JP3424913B2 - Photo processing method and photo processing device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a photographic processing method including image processing and a photographic processing apparatus having an image processing section.

[0002]

2. Description of the Related Art Conventionally, in an image display medium, an image of a human face or the like is subjected to image processing by using pattern recognition technology, and the image is corrected into an image which is easy to see on the image display medium. There is an image processing technique for modifying the image to protect the portrait right or adding artistic decoration.

According to such an image processing technique, for example, in a technical field such as computer graphics, digital data is used as data relating to an image, and based on this data, enlargement / reduction, rotation, inversion, projection of the image are performed. It is possible to perform various three-dimensional operations such as shadowing.

Further, in the technical field of photographic processing apparatus, for example, the color of the pupil of the eye is corrected in the case of being affected by strobe light at the time of taking a picture or wearing a color contact. Since there is a demand for producing color prints after that, the image processing technique as described above has been required.

Also in the technical field of photographic processing apparatus, image information recorded on a film in an analog manner is read by using a scanner or the like to convert the image information into an electric signal, and based on this electric signal. A technique for producing a print by exposing and developing the obtained digital data by using image data obtained by performing image processing has become widespread.

By the way, in such an image processing technique, as a pattern recognition technique advances, there are various techniques for identifying and extracting a certain part of a human face, for example, only the eye, and performing image processing on only that region. Although proposed, a simple method that can be executed by an unskilled person without specialized knowledge as described later by a general-purpose inexpensive device such as a personal computer equipped with an image display means. Is hard to get.

That is, in the technical field of photographic processing, photographic processing apparatuses for automatically developing and printing photographic light-sensitive materials are rapidly spreading, and unskilled persons who do not have specialized knowledge of photographic processing and image processing are Since it has become possible to produce color prints in a short time and at low cost by operating such an automatic photographic processing apparatus, it is required that even an unskilled person can easily operate the equipment to perform image processing. ing.

[0008] For example, the outline of a technique for performing some image processing by specifying only the eyes of the image of the human face displayed on the image display medium will be described with reference to FIG. Reference numeral 10 indicates an eye, and 20 indicates an eyebrow.

In the image as described above, the operator visually recognizes the part of the eye 10 and, for example, with a pointing device such as a mouse, around the eye region,
For example, points R _{1 to} R ₈ are designated as a plurality of points at positions as close as possible to the eye 10, and a region surrounded by a polygon formed by sequentially connecting the plurality of points R _{1 to} R ₈ is defined. Area of the eye 10 on the image is specified.

At this time, since the image display medium temporarily digitizes the image obtained by the image pickup device as image data and re-images it, the image data is merely a numerical value. Therefore, when the plurality of points are specified on the image display medium, if the area other than the eye is specified widely, the area other than the eye is excluded when the image processing is performed after the area is specified. It takes a lot of time to process the data. Therefore, when specifying a plurality of points as described above, it is necessary to specify the points as close to the eyes as possible in order to perform data processing quickly and accurately.

However, there is a problem in that skillfully specifying the plurality of points by the pointing device on the image display medium requires skill, and it takes a long time to fail to specify them. Moreover, even if it takes a long time, the designation accuracy does not improve.

[0012]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and specifies and extracts, for example, a pupil region from the eye region of a display image without requiring skill. It is an object of the present invention to provide a photographic processing method and a photographic processing device that perform image processing by performing image processing with high accuracy in a short time.

[0013]

In order to solve the above-mentioned problems, a photographic processing method according to claim 1 of the present invention comprises a display step 33 for displaying an image of a subject and an image processing of the displayed images. In the photographic processing method, including a specifying step 36 for specifying a processing area and an image processing step 37 for image-processing the processing area to form image data for forming a photograph, in the specifying step 36, the processing area Are specified, and a rectangular area determined based on the two points and a processing area standard pattern located in the center of the rectangular area are displayed as an image, and the processing area standard pattern and the displayed image are displayed. Pattern matching is performed to specify the processing area.

