JP2005235136A - System, method and program for objectively evaluating face - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide objective evaluating system/method/program of a face, which can be used for objective evaluation of a face image. <P>SOLUTION: The objective evaluation system of the face has a face image acquiring means for acquiring the face image, a face image storing means for storing the acquired face image, a calculating means for calculating lightness in respective parts of the acquired face image, a face lightness graph creating means for creating a face lightness graph where lightness at every part of the face image can be compared based on a calculation result by the calculating means and a display means for displaying the face lightness graph created in the face lightness graph creating means. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an objective evaluation system for a face, an objective evaluation method for a face, and an objective evaluation program for a face, and more specifically, an objective evaluation system for a face that can be used for objective evaluation of a face in a digital image, The present invention relates to an objective evaluation method and a face objective evaluation program.

  Conventionally, facial impressions have been evaluated with human subjective impressions such as “clear face” and “blurred face” depending on the size of each facial part in addition to the shape of each facial part.

  Further, Patent Document 1 describes a skin color classification system as a system for classifying the skin color of a face, but the evaluation of a face cannot be evaluated.

  For example, in a photo studio that takes a commemorative photo of a customer and obtains a digital image, the customer cares not only about the costume and background at the time of shooting, but also about the photograph of his face. Photographing can be evaluated by evaluating a digital image, and customer satisfaction also depends on the quality of the face image from which the photograph of the face is evaluated. Therefore, it is conceivable to apply makeup so that the photograph will satisfy the customer.

  However, even if the makeup is finished well, the customer has only to subjectively evaluate the appearance of the facial image by looking at the facial image. The evaluation was also sensuous, and only an ambiguous evaluation such as “Somehow it became beautiful” or “Somehow it became a clear face” was obtained. For this reason, the customer's satisfaction cannot be sufficiently obtained even though the makeup is performed so that the photograph is satisfactory.

If objective evaluation is possible for such face images without subjective evaluation, the evaluation of the makeup that has been made can be shown as objective evaluation, which can improve customer satisfaction. it can.
JP 2000-4939 A

  The present inventor researched a method for objectively evaluating a face image. When a face image is generally printed and expressed as a black and white image, the dot density corresponding to the brightness is changed and expressed. Focusing on this point, that is, finding that there is brightness as one index that greatly affects the impression of the face, has led to the present invention.

  SUMMARY OF THE INVENTION An object of the present invention relates to a face objective evaluation system, a face objective evaluation method, and a face objective evaluation program that can be used for objective evaluation of face images.

  Other problems of the present invention will become apparent from the following description of the present specification.

  The above problems are solved by the following inventions.

(Claim 1)
Based on the face image acquisition means for acquiring the face image, the face image storage means for storing the acquired face image, the calculation means for calculating the brightness at each part of the acquired face image, and the calculation result by the calculation means The apparatus includes: a face brightness graph creating unit that creates a face brightness graph that enables comparison of brightness for each part of the face image; and a display unit that displays the face brightness graph created by the face brightness graph creating unit. An objective evaluation system for the face.

(Claim 2)
The objective evaluation system for a face according to claim 1, wherein the face brightness graph is a bar graph showing the brightness of each part of the face image.

(Claim 3)
The objective evaluation system for a face according to claim 1, wherein the face brightness graph is a graph of a correlation curve between each part of a face image and the brightness of each part.

(Claim 4)
It has a graph storage means for storing in advance a standard face brightness graph, which is a face brightness graph for a face image in a standard state for each subject, and the display means is stored in advance as a face brightness graph for calculation. The objective evaluation system for a face according to claim 1, 2 or 3, wherein the standard face brightness graph is compared and displayed.

(Claim 5)
A graph storage unit that stores in advance an ideal facial brightness graph that is a facial brightness graph for a facial image of an ideal face, and the display unit includes a calculated facial brightness graph and a pre-stored ideal facial brightness graph; The objective evaluation system for a face according to claim 1, 2 or 3, characterized by comparing and displaying.

(Claim 6)
The brightness of each face position of the face image is calculated, and the calculation result is used as a face brightness graph indicating the brightness for each face position, and the face is evaluated based on the face brightness graph. Objective evaluation method.

