to make up, to use for surgical planning simulations in plastic surgery and cosmetic surgery, and to automatically identify individual faces Its purpose is to make it available, and to use it for objective evaluation of facial expressions. More specifically, it is an object of the present invention to provide a face determination method and a face determination support device using the above-described knowledge.
[0007]
[Means for Solving the Problems]
In the face determination method according to the present invention, a determination face image obtained by processing a face image brightly and darkly is used. For this purpose, first, the face to be determined is imaged to obtain a face image. Next, the face image is subjected to image enhancement processing, in particular, image enhancement processing relating to light and dark, thereby creating a determination face image represented by a plurality of regions distinguished in terms of brightness. Then, based on the contour shape of the area in the determination face image and the state of the boundary between these areas, the face is determined for each of the above-mentioned objective matters such as the beauty degree and aging degree of the face.
[0008]
For the image enhancement processing in such a face determination method, any method can be used as long as the face is divided into bright and dark, but for example, use of a density gradation conversion method is convenient. Among them, the threshold processing method is easy to use, but other methods such as a sharpening method, a histogram method, an edge enhancement method, a filtering method, and a contour line processing method can also be used.
[0009]
The image enhancement processing is intended for facial image segmentation processing based on brightness and darkness, and needs to be at a level at which it is easy to discriminate the contours of regions having different brightness. Such image enhancement processing is binarization using one threshold value in the case of the threshold processing method, but the number of threshold values can be increased to about three. The number of areas related to brightness corresponds to the number of thresholds and the like. Therefore, in the case of binarization, there are two areas, a bright area and a dark area.
[0010]
By the way, the light and darkness of the face has a contour distribution according to the three-dimensional shape of the face. For this reason, it can be said that the contour of the region obtained by the image enhancement process corresponds to the light and dark contour lines. Therefore, in the case of the threshold processing method, the outline of the region can be moved from the front of the face to the rear, for example, by sequentially changing the threshold from the higher to the lower. A region can be obtained in a scanning manner for the entire face.
[0011]
A determination support apparatus according to the present invention for supporting the face determination method as described above includes an imaging unit that captures an image of a face to be determined, and performs image enhancement processing on the face image obtained by the imaging unit, thereby increasing brightness. An image processing means for creating a determination face image represented by a plurality of areas distinguished from each other and a display means for displaying the determination face image. Note that it is possible to further add a determination means for automatically determining the ratio between the bright area and the dark area, the disturbance of the boundary shape, and the like based on the displayed determination face image. According to this, it is possible to easily quantify the boundary state between the bright region and the dark region, and it becomes possible to perform more objective determination regarding an ambiguous element when using the face determination method.
[0012]
[Embodiment]
According to one embodiment of the present invention, the determination support apparatus includes an imaging unit 1, an image processing unit 2, a threshold setting unit 3, and a display unit 4, as shown in FIG. For the imaging means 1, it is preferable to use a video camera or a digital camera using, for example, a CCD. The image processing means 2 includes a memory for storing the output from the imaging means 1 as a face image, a microcomputer for performing image enhancement processing on the face image, specifically threshold processing. As the display means 4, a CRT display device, a liquid crystal display device, or the like can be used.
[0013]
In order to determine a face using such a determination support apparatus, first, the face of the person to be determined is imaged by the imaging means 1 from the front, for example. The output from the imaging means 1 is input to the image processing means 2 and stored in the memory there. If a face image is obtained in this way, image enhancement processing by threshold processing is performed on the face image. For this purpose, a threshold value is first set by the threshold value setting means 3. The threshold value is set using, for example, 256 gradation levels in a general image display system. If the threshold value is set to 180, for example, in the case of a black and white image, all pixels greater than 180 are white and all pixels lower than 180 are black. By such image processing, a determination face image represented by a plurality of regions distinguished with respect to brightness is obtained. This determination face image is displayed on the screen of the display means 4, and the appearance of the face and the degree of aging are determined while looking at the image.
[0014]
FIG. 2 and FIG. 3 show examples of binarization type determination face images for determining the makeup method. FIG. 2 shows a case where the threshold value is raised, and FIG. 3 shows a case where the threshold value is lowered. In each figure, an ideal makeup is applied based on the face image for determination (left side of the figure) for the face before applying the ideal makeup and the makeup method obtained from the determination for the face image for determination. The face images for determination after having been performed are shown side by side. As shown in FIG. 2, in this example, the bright region is basically the above-described clover leaf type, but the shape of the bright region on the face before applying ideal makeup is compared to the ideal shape in the clover leaf type. It has collapsed. Further, the boundary between the bright region and the dark region is not smooth, and there is a fine disturbance. Such a collapse of the shape of the bright region and the presence of fine disturbances at the boundary between the bright region and the dark region can also be applied to FIG. From this, it can be determined that the face before applying ideal makeup is low in beauty, and makeup is applied in a direction that eliminates the above-mentioned disruption of the shape of the bright region and the disturbance at the boundary between the bright region and the dark region. Thus, it can be determined that it is possible to increase the aesthetic degree to the maximum. On the other hand, looking at the face image for judgment after applying the ideal makeup, the shape of the bright area is close to the ideal shape in the cloverleaf type, and the boundary between the bright area and the dark area is smooth, It can be determined that the beauty has been improved by ideal makeup. According to the actual judgment example, the judgment by these almost coincided with the judgment result of the beauty evaluation by the person and a third party using an ordinary photograph.
[0015]
Here, as can be understood from a comparison between FIG. 2 and FIG. 3, the higher the threshold value, the narrower the bright region range, and the lower the threshold value, the wider the bright region range. This can be compared to a contour line. Therefore, by sequentially changing the threshold value from the higher side to the lower side, the outline of the bright region can be moved from the front of the face to the rear, thereby scanning the entire face.
[0016]
FIG. 4 shows an example of an aging degree determination face image. In the figure, the determination face image (left side of the figure) for the face before the wrinkle removal operation and the determination face image after the wrinkle removal operation are shown side by side. As shown in FIG. 4, the face before the wrinkle removal operation has a large meandering in the outline of the bright area, there are many fine disturbances at the boundary between the bright area and the dark area, and there is a small dark area in the bright area. Scattered. These reflect wrinkles, skin slack, and the like, and from these, it is possible to determine the level of aging. On the other hand, in the face after the wrinkle removal operation, the contour of the bright region is a curve with less meandering. In addition, the disturbance of the boundary between the light area and the dark area is reduced, and the small dark area disappears in the light area. From this, it can be determined how much the level of aging has been improved.
[0017]
Here, the level of aging can be roughly quantified from the threshold value and the degree of dark areas scattered in the bright area. That is, for example, if there is a dark area of a certain level or more in the bright area at the threshold value R1, and if it disappears at the threshold value R2, then the aging level is A, and if the threshold value R2 is greater than a certain value If the dark region is in the bright region and disappears if the threshold value is set to R3, the aging level is B. This is to quantify the darkness of shadows caused by large and small unevenness, and it is possible to quantify the beauty level and the unevenness level of the skin as well as quantifying the darkness of shadows that create facial expressions. Can be quantified.
[0018]
【The invention's effect】
As described above, according to the present invention, the knowledge that the impression about the beauty degree or aging degree of a face can be measured from the state of light and darkness on the face is obtained as a surgical plan simulation in facial makeup and cosmetic surgery, and further, It can open the way for effective use for identification and objective evaluation of facial expressions.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a determination support apparatus according to an embodiment.
FIG. 2 is a diagram illustrating an example of a determination face image.
FIG. 3 is a diagram showing an example of a determination face image with a threshold value different from that in FIG. 2;
FIG. 4 is a diagram of another example of a determination face image.
FIG. 5 is a diagram schematically showing a typical bright and dark area of a face.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Imaging means 2 Image processing means 3 Threshold value setting means 4 Display means