With the above-described structure, the pattern matching of the processing area can be performed only by designating two points sandwiching the processing area at positions near the processing area that are easy to specify.

According to the photographic processing method of the second aspect of the present invention, the subject is a human eye 10, the processing area is at least one of a pupil area 13 and an iris area 12, and Since the points are located in the white-eye area 11, it is possible to perform processing for each eye, and it is easy to specify two points that sandwich the processing area.

According to the photographic processing method of the third aspect of the present invention, since the processing area standard pattern is the template 17 and the processing area can be specified by performing template matching, only a small number of templates are held. Thus, the area of the pupil or iris can be specified accurately in a short time.

The photographic processing method according to claim 4 of the present invention is
A first display step 33 for displaying an image of a subject, a target area designating step 34 for designating a target area including a processing area for image processing in the displayed image, and extracting the target area and enlarging the image. Second display step 35 of displaying and specifying step 3 of specifying the processing area from the target area
6 and an image processing step 37 of forming an image data for forming a photograph by performing image processing on the processing area, and in the specifying step 36, two points sandwiching the processing area are designated, The processing area is displayed by displaying a rectangular area determined based on the two points and a processing area standard pattern located in the center of the rectangular area, and pattern matching is performed between the processing area standard pattern and the displayed image. Is specified.

With the above-mentioned configuration, even if the image of the eye is small, the pattern matching of the processing area can be performed only by designating two points sandwiching the processing area at positions which are easy to designate in the vicinity of the processing area.

According to the photographic processing method of claim 5 of the present invention, the subject is a human face, the target area is the human eye 10, and the processing area is a pupil area 13 and an iris area 12. Since at least one of them is located in the white-eye area 11, it is possible to perform processing for each eye, and it is easy to specify two points sandwiching the processing area.

According to the photographic processing method of the sixth aspect of the present invention, the processing area standard pattern is used as the template 17, and only a small number of templates that specify the processing area by performing template matching are held, The area of the pupil or iris can be specified accurately in a short time.

According to the photographic processing method of the seventh aspect of the present invention, in the target area designating step 34, predetermined two points are designated, and a line segment defined by the two points,
Since the target area is specified based on the positional relationship with the target area, the target area can be easily specified by specifying only two predetermined points, and it is not necessary to specify a plurality of points in the vicinity of the target area.

According to the photographic processing method of claim 8 of the present invention, in the target area designating step 34, predetermined four points are designated, and a line segment defined by the four points and the target area are designated. The target area can be easily specified only by specifying two predetermined points that specify the target area based on the positional relationship between them, and it is not necessary to specify a plurality of points in the vicinity of the target area.
It is possible to display an icon that clearly displays the target area.

A photographic processing apparatus according to claim 9 of the present invention is
In the photographic processing device provided with an image processing unit 50 capable of performing image processing on image data of a subject, the image processing unit 5
0 is an image display unit 51 for displaying an image of the subject,
A pointing device 52 capable of designating an arbitrary point and area on the image display unit, an arithmetic processing unit 53 for arithmetically processing image data for image processing, and a storage unit 54 for storing predetermined data regarding the image. A processing area for image processing of the image is displayed, and when two points sandwiching the processing area are designated by the pointing device 52, a rectangular area defined by the two points and a central portion of the rectangular area are positioned. It is characterized in that the processing area standard pattern is displayed, and the processing area can be specified by performing pattern matching to perform image processing.

According to the above-mentioned structure, the processing area is sandwiched between 2
The area of the pupil or iris can be specified easily in a short time and with high accuracy simply by designating the points by a normal image processing means.

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the accompanying drawings.

In the present specification, with respect to an image of a human face, the head of the face (including the head) is placed on the upper side, the jaw is placed on the lower side of the human body, and the right side of the face is directed toward the image. Or left. Therefore, in this specification, the left or right of the face does not mean the left or right of the person who is the subject.