(Claim 7)
The objective evaluation method for a face according to claim 6, wherein the face brightness graph is a bar graph indicating the brightness of each part of the face image.

(Claim 8)
The objective evaluation method for a face according to claim 6, wherein the face brightness graph is a graph of a correlation curve between each part of the face image and the brightness of each part.

(Claim 9)
A standard facial brightness graph that is a facial brightness graph for a facial image of a face in a standard state for each subject is stored in advance, and the evaluation of the face is a facial brightness graph for calculation and a pre-stored standard facial brightness graph 9. The method for objective evaluation of a face according to claim 6, 7 or 8, characterized in that the smaller the difference between the two, the higher the evaluation of the face.

(Claim 10)
An ideal facial brightness graph, which is a facial brightness graph for a facial image of an ideal face, is stored in advance, and the evaluation of the face is performed by comparing a calculated facial brightness graph with a pre-stored ideal facial brightness graph. 9. The method for objective evaluation of a face according to claim 6, 7 or 8, wherein the smaller the difference is, the higher the evaluation of the face is.

(Claim 11)
A computer that calculates the brightness of each part of the face image acquired by the face image acquisition means, and the brightness of the face image that enables comparison of the brightness of each part of the face image based on the calculation result of the calculation unit An objective evaluation program for a face, which functions as a facial brightness graph creation means for creating a graph.

(Claim 12)
The objective evaluation program for a face according to claim 11, wherein the computer functions as display control means for displaying the facial brightness graph on a display means.

(Claim 13)
The objective evaluation program for a face according to claim 11 or 12, wherein the facial brightness graph is a bar graph indicating the brightness of each part of the face image.

(Claim 14)
The objective evaluation program for a face according to claim 11 or 12, wherein the face brightness graph is a graph of a correlation curve between each part of a face image and the brightness of each part.

(Claim 15)
A computer is caused to function as a graph storage unit that stores in advance a standard facial brightness graph that is a facial brightness graph for a facial image of a face that is standardized for each subject, and the display control unit includes a facial brightness graph for calculation 15. The objective evaluation program for a face according to claim 12, 13 or 14, wherein the standard face brightness graph stored in advance is compared and displayed on the display means.

(Claim 16)
The computer is caused to function as a graph storage unit that stores in advance an ideal facial brightness graph that is a facial brightness graph for a facial image of an ideal face, and the display control unit stores the facial brightness graph for calculation and the ideal image stored in advance. 15. The objective evaluation program for a face according to claim 12, 13 or 14, wherein the display is compared with a face brightness graph and displayed on the display means.

  ADVANTAGE OF THE INVENTION According to this invention, the objective objective evaluation system of the face which can be utilized for the objective evaluation of a face, the objective objective evaluation method of a face, and the objective evaluation program of a face can be provided.

  In addition, according to the present invention, it is possible to objectively recognize how close the makeup face is to the person's standard face, and it can be used to bring the facial impression closer to the person's standard face. .

  In addition, according to the present invention, it is possible to objectively recognize how close a makeup face is to the ideal face, and it can be used to bring the facial impression closer to the ideal face.

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

  FIG. 1 is a schematic hardware configuration diagram showing an example of an objective evaluation system for a face for carrying out the objective evaluation method for a face of the present invention.

  In FIG. 1, the objective evaluation system for a face of the present invention is a computer 100 for controlling each part of the objective evaluation system for a face, a flatbed scanner 200 as an example of an apparatus for acquiring a face image, and an example of a display means. And a keyboard 400 and a mouse 500 as input devices.

  The computer 100 implements the objective evaluation method of the face by controlling the flatbed scanner 200, the monitor 300, the keyboard 400, and the mouse 500 by installing the objective evaluation program for the face of the present invention. The computer 100 may be configured to be able to acquire an image from a server via a network as necessary.

  The flatbed scanner 200 is an apparatus used as an example of an apparatus that acquires a face image. By operating and instructing the keyboard 400 and the mouse 500, for example, a face image can be acquired by scanning a photograph of a face photographed. As an apparatus for acquiring a face image, in addition to the flatbed scanner 200, other scanner apparatuses such as a film scanner, and a recording capable of acquiring an image recorded on an external recording medium such as a CD-R or a memory card. A network interface or the like that can acquire an image recorded in a medium reading device or a server connected on a network such as the Internet can be used.