Claims (4)

A method executed by an apparatus for determining the beauty of the face of a person who has captured the face image by performing image enhancement processing on the face image obtained by imaging the face to be determined,
The device performs,
A process of creating a determination face image represented by a plurality of regions distinguished from each other in terms of brightness, from the face image;
Comparing the determination face image with an ideal face image for an ideal face and determining how close the determination face image and the ideal face image are;
Based on the result of the determination, the step of quantifying the degree of beauty of the face of the person who captured the face image, assuming that the degree of beauty is higher as the determination face image is closer to the ideal face image;
A method for judging the beauty including A method executed by an apparatus for determining the beauty of the face of a person who has captured the face image by performing image enhancement processing on the face image obtained by imaging the face to be determined,
The device performs,
A process of creating a determination face image represented by a plurality of regions distinguished from each other in terms of brightness, from the face image;
A process of determining the smoothness of boundaries between the plurality of regions in the determination face image;
Based on the result of the determination, a process of quantifying the beauty of the face of the person whose face image has been captured, assuming that the beauty is higher as the boundary line is smoother,
A method for judging the beauty including A beauty degree determination device that determines the beauty degree of a face of a person who has captured the face image by performing image enhancement processing on a face image obtained by imaging the face to be determined,
Means for creating, from the face image, a determination face image represented by a plurality of areas distinguished with respect to brightness;
Means for comparing the determination face image with an ideal face image for an ideal face and determining how close the determination face image and the ideal face image are;
Means for quantifying the degree of beauty of the face of the person whose face image has been captured, based on the result of the determination, assuming that the degree of beauty is higher as the determination face image is closer to the ideal face image;
A beauty degree determination device. A beauty degree determination device that determines the beauty degree of a face of a person who has captured the face image by performing image enhancement processing on a face image obtained by imaging the face to be determined,
Means for creating, from the face image, a determination face image represented by a plurality of areas distinguished with respect to brightness;
Means for determining smoothness of boundaries between the plurality of regions in the determination face image;
Means for quantifying the aesthetics of the face of the person whose face image has been captured, based on the result of the determination, assuming that the more beautiful the boundary is, the higher the aesthetics;
A beauty degree determination device.

Images