FIG. 1 is a block diagram schematically showing the construction of a photographic processing apparatus according to the present invention. The photographic processing apparatus according to the present invention includes an image processing unit 5 described in detail below.
The processing device 61 automatically performs various chemical treatments such as development, fixing, washing with water, and drying in a series of steps as in the case of a conventionally known photographic processing device except that 0 is provided. Alternatively, color separation is performed by separating and metering each pixel of a color image into three primary colors of blue, green, and red, and for each pixel, the light intensity of blue, green, and red is used as a color signal for each color. A photo processing unit 60 including a color scanner device 62 and the like for sequentially reading and a control unit 70 are provided.

Then, after converting each color signal into a digital signal, image processing is performed as described later, and then the photographic light-sensitive material is scanned and exposed based on the image data subjected to the image processing, and ordinary photographic processing is performed. Get color prints. The description of the processing device 61, the color scanner device 62 and the like is omitted, and the image processing unit 50 according to the present invention will be described below.

The image processing section 50 includes an image display section 51.
The pointing device 52, the arithmetic processing unit 53, and the storage unit 54 control the operation of the photo processing apparatus 1 as a whole.

First, the image display section 51 reads image information recorded on a photographic light-sensitive material such as a film as image information by a color scanner device 62 or the like, and the image data read as digital data is displayed on a display device (hatched in the figure). The image is displayed on the display) and is composed of a CRT, a liquid crystal display device, or the like. An interface (not shown) that allows input / output of statistical data regarding a target image, and additional instructions and information regarding arithmetic processing and image processing described later is provided in association with the image display unit 51. Good.

Further, the pointing device 52 has a function of designating an arbitrary point on the image display unit 51, and a rectangle calculated by the arithmetic processing unit based on the two points when two points are designated as described later. It has a function of acting on the arithmetic processing unit and the image display unit so as to generate an area and a processing area standard pattern for performing pattern matching, and is configured by a so-called mouse or trackball.

Further, the arithmetic processing unit 53 is a process for specifying a region for performing image processing according to the present embodiment on the image data (hereinafter referred to as "image processing region specifying process").
And calculation processing for calculating the rectangular area and the processing area standard pattern, color correction and density correction,
Performs image processing for processing unique to photographic images, which will be described later, and arithmetic processing for performing various image processing such as enlargement / reduction, rotation, inversion, projection, shading, composition, and division that are generally performed on images. And is composed of various logic circuits, arithmetic circuits, registers, and the like.

Further, the storage section 54 calculates not only basic data based on predetermined data or statistical data relating to the target image according to the present embodiment but also the image data, data obtained by processing the basic data, and the like. It stores data in the process of processing and is configured by a RAM, a ROM, and the like. As described above, since the storage unit 54 stores the basic data based on the predetermined data and the statistical data related to the target image, it does not require a particularly complicated calculation or the like and is based on the predetermined data and the statistical data. Therefore, the position of the processing area can be estimated. Since the storage unit 54 does not require any special device or processing, the storage unit can be realized at low cost.

Next, the image processing unit 5 according to the present embodiment.
The basic concept of the image processing area specifying process performed by 0 will be described. In the present embodiment, a subject is an image of a person, and an image of a pupil or an iris among the images is subjected to image processing as an object of image processing. Therefore, first, the eye area is specified as the "target area", the eye image is enlarged and displayed, and then the pupil or iris, which is the area to be image-processed, is designated as the "processing area", one point for each white-eye area. A rectangular area is displayed by designating points and the pupil or iris is specified. Then, image processing is performed on at least a part of the identified pupil region or iris region. The details will be described below. In the following description, both references such as "pupil or iris" may be used.

First, a method of designating an eye region as a target region will be described. In the present embodiment, a target area which is a premise for specifying the image processing area is designated. 2 for that
A point is designated, a line segment defined by the two points and a region having a predetermined positional relationship with the line segment are estimated to specify the target region. Alternatively, when the area spread of the face image can be used, that is, when the designated points can be designated also on the crown and the chin, as will be described later, four points are designated and a line segment determined by the four points and a predetermined positional relationship. Area is estimated, an icon corresponding to the area is displayed, and the target area is specified.