  The monitor 300 is a display device that functions as a display unit and displays the face image acquired by the flatbed scanner 200 or the like based on the control of the computer 100, operation details, and the like as necessary. The monitor 300 has a touch panel function as necessary, and can also serve as an input device.

  FIG. 2 is a block diagram illustrating an example of a functional configuration of a computer that can be used in the objective objective evaluation system for a face.

  In FIG. 2, a computer 100 stores a main control unit 101 for controlling each part of the computer 100, an image input unit 102 for inputting an image, images and various data, and an objective evaluation program for a face of the present invention. A main storage unit 103 for acquiring a facial image, a facial image acquisition processing unit 104 for acquiring a facial image, a facial brightness calculating unit 105 that scans the acquired facial image and calculates the brightness of each part of the face, and the calculation by the calculating unit Based on the result, a face brightness graph creation processing unit 106 that creates a face brightness graph that makes it possible to compare the brightness of each part of the face image, a data input / output unit 107 that inputs and outputs data, and a display that is displayed on the monitor 200 A display control unit 108 for controlling is provided. Reference numeral 109 denotes an external communication unit for exchanging data via a network, which is preferably provided.

  The image input unit 102 and the face image acquisition processing unit 104 constitute an image acquisition unit in cooperation with the flatbed scanner 200. Specifically, the face image acquisition processing unit 104 reads an image from a recording medium such as a photograph with the flatbed scanner 200 and inputs the read image to the computer 100 via the image input unit 102 to thereby convert the face image. get. The input image can be stored in the main storage unit 103 as a face image storage unit. Of course, other scanner devices, recording medium reading devices, network interfaces, and the like can be used instead of the flatbed scanner 200 constituting the image acquisition means.

  The main storage unit 103 functions as a face image storage unit that stores a face image and a graph storage unit that stores a standard face brightness graph and / or an ideal face brightness graph in advance.

  The main storage unit 103 may store the ideal face brightness graph and / or the ideal face brightness graph in advance when the program is installed, acquire a standard or ideal face image, and The face brightness of the face image may be calculated, and a standard face brightness graph and / or an ideal face brightness graph may be created respectively.

  The standard face image is a face image of an arbitrarily selected subject. For example, it is possible to obtain a photograph that the subject feels that the photograph is good and read the photograph with the flood head scanner 200.

  The ideal face image is a face image other than the arbitrarily selected subject. For example, it can be acquired by reading a picture of an actress or an idol with the flatbed scanner 200.

  The facial brightness calculation unit 105 functions as a calculation unit. Specifically, the brightness of each part of the face image is calculated by scanning the face image acquired by the image acquisition unit according to an aspect described later. Based on the calculated brightness of each part of the face image, the face brightness graph creation processing unit 106 as a face brightness graph creating unit 106 creates a face brightness graph that makes it possible to compare the brightness of each part of the face image. A graph is created.

  The face lightness graph created by the face lightness graph creation processing unit 106 is a graph in which the lightness of each part of an arbitrary face image can be compared.

  The parts of the face image are elements constituting the face such as “hair”, “forehead”, “eyebrows”, “eyes”, “cheek”, “lips”, “chin”, “neck”, etc. Can be arbitrarily defined.

  The lightness of each part can be compared by representing the lightness of at least two parts in one graph.

  The data input / output unit 107 is a port or the like for serving as a data entry / exit port when input from the keyboard 400 and the mouse 500 or other various input / output devices.

  The display control unit 108 functions as a display control unit and controls display contents on the monitor 300.

  The external communication unit 109 is a device that controls input / output of signals when performing communication with an external device via a network as a network interface.

  FIG. 3 is a diagram showing an operation flow of the objective objective evaluation system for face controlled by the objective objective evaluation program for face according to the present invention.

  The objective evaluation system for a face according to the present invention is used, for example, when a photographer or the like takes a photograph of a subject to explain how much the objective impression before and after makeup is different by an operator. Therefore, description will be made based on such an example.

  In FIG. 3, the face image of the subject is acquired by the face image acquisition means (S1).

