JP2001111806A - Image generator and device and method for recognizing top and bottom of image - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image generator that automatically recognizes an image of an object such as a person included in an image photographed by a digital camera or the like, extracts and prints out it and to provide a device and a method that recognize a top and a bottom of an image applicable to the image generator as above. SOLUTION: An area of nearly channel shape is set along 3 sides in 4 sides of an outer circumference of a designated image as a decision area to decide a top and a bottom of the image, a maximum value and a minimum value of pixel value of pixels in the decision area are obtained, a top and bottom direction of the image is decided on the basis of the difference between the maximum value and the minimum value, the image is rotate so that the top and bottom direction of the image is a prescribed reference direction on the basis of the result of decision (S54, S55, S62, S63), object image parts are identified from the rotated image (S65-S58, S65), a prescribed image part including the part of the identified target image is extracted (S61), and the extracted image part is printed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image forming apparatus for extracting and printing a target image portion such as a human image from image data recorded by a digital camera or the like, and further applicable to such an image forming apparatus. The present invention relates to an apparatus and a method for recognizing upper and lower sides of an image.

[0002]

2. Description of the Related Art In recent years, digital cameras for recording captured images as electronic data on a recording medium have rapidly become widespread. In the future, with further technological innovation, digital cameras with lower cost and higher performance will be provided. Is expected.

[0003] A user can freely process, edit, and print data photographed by a digital camera on a personal computer, and the range of use is wide.
For example, it is conceivable to create an identification photograph for obtaining a passport or a license from image data obtained by photographing a person.

[0004]

Generally, ID photos are:
It is necessary to satisfy various conditions, such as the size of the image and the ratio of the person portion (particularly, the face portion) in the entire image. Therefore, in order to create an ID photo from image data, the user manually changes the size (enlargement / reduction) or meets the conditions while viewing the image data displayed on the display device. It is necessary to set a cutout area. At this time, in the image data, the person image included in the image data may be displayed horizontally or downward, and the image needs to be rotated as necessary. Since these operations are very complicated operations for the user, a device for automatically recognizing, extracting, and printing a person image included in the image is desired.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and an object of the present invention is to convert an image photographed by a digital camera or the like and recorded as electronic information into a person or the like included in the image. An object of the present invention is to provide an image creating apparatus for automatically recognizing an image of a target, extracting and printing the image, and an apparatus and method for recognizing the upper and lower sides of an image applicable to the image creating apparatus.

[0006]

According to the present invention, there is provided an image forming apparatus comprising: an input unit for inputting an image including a target image; and a four-sided image forming an outer periphery of the image as a determination area for determining the upper and lower sides of the input image. An area setting means for setting a substantially U-shaped area along three sides of the three sides, obtaining a maximum value and a minimum value of a pixel value for a pixel in the determination area, and based on a difference between the maximum value and the minimum value. Determining means for determining the vertical direction of the image, rotating means for rotating the image based on the determination result so that the vertical direction of the image is a predetermined reference direction, and identifying means for identifying a target image portion from the rotated image Extracting means for extracting a predetermined image portion including a part of the identified target image; and printing means for printing the extracted image portion.

In the image forming apparatus, when the vertical direction of the image cannot be determined by the vertical direction determining means, the rotating means, the identifying means, the extracting means, and the printing means may not perform the processing.

According to the image vertical recognition apparatus of the present invention, a substantially U-shaped area is set as a determination area for determining whether the image is up or down along three of the four sides constituting the outer periphery of the image. Setting means; and determining means for determining the maximum value and the minimum value of the pixel values for the pixels in the determination area, and determining the upper and lower sides of the image based on the difference between the maximum value and the minimum value of the pixel values.

In the image vertical recognition device, the U-shaped region is formed by a first region provided along one of the above four sides and a second region provided along a side contacting one end of the one side. A third region provided along a region and a side contacting the other end of one side thereof
Area. At this time, preferably, the lengths of the second region and the third region are shorter than the lengths of the sides along which the respective regions are provided.