The case of designating four points will be described below. When designating two points, no icon or the like is used, and four points are designated to specify the eye region only from the statistical positional relationship described later. Other than that, the others are the same. In the present embodiment, four points for designating a target area on which image processing area is specified are designated, and a line determined by the four points is utilized by using the positional relationship of each part of the face based on statistical data. A circle corresponding to the eye shape, a tool such as a spectacle-shaped icon, a mask-shaped icon, or the like is displayed at a position where the eye area is statistically estimated based on the minutes, and the eye area is specified. Hereinafter, the statistical data will be described first.

FIG. 2 shows an eye 10 and a nose 3 of a human face.
It is a plane explanatory view which showed roughly the result of having investigated statistically the mutual positional relationship which 1, the mouth 32 grade | etc., Exist in. In general, it is known that the positional relationship of the existence regions, where the left and right eyes 10 are present in the human face, is statistically substantially constant.

That is, as shown in the figure, when the line segment CD is set so as to be located on the line segment passing through the lower end of the left eye and the lower end of the right eye, the positions of both ends of the right eye (the inner and outer corners of the eye) and The positions of both ends of the left eye are four points P ₁ , P ₂ , P that divide the line segment CD into five equal parts.
_It is almost equivalent to the positions of ₃ and P ₄ . That is, h ₁ ≈h ₂ ≈h ₃ ≈h ₄ ≈h ₅ (1). The line segment AC is divided into two approximately equal parts by the line segment BD. Therefore, it is AF ≒ DF ≒ FB ≒ FC ≒ v 1 ≒ v 2 ≒ 0.5V (2). Then, similarly, a, b, c, and d are line segments B similarly as statistically obtained constants (setting values can be changed).
The size L of the eye itself in the D direction (longitudinal direction of the face) is L≈h ₁ ≈aH (a≈0.2) (3), and the eye sees the point F from bh ₁ (b≈0. A positional relationship (positional relationship 1) of being located in the area from the point 5) to the point ch ₁ (c≈0.5) is obtained.

Then, the size W of the eye itself in the direction of the line segment AC (the lateral direction of the face) is y ₁ ≈y ₂ ≈0 for the lower end P _{5 of} the eyebrows, the lower end P _{6 of} the nose and the chin C. 5Y Since (4), W ≈ dy ₁ (d ≈ 0.3) (5), and the eye is said to be located in the region from the position touching the line segment BD to the line segment to dy ₁ A positional relationship is obtained (positional relationship 2).

Therefore, the length of the line segment AC and the length of the line segment BD are respectively obtained on the displayed image, and based on the length data, the above-mentioned formula ( According to 1) to (5) and the positional relationships 1 to 2, it is possible to calculate and calculate an area where the presence of the eye is estimated, and display the image.

Using the statistical data on the positional relationship of each part of the face described above, the following four points are designated and the line segment defined by the four points is determined.

That is, first, on a straight line passing vertically through substantially the center of the face, a point A on the top of the head, a point B on the chin, and a straight line connecting the lower ends of both eyes and passing the face left and right, A point C on the right end of the face and a point D on the left end of the face are designated. The point C is located between the right eye and the edge of the face,
It is assumed that the point D is located between the left eye and the edge of the face.
If the above four points are specified, the line segment A connecting the points A and B
A line segment CD connecting the points B, C and D is determined, and further, an intersection point F between the line segment AB and the line segment CD is determined. At this time, 4
The points are set so that the quadrilateral shape can be designated, and the diagonal lines are in a positional relationship in which they are substantially orthogonal to each other. Since four points are designated at such positions, the designation operation is easy without requiring any skill.