  The acquired face image may be a face image 10 as shown in FIG. 4, for example, and may be an image including at least a face portion having a height including the jaw and a width including both ears from the forehead. In this embodiment, since the objective impression of the face image before makeup and the face image after makeup is compared, the face image before makeup and the face image after makeup are acquired. As for the acquisition timing, two images may be acquired at the same time, or after the face image before makeup is processed, the face image after makeup may be acquired. An example in which the face image after makeup is acquired after the processing of the face image will be described as an example.

  As a trigger for acquiring a face image, for example, an input of an operator's instruction to acquire a face image is accepted by an input unit such as a keyboard 400 or a mouse 500, or the computer 100 automatically determines that the timing is predetermined. It is mentioned.

  When there is a face image acquisition trigger, the face image acquisition processing unit 104 controls the flat bed scanner 200 to scan a recording medium such as a photograph and acquires the face image via the image input unit 102.

  As a mode of acquiring the face image, in addition to a mode of acquiring by scanning the recording medium with the flatbed scanner 200, a mode of acquiring with a recording medium reading device or a network interface may be used.

  The acquired face image is stored by the face image storage means (S2). That is, the acquired face image is stored in the main storage unit 103.

  The brightness of each part of the face in the face image stored in S2 is calculated by the calculating means (S3).

  Specifically, the face brightness calculation unit 105 continuously scans the face image stored in the main storage unit 103 in S2, and calculates the brightness of each part. 5 and 6 are diagrams illustrating an example of a scanning method.

  Scanning means obtaining information for calculating the brightness of each part of the image.

  Scanning continuously means scanning so that the brightness of at least two parts of the same face image can be calculated.

  The mode of scanning in the face brightness calculation unit 105 is performed by continuously scanning each part of the face image, and is not particularly limited, but from the viewpoint that scanning can be easily performed. As shown in FIG. 5A and FIG. 5B, it is preferable to scan in a straight line. Further, from the viewpoint of expressing features better, it is generally preferable to perform scanning so that as many portions as possible of facial features are included as shown in FIGS. 6A and 6B. In FIG. 5A, one side of the face image is scanned in a straight line, and in FIG. 5B, the face image is scanned in a straight line obliquely. In FIG. 6 (a), the facial feature points of both eyes, cheeks and lips are scanned once in a line formed by connecting a plurality of straight lines. In FIG. 6 (b), The face feature points, forehead, left eye, left cheek and lips are scanned once in a line composed of multiple straight lines, and multiple points for the forehead, right eye, right cheek and lips are similarly scanned. A value obtained by scanning once with a line formed by connecting these straight lines and averaging the results of a plurality of scans for each part is adopted as the scan result.

In the present invention, the lightness may be an index representing the degree of brightness. For example, in the L ^* a ^* b ^* color system, the lightness index of the lightness index L ^* and the chromaticness index a ^* b ^* index. L ^* can be employed, and in the Munsell color system, the lightness V of hue H, lightness V, and saturation C can be employed, but in other color systems, the degree of brightness is also expressed. An index can be adopted and is not particularly limited.

  When the brightness of each part of the face image is calculated in S3, a face brightness graph that makes it possible to compare the brightness of each part of the face image based on the calculation result is created by the face brightness graph creation processing unit ( S4).

  The face brightness graph is a graph showing the correlation between each part of the face image and the brightness of the face image of the part, and the brightness of each part can be compared. For example, FIG. It can be set as a graph as shown in. In FIG. 7A, each part and the brightness value of the part are represented as a graph of a correlation curve. In FIG. 7B, the brightness value of each part and the part is shown for each part of the face. It represents as a bar graph shown in.

  In the graph shown in FIG. 7A, the vertical axis indicates each part, and “hair”, “forehead”, “brow”, “eyes”, “cheek”, “lips” are the parts in the scanning order from the vicinity of the origin. "," Chin "and" neck "are plotted. On the other hand, the horizontal axis shows the lightness, and each of “hair”, “forehead”, “brow”, “eyes”, “cheek”, “lips”, “chin”, “neck” shown on the vertical axis The brightness values at the sites are plotted and each value is connected by a correlation curve. Also in the graph shown in FIG. 7B, as in the graph of FIG. 7A, the vertical axis indicates “hair”, “forehead”, “brow”, “eyes”, “cheek”, “ Lips, chin, and neck are plotted, and lightness values at each part are plotted on the horizontal axis, and each value is shown as a bar graph. In FIG. 7, “hair”, “forehead”, “eyebrow”, “eyes”, “cheek”, “lips”, “chin”, and “neck” shown as each part are examples, and scanning It depends on the method.