Further, in the image vertical recognition apparatus, at least one of the first area, the second area, and the third area is provided.
One may consist of discontinuous regions.

[0011] In the image vertical recognition method according to the present invention,
As a determination area for determining the upper and lower sides of the image, a substantially U-shaped area is set along three of the four sides constituting the outer periphery of the image, and the maximum and minimum pixel values of the pixels in the determination area are set. , And based on the difference between the maximum and minimum pixel values,
The top and bottom of the image are determined.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an image forming apparatus according to the present invention will be described below with reference to the accompanying drawings. In addition,
In the drawings, the same reference symbols indicate the same or equivalent ones.

The image forming apparatus of the present embodiment prints out an image recorded on a recording medium in various formats according to a user's request. The image creating apparatus has, as one of various functions, a function of creating a certification photograph from an image recorded on a recording medium and printing out the photograph.

(External Appearance of Image Creation Apparatus) FIG. 1 shows an external appearance of an image creation apparatus 1 according to an embodiment of the present invention. This image creation device 1 is a coin vender type device in which a user inputs a fee, checks an image by himself, and obtains a print. The fee is input to the input port 8. A media loading unit 2 is provided at the top of the apparatus, and the user can select one of four types of image recording media (smart media (registered trademark), compact flash (registered trademark), optical disk, and floppy (registered trademark) disk). Can be loaded. The media loading section 2 is provided with a dustproof cover 3. In addition,
A dustproof shutter may be provided at each loading port instead of the cover. A panel 7 is arranged beside the media loading section 2 to display the types of media that can be loaded. The monitor 4 is provided to display images and operation instructions, and has a touch panel formed on the surface. The keyboard 5 is used for giving various instructions. A silver halide printer 16 for printing a normal photograph and a sublimation printer 18 for printing a seal print are mounted inside the apparatus (see FIG. 3). The print is discharged from the discharge port 6. The user enters a fee into the slot 8 and operates the apparatus with the monitor 4 and the keyboard 5.

FIG. 2 shows the recording medium loading section 2 and the display panel 7 in more detail. First, the display panel 7 displays the shapes of four types of image recording media that can be loaded in the apparatus. Loadable media are smart media (SM), compact flash (CF), optical disc (C
D) and a floppy disk (FD). The number to the left of each symbol corresponds to the number of the loading port. The user checks the type of the recording medium with the display on the panel 7 and selects the corresponding medium. As a means for selection, a touch switch may be provided on the panel 7 or a number may be input on the keyboard 5. Further, the touch panel on the monitor 4 may be selectable. When the user selects a medium, a display 21 provided beside the loading port corresponding to the selected medium is turned on. In the example shown in FIG. 2, the display 21 corresponding to the smart media is turned on. The user loads the recording medium into the loading slot where the display 21 is lit. By doing so, the user can identify his or her own medium, and it is not necessary to load the medium into a wrong loading slot.

FIG. 3 is a block diagram of the inside of the image forming apparatus. The entire apparatus is controlled by the overall control unit 20. The drive 10 reads an image recorded on the smart media. Drive 11
Reads an image recorded on a compact flash. The drive 12 reads an image recorded on an optical disk. The drive 13 reads an image recorded on a floppy disk. The image signal processing unit 14
The digital image signal from each medium received from each of the drives 10 to 13 is processed for monitoring and printing,
The memory unit 15 records the signal-processed image. The contents recorded in the memory 15 are reproduced by the monitor 4 and the printers 16 and 18.

The silver halide printer 16 performs digital image exposure on silver halide paper, and then performs development, fixing, and stabilization processing to create a digital silver halide print. In this apparatus, a PLZT optical shutter array (not shown) is used as a silver halide exposure control unit. First interface
An (IF) 17 is an interface with the silver halide printer 16. Through the first interface 17, the memory 1
5 is sent to the silver halide printer 16, a print start command is sent from the overall control unit 20 to the silver halide printer 16, and the overall control unit 20 sends a print start instruction to the silver halide printer 16. Print status data is sent.