Next, using the line segments, the eye region is estimated based on the positional relationship of each part of the face based on the statistical data with reference to the line segments defined by the four points (hereinafter referred to as the eye region). , "Estimated region") is displayed to identify the target region.

That is, it is estimated that the left eye is above the line segment DF and the right eye is above the line segment CF.
The position is determined in association with the line segments DF and CF according to the positional relationship of each part shown in FIG. Based on the positional relationship of each portion, the estimated region of the eye is calculated and calculated according to the positions of the four points and the formulas (1) to (5). At this time, the size of the estimated region is set so as to include the variation in the designated positions when the four points are designated and the estimated region may include the true region. After the eye area is specified as described above, the pupil area or the iris area is specified from the eye areas. FIG. 3 is an explanatory diagram schematically showing a method of identifying a pupil or an iris.

That is, of the areas of the eye 10 including the eyelashes 14, one point A and B are designated in both areas of the white eye area 11. A rectangular area 16 including these two points A and B
Then, a standard pattern described later is set in the center of the rectangular area 16, and the standard pattern and the displayed image are subjected to pattern matching to specify the processing area and display the image processing.

At this time, the rectangular area 16 is set so that the long side of the rectangular area 16 is parallel to the horizontal direction of the eye and the two points are located on the short sides. And
The iris and the pupil are concentrically located in the center of the line segment defined by the two points, and an arc corresponding to the length of the short side is obtained, and is stored for each pixel in the iris 12 area or the pupil 13 area. Iris 12 and pupil 13 stored in section 54
Processing area standard pattern 17 (hereinafter simply referred to as “standard pattern”) in which color information and the like are set based on predetermined data regarding
Is preliminarily set so as to be located in the central portion of the rectangular area 16, and pattern matching is performed with the displayed image to specify the processing area.

As a concrete method of this pattern matching, for example, the standard pattern 17 is formed as a template, and the iris 12 is formed by performing template matching on the displayed image according to a known method.
The area or the pupil 13 area is specified. The pattern matching method is not limited to template matching.

In either case, according to the present embodiment, the iris 1
Since the two areas and the pupil area 13 are narrowed down considerably, the number of standard patterns 17 to be held can be small, and the estimation area can be specified as a processing area by a simple calculation process.

As described above, the iris 12 region or the pupil 13
Since the area is specified, the processing area can be estimated and specified by performing the arithmetic processing only by using a well-known arithmetic circuit or the like. Therefore, the arithmetic processing unit and the like can be realized at low cost. Moreover,
Since the estimated positions of the target area and the processing area are obtained by performing arithmetic processing based on the line segment, the processing area can be estimated at high speed with simple arithmetic processing, and the target area and the processing area can be estimated in a short time without using a special complicated device. The processing area can be specified. Further, since the target area is estimated based on the statistical data, the target area can be accurately estimated with statistical certainty.
Furthermore, since the processing area is considerably narrowed down when performing pattern matching, pattern matching can be performed accurately in a short time.

FIG. 4 is a flow chart showing, in the order of steps, an actual procedure for performing image processing according to the basic concept of the image processing area specifying processing described above. First, a target face image is captured as image data by a scanner device or the like from an image recording unit such as a film (step 3
2), an image is displayed on the image display unit 51, which is an image display medium, as a first display step (step 33). If the target image is an eye, the processing area is a pupil or an iris, and the white-eye area that sandwiches the processing area can be displayed sufficiently large, and two points that sandwich the processing area can be easily specified in this white-eye area. , A target area designation step 34 and a second display step 3 which will be described later.
Step 5 is omitted, and the white-eye region is used as it is as the processing region.
6 is executed.

In step 33, if the face is facing diagonally or is lying down, the two points that are the preconditions for designating the target area by rotating the face to a position that is almost front in the image display, or The image may be corrected and displayed on the image display unit 51 in such a direction that it becomes easy to specify four points.