  The face brightness graph created in this way is continuously scanned for each part of the face, and the brightness is expressed as a face brightness graph in which the brightness of each part can be compared. (Tone difference) can be objectively captured.

  The face brightness graph created in S4 is displayed by the display means (S5).

  In this way, after the processing of the face image before makeup is completed, the face image after makeup is processed in the same manner as the processing of the face image before makeup.

  The created face brightness graph is displayed on display means such as the monitor 300 by the display control unit 108. FIG. 8 is a diagram showing an example in which a face brightness graph before makeup and a face brightness graph after makeup are compared and displayed. In FIG. 8, the face brightness graph before makeup is shown by a dotted line, and the face brightness graph after makeup is shown by a solid line. In FIG. 8, the portion where the brightness is increased from before the makeup is indicated by diagonal lines, and the portion where the brightness is decreased is indicated by a stitch pattern.

  With the objective evaluation system for a face of the present invention, a face brightness graph before makeup and a face brightness graph after makeup can be displayed, and advice can be given to the subject based on the displayed face brightness graph. Each face lightness graph is objectively expressed as a graph of lightness and darkness (tone difference) for each face, so that the impression of the whole face changed before and after makeup, Since the degree of change can be visually confirmed with a graph, it is easy for the operator to explain to the subject. Further, it is preferable to appropriately perform processing such as changing the scale of the axis of the graph or color-coding the graph because the tone difference can be more clearly visually recognized.

  Further, in this embodiment, the face brightness graph is created and displayed for the face image before makeup and the face image after makeup, but this is a face brightness graph for the face image of the face that is the standard of the person in advance. A standard facial brightness graph may be created, the facial brightness graph may be stored in the main storage unit 103, and the standard facial brightness graph and the created facial brightness graph may be compared and displayed. At this time, the face image of the person's standard face is obtained, for example, by having the subject bring a photograph taken at the time of the photograph taken by the subject and reading the photograph with the flatbed scanner 200. Also good. Then, make up with reference to the photo, compare the face brightness graph after makeup with the standard face brightness graph, and objectively see how close the face image was when the photo was taken with this makeup. I understand.

  In that case, for example, by applying makeup so that the created face brightness graph is the same as the standard face brightness graph, it can be used to bring the facial impression closer to the standard face of the person.

  In addition, when performing makeup, by comparing a face brightness graph created based on a face image before makeup with a standard face brightness graph, for example, an objective evaluation such as “cheak brightness is insufficient” is performed. Therefore, cosmetics can be selected to correct cheek brightness based on this evaluation. Therefore, it is easy to apply makeup so as to obtain a favorite photograph.

  Also, an ideal facial brightness graph, which is a facial brightness graph for a facial image of a face to be aspired, such as an actress's facial image, is created and stored in the main storage unit 103 in advance. You may compare and display with the created face brightness graph. By comparing the face brightness graph after makeup and the ideal face brightness graph, it is possible to objectively know how close the face image is to the ideal. In that case, makeup can be applied so that the created face brightness graph is the same as the ideal face brightness graph, so that the facial impression can be made closer to the ideal face.

  In this way, the present invention enables objective evaluation and comparison of the impression of a face image without being influenced by the shape and skin color of each part of the face. This is because the impression of the face depends on the difference in brightness of the face (tone difference). This is due to new knowledge obtained by the present inventors.