Another printer 18 is a sublimation printer for producing a seal print. This printer 18
As long as the printer is suitable for seal printing, another type of printer (for example, an inkjet type or a hot-melt type printer) may be used. The second interface (IF) 19
This is an interface with the sealing printer 18 and has almost the same function as the first interface 17.

(Overall Control Sequence) Next, a control sequence by the overall control unit 20 in the image forming apparatus will be described.

FIG. 4 is a flowchart from the service menu display to the index display. When the present apparatus is started, first, a service menu (see FIG. 6A) usable in the present apparatus is displayed on the monitor 4 (step S11; hereinafter, "step" is omitted). The user selects a desired service on the touch panel. In the present embodiment, the following six services (print modes) can be used. (1) All prints (mode selected when printing all images recorded on media) (2) Individually designated print (mode for individually designating and printing images recorded on media) (3) Index print (Mode for performing index printing) (4) Sticker printing (mode for performing sticker printing) (5) ID photo printing (mode for individually designating images recorded on media and printing them as ID photos) (6) Data Save (mode to save data to media CD-R, FD, MO)

No matter which service is selected from the service menu, the processing up to the index display is common. First, in order to capture an image, the user selects a recording medium for inputting an image (S12). The media may be displayed on the monitor 4 so that the media can be selected.
Alternatively, the selection may be made with the keyboard 5. When the user makes a selection, the display 21 of the loading port of the selected medium is lit on the display panel 7 (S13).

Next, it waits for the user to load a medium (S14). When media is loaded by the user,
A subroutine for data input / decompression is called to decompress and read image data from the medium (S15). The original image data is stored in the memory. This is because the original image data is stored until the operation is completed, so that high-speed access is always possible. Next, the process waits for the user to remove the media (S16), and then inputs the selection of the service content by the user (S17), and performs a predetermined process according to the selected service content (S18 to S2).
3).

That is, if the selected service is "print all", a predetermined process for printing all the images recorded on the medium is performed (S18). If the selected service is "individual designation print", a predetermined process for individually designating and printing images recorded on the medium is performed (S19). If the selected service is "index print", a process for performing index print is performed (S20). If the selected service is "seal print", a process for performing a seal print is performed (S21). If the selected service is "ID photo print",
A predetermined process of individually designating the images recorded on the medium and printing them as a photograph for certification is performed (S22). If the selected service is "data storage", a process for storing data in the media CD-R, FD, and MO is performed (S23).

Here, the processing for each of the services "print all", "designated print", "index print", "sticker print", and "data storage" may be any known processing. In the following, in particular, control for realizing the “certification photo print” process will be described in detail.

(Proof Photo Print Processing) FIG. 5 is a flow chart showing the control when the verification photo print is selected. In this process, a process of creating an image for an ID photograph from the image data recorded on the medium loaded by the user and printing the image is performed.

First, a selection screen for the user to select the type of ID photo is displayed (S31). For example, a selection menu as shown in FIG. 6B is displayed, and the user selects a type of photograph, such as a general type, a passport type, or a license type, according to the purpose. Next, information of the type of the photograph designated by the user and the number of prints are fetched (S3).
2). The number of prints is also input on the monitor 4 by the user as in the case of the type of photo. The amount under the current setting is calculated and displayed on the monitor 4 (S33). The user performs setting while checking the display, and when the setting is completed, presses an OK button (not shown) displayed on the monitor on the monitor 4. It is determined that the user has pressed the OK button (S34), and the process proceeds to the next step.

Next, the service contents and the final fee are displayed (S35, S36), and the user waits for a deposit, and when there is a deposit, calculation of change, payment processing, etc. are performed (S3).
7). Thereafter, the subroutine of the ID photo image creation process is called, and the ID photo image is generated from the image stored in the medium loaded in the loading slot by the user (S3).
8). Subsequently, a process of printing the created ID photo image on predetermined paper is performed (S39), and the process returns. At this time, the created ID photo image may be stored in a medium.