Next, in order to specify the eye region, the above-mentioned two points or four points are designated from the displayed face image by the pointing device 52 such as a mouse (step 34). At this time, when designating four points, the four points are designated in order around the predetermined direction from the crown. However, 4
The order itself is not limited as long as it is appropriately defined which of the points is the crown or the chin. The specified position needs to be inside the face region, but if it is inside, the position may be slightly deviated. For example, in the case of the crown, the position on the contour line of the face is preferable, but the position may be slightly deviated, that is, the hair portion may be used. By specifying 2 points or 4 points in this way, as described above, the positional relationship and the line segment obtained from the statistical data based on the positional relationship of each part of the face are calculated and determined, and this line segment is determined. The position of the eye is statistically estimated based on the reference and displayed as an estimated region in an image.

At this time, when four points are designated, in order to make the relationship between the estimated region of the eye and the contour of the face easy to see on the image display unit, at the same time as displaying the face image, an icon indicating the estimated region is displayed. Can be displayed. This icon
As shown in (a) to (c) of FIG.
(a)) or an icon in which the estimated area is in the shape of glasses (see FIG. 5).
(b)), or a mask-shaped icon ((c) in FIG. 5) having an opening in the estimated region can be set. The shape of the estimated region may be a circle or an ellipse, or may be a shape that approximates the shape of the eye and is close to the contour of the eye.

If the shape of the estimated area is a circle or an ellipse, the icon can be easily set and the storage capacity of the storage unit is not overwhelmed. Further, if the shape of the estimated region is a spectacle shape, the facial expression of the eyes can be visually imagined when performing image processing. Furthermore, if the estimated region is a mask-shaped icon having an opening, it is easy to see because it is not affected by the skin color part.

At this time, the position and size of the eyes and the positional relationship between the eyes may be deviated from the estimated region due to variations in the age and sex of the person to be photographed, so that the pointing device 52 and the like are well known. By such an operation, the position of the eyeglasses or the position of the opening of the mask-like icon can be displaced, or the size can be enlarged or reduced to be closer to the position and size of the display image. Therefore, it is possible to adjust the size and position of the estimated region by displaying the estimated region so that the positional relationship is easy to understand with respect to the display image.

After that, the eye portion is extracted and enlarged to be displayed as the second display (step 35). Next, as two points sandwiching the processing region, one point is designated for each of the two white eye portions. Is designated (step 36a), and the rectangular area 16 and the standard pattern 17 are displayed based on these two points (step 3).
6b). As described above, the template is previously formed as the standard pattern 17 set in accordance with the iris 12 or the pupil 13 in the central portion of the rectangular area 16, and the template can be displayed in accordance with the rectangular area 16. With respect to the image, template matching is performed on the standard pattern 17 and the image in the central portion of the rectangular area 16 to perform pattern matching (step 36c) to identify the iris 12 or the pupil 13 (step 36d).

After the processing area is specified as described above, at least a part of the processing area, for example, the pupil 12 or the iris 13 is extracted and the color is changed, or the part is enlarged or enlarged. Image processing can be performed by reducing or adding a desired partial image (step 3).
7).

That is, it is possible to judge that the circular area inside the eye is the pupil and arbitrarily change the color of the pupil to remove the red eye, or to wear a color contact. In addition, the contrast of the pupil region and its adjacent region can be enhanced by black and white. In addition, a white dot can be added to the pupil to create a so-called catch light effect.

As described above, a simple algorithm is used to specify a smaller number of designated points than in the conventional method.
It is possible to specify the area to be image-processed in a relatively short time by specifying it at a position that is easy to specify. Moreover, since the positions of the two designated points sandwiching the processing region are not particularly complicated positions, it is possible to easily specify the region to be image-processed without requiring skill even in the case of each eye or only one eye. You can

In the above, the case where the processing area is the pupil or iris of the person as the subject has been described in the present embodiment, but the present invention is not limited to this configuration, and each portion of a certain object is processed. If the positional relationship is quantitatively obtained and the area of each part is estimated by the positional relationship, and the processing area can be sandwiched by two points, the target image is not only a person, but also a living thing such as animals and plants. The area can be specified even for inanimate objects such as buildings and articles.