Hardware schematic block diagram showing an example of an objective evaluation system for a face for carrying out the objective evaluation method for a face of the present invention A block diagram showing an example of a functional configuration of a computer that can be used in an objective evaluation system for a face The figure which shows the operation | movement flow example of the objective evaluation system of the face controlled by the objective evaluation program of the face of this invention The figure which shows an example of the acquired face image (A), (b) is a figure which shows the example of a scanning method, respectively. (A), (b) is a figure which shows the example of a scanning method, respectively. (A), (b) is a figure which shows the example of a face brightness graph, respectively The figure which shows the example which compared and displayed the face brightness graph before makeup, and the face brightness graph after makeup

Explanation of symbols

DESCRIPTION OF SYMBOLS 100: Computer 101: Main control part 102: Image input part 103: Main memory part 104: Face image acquisition process part 105: Face lightness calculation part 106: Face lightness graph creation process part 107: Data input / output part 108: Display control part 109: External communication unit 200: Flatbed scanner 300: Monitor 400: Keyboard 500: Mouse

Claims (16)

  Based on the face image acquisition means for acquiring the face image, the face image storage means for storing the acquired face image, the calculation means for calculating the brightness at each part of the acquired face image, and the calculation result by the calculation means The apparatus includes: a face brightness graph creating unit that creates a face brightness graph that enables comparison of brightness for each part of the face image; and a display unit that displays the face brightness graph created by the face brightness graph creating unit. An objective evaluation system for the face.   The objective evaluation system for a face according to claim 1, wherein the face brightness graph is a bar graph showing the brightness of each part of the face image.   The objective evaluation system for a face according to claim 1, wherein the face brightness graph is a graph of a correlation curve between each part of a face image and the brightness of each part. It has a graph storage means for storing in advance a standard face lightness graph that is a face lightness graph for a face image of the face as a standard for each subject,
The objective evaluation system for a face according to claim 1, 2 or 3, wherein the display means compares and displays a face brightness graph for calculation and a standard face brightness graph stored in advance. A graph storage unit that stores in advance an ideal facial brightness graph that is a facial brightness graph for a facial image of an ideal face;
The objective evaluation system for a face according to claim 1, 2 or 3, wherein the display means compares and displays a face brightness graph for calculation and an ideal face brightness graph stored in advance.   The brightness of each face position of the face image is calculated, and the calculation result is used as a face brightness graph indicating the brightness for each face position, and the face is evaluated based on the face brightness graph. Objective evaluation method.   The objective evaluation method for a face according to claim 6, wherein the face brightness graph is a bar graph indicating the brightness of each part of the face image.   The objective evaluation method for a face according to claim 6, wherein the face brightness graph is a graph of a correlation curve between each part of the face image and the brightness of each part.   A standard facial brightness graph that is a facial brightness graph for a facial image of a face in a standard state for each subject is stored in advance, and the evaluation of the face is a facial brightness graph for calculation and a pre-stored standard facial brightness graph 9. The method for objective evaluation of a face according to claim 6, 7 or 8, characterized in that the smaller the difference between the two, the higher the evaluation of the face.   An ideal facial brightness graph, which is a facial brightness graph for a facial image of an ideal face, is stored in advance, and the evaluation of the face is performed by comparing a calculated facial brightness graph with a pre-stored ideal facial brightness graph. 9. The method for objective evaluation of a face according to claim 6, 7 or 8, wherein the smaller the difference is, the higher the evaluation of the face is.   A computer that calculates the brightness of each part of the face image acquired by the face image acquisition means, and the brightness of the face image that enables comparison of the brightness of each part of the face image based on the calculation result of the calculation unit An objective evaluation program for a face, which functions as a facial brightness graph creation means for creating a graph.   The objective evaluation program for a face according to claim 11, wherein the computer functions as display control means for displaying the facial brightness graph on a display means.   The objective evaluation program for a face according to claim 11 or 12, wherein the facial brightness graph is a bar graph indicating the brightness of each part of the face image.   The objective evaluation program for a face according to claim 11 or 12, wherein the face brightness graph is a graph of a correlation curve between each part of a face image and the brightness of each part.   A computer is caused to function as a graph storage unit that stores in advance a standard facial brightness graph that is a facial brightness graph for a facial image of a face that is standardized for each subject, and the display control unit includes a facial brightness graph for calculation 15. The objective evaluation program for a face according to claim 12, 13 or 14, wherein the standard face brightness graph stored in advance is compared and displayed on the display means.   The computer is caused to function as a graph storage unit that stores in advance an ideal facial brightness graph that is a facial brightness graph for a facial image of an ideal face, and the display control unit stores the facial brightness graph for calculation and the ideal image stored in advance. 15. The objective evaluation program for a face according to claim 12, 13 or 14, wherein the display is compared with a face brightness graph and displayed on the display means.

Images