(Proof Photo Image Creation Process) FIG. 7 is a flowchart showing the identification photo image creation process (S38). This processing is divided into two parts: processing to determine the top and bottom of the image to position the image in the correct direction, processing to detect the outline part of the human image in the image, and processing for the ID photo centering on the detected outline part. And editing the image. Hereinafter, these will be described specifically.

First, image file information designated by the user is read, an image corresponding to the designation is read from the memory, and displayed on the monitor 4 (S51). Here, the user specifies an image to be used for the ID photograph on the monitor 4.
FIG. 11A shows a display example of image data in this case. Next, a frame (cutting frame) indicating an area to be cut as an ID photo is displayed on the monitor 4 (S52). FIG.
2 shows an example of the cutout frame (in this figure, B indicates the outline of the face). The cutout frame A has information indicating the center of the face, the position of the head, and the position of the chin. Cutting frame A
Are initially displayed at the center of the image as shown in FIG. Thereafter, the vertical direction of the designated image is determined, and the image is rotated so that the vertical direction is correct (S53 to S55, S62, and S63).

Specifically, first, the size of the designated image, that is, the vertical and horizontal sizes, is read from the tag information of the designated image (S53). Next, a determination area for determining whether the image is up or down is set (S54). Here, a determination area for up / down determination will be described.

The determination area is set to a substantially U-shaped area having a predetermined width provided along three of the four sides constituting the outer periphery of the image. For example, the substantially U-shaped region includes at least a part of a region along one of four sides forming an outer periphery of an image and at least a part of a region along each of two sides contacting both ends of the one side. Is set to an area of 1 dot width including. In the present embodiment, the U-shaped area painted black in FIG. 8A is set as the determination area. That is,
In FIG. 8A, the U-shaped determination region is a region that is parallel to the upper side and a region that is parallel to the left side of the upper side in the one-dot width region on the outermost periphery of the image. And a part of a region parallel to a side that is in contact with the right side of the upper side. More specifically, the region parallel to the two left and right sides extends from a position in contact with the upper side, and has a half length of the vertical size of the image. Therefore, if the image is an image having a size of vertical y × horizontal x dots, the determination region has one side of a (= x) dot in the horizontal direction and b in the vertical direction in contact with the side. It becomes a U-shaped area composed of (= y / 2) dots.

The reason why the determination area is set as a U-shaped area along the outer periphery of the image as described above is as follows. That is, in the case of a photograph in which a person such as an identification photograph is taken, the upper side and the side (left and right) sides of the person are usually backgrounds, so that the colors of these areas are considered to be substantially uniform.
Therefore, the upper end and the side (left and right) of the person in the image
The colors of the dots at the side edges are considered to be substantially uniform. On the other hand, since the color of the dot at the lower end of the image is the color of the clothes or the like worn by the person, it is generally considered that the color of the dot at the lower end, which is the background color, will be largely different from the color of the upper end, left and right end . Therefore, in the U-shaped region provided in the peripheral portion of the image, if the colors of the pixels in the region are substantially the same, the side located in the middle of the three sides constituting the U-shaped region Indicates the upward direction of the image, so that the upper and lower sides of the image can be determined.

It should be noted that the U-shaped region of the present embodiment corresponds to FIG.
(A) is set as a continuous area as shown in FIG. 8 (b) and (b) in FIG. c), (d), (e)
As shown in FIG. Further, the width of the U-shaped region is not limited to one dot, and may be a sufficiently small dot width that does not include the central target image. Similarly, if the U-shaped region is provided at the peripheral portion of the image, it may not be provided at the outermost periphery.

Returning to FIG. 7, after the U-shaped determination area is set, the color of each pixel in the set U-shaped area for the image, that is, the R, G, B colors for each pixel (Hereinafter