[0061]

According to the photographic processing method of the present invention,
In the specifying step of specifying the processing area to be image-processed in the displayed image, two points sandwiching the processing area are designated and
A rectangular area defined based on points and a processing area standard pattern located in the center of the rectangular area are displayed as an image, and the processing area standard pattern and the displayed image are subjected to pattern matching to define the processing area. With the configuration of specifying, it is possible to perform pattern matching of the processing area only by specifying two points sandwiching the processing area at positions that are easy to specify in the vicinity of the processing area. Therefore, the specified points can be specified without skill. It is possible to process images with high accuracy and in short time.

According to the photographic processing method of the present invention, the subject is a human eye, the processing area is at least one of a pupil area and an iris area, and the two points can be located in the white eye area. Since it is possible to perform processing for each eye and it is easy to specify two points that sandwich the processing area, the specified points can be specified without requiring skill.

According to the photographic processing method of the present invention, the processing area standard pattern can be used as a template, and the processing area can be specified by performing template matching. Therefore, a small number of templates can be held. Since the pupil or iris region can be specified accurately in a short time, the image processing of the pupil or iris can be performed without requiring skill even with each eye or only with one eye.

According to the photographic processing method of the present invention, when an image of a subject is displayed as an image to specify a target area, and the target area is extracted and enlarged to display an image, the processing area is specified. Two points sandwiching the processing area are designated, a rectangular area defined by the two points and a processing area standard pattern positioned in the center of the rectangular area are displayed, and the processing area standard pattern and the displayed image are displayed. With the configuration in which the processing area is specified by performing the pattern matching, the pattern matching of the processing area can be performed only by specifying two points sandwiching the processing area at positions that are easy to specify in the vicinity of the processing area. Even if the image to be processed is small, a specified point can be specified without requiring skill, and image processing can be performed in a short time and with high accuracy, and photo processing can be performed.

According to the photographic processing method of the present invention, the subject is a human face, the target area is the human eye, and the processing area is at least one of a pupil area and an iris area. Two points can be located in the white-eye area, processing can be performed for each eye, and since two points sandwiching the processing area can be easily specified, the specified points can be specified without requiring skill.

According to the photographic processing method of the present invention, since the processing area standard pattern is used as a template and the processing area can be specified by performing template matching, a small number of templates can be held for a short time. The pupil or iris region can be accurately specified, and the pupil or iris region can be accurately specified in a short time by holding a small number of templates. Iris image processing is possible.

According to the photographic processing method of the present invention, in the target area designating step, the target area is designated by designating two predetermined points, so it is only necessary to designate two predetermined easy-to-designate points. The target area can be specified easily with and there is no need to specify multiple points in the vicinity of the target area. Therefore, image processing of the pupil or iris can be performed without requiring skill.

According to the photographic processing method of the present invention, in the target area designating step, the target area is designated by designating the predetermined four points, so that only the predetermined four points which are easy to designate are designated. The target area can be easily specified with, there is no need to specify a plurality of points in the vicinity of the target area, and an icon for easily displaying the target area can be displayed. Therefore, image processing of the pupil or iris can be performed without requiring skill.

The photographic processing apparatus according to the present invention comprises an image display section, a pointing device, an arithmetic processing section, and a storage section 5.
And a processing area for image processing is displayed, and when two points sandwiching the processing area are designated by the pointing device, a rectangular area and a processing area standard pattern defined by the two points are displayed. Then, pattern matching is performed to specify the processing area and image processing can be performed. Therefore, it is possible to realize a photographic processing device which can accurately specify a processing area in a short time and at low cost only by designating two points sandwiching the processing area by a normal image processing means.

[Brief description of drawings]

FIG. 1 is a block diagram schematically showing the configuration of a photographic processing apparatus according to an embodiment of the present invention.

FIG. 2 is an explanatory view conceptually showing an image processing area specifying method according to an embodiment of the present invention.

FIG. 3 is an explanatory diagram conceptually showing a template according to an embodiment of the present invention.

FIG. 4 is a flowchart showing an image processing area specifying method according to an embodiment of the present invention in process order.

5A and 5B are explanatory views conceptually showing an image processing area specifying method according to an embodiment of the present invention, in which FIG. 5A is a circular icon, FIG. 5B is a spectacle-shaped icon, and FIG. Explanatory drawing which shows the mask-shaped icon which has.

FIG. 6 is an explanatory view conceptually showing a conventional method of designating the vicinity of an eye.

[Explanation of symbols]

10 ... Eye, 16 ... Rectangular area, 17 ... Template, 50
… Image processing unit

Continuation of the front page (56) Reference JP-A-10-91761 (JP, A) JP-A-9-261580 (JP, A) JP-A-10-233929 (JP, A) JP-A-9-81732 (JP , A) (58) Fields investigated (Int.Cl. ⁷ , DB name) H04N 1/387 H04N 1/40-1/409 G06T 7/00

Claims (9)

(57) [Claims] 1. A display step (33) for displaying an image of a subject.
And a specifying step (36) of specifying a processing area to be image-processed in the displayed image, and an image processing step of image-processing the processing area to form image data for forming a photograph.
(37) In the photograph processing method including
In (36), two points sandwiching the processing area are designated, and a rectangular area determined based on the two points and a processing area standard pattern located in the central portion of the rectangular area are displayed as images.
A photographic processing method characterized in that the processing area is specified by performing pattern matching between the processing area standard pattern and a displayed image. 2. The subject is a human eye (10), the processing region is at least one of a pupil region (13) and an iris region (12), and the two points are white eye regions (11). The photographic processing method according to claim 1, which is located at. 3. The photograph processing method according to claim 1, wherein the processing area standard pattern is used as a template (17), and template processing is performed to identify the processing area. 4. A first display step (33) for displaying an image of a subject, and a target area specifying step (34) for specifying a target area including a processing area for image processing in the displayed image.
And extracting the target area and enlarging and displaying the image.
A display step (35), an identification step (36) of identifying the processing area from the target area, and an image processing step (37) of image-processing the processing area to form image data for forming a photograph. ) And a processing area located at the center of the rectangular area defined by the two points that sandwich the processing area in the specifying step (36) A photographic processing method characterized by displaying a standard pattern and performing pattern matching between the standard pattern of the processing area and the displayed image to identify the processing area. 5. The subject is a human face, the target region is the human eye (10), and the processing region is at least one of a pupil region (13) and an iris region (12). The photographic processing method according to claim 4, wherein the two points are located in the white-eye area (11). 6. The photographic processing method according to claim 5, wherein the processing area standard pattern is a template (17), and the processing area is specified by performing template matching. 7. In the target area specifying step (34),
5. The photographic processing method according to claim 4, wherein two predetermined points are designated, and the target area is designated based on a positional relationship between the line segment defined by the two points and the target area. 8. In the target area specifying step (34),
The photographic processing method according to claim 4, wherein the predetermined four points are designated, and the target area is designated based on a positional relationship between the line segment defined by the four points and the target area. 9. A photographic processing device provided with an image processing unit (50) capable of performing image processing on image data of a subject,
The image processing unit (50) displays an image of the subject, an image display unit (51), a pointing device (52) capable of designating an arbitrary point and area on the image display unit, and for image processing. Arithmetic processing unit (53) for arithmetically processing image data
And a storage unit that stores predetermined data regarding the image
(54) A rectangular area defined by the two points when a processing area for image processing of the image is displayed and two points sandwiching the processing area are designated by the pointing device (52) and the rectangular area A photographic processing apparatus capable of displaying a processing area standard pattern positioned at the center of the area, performing pattern matching to identify the processing area, and performing image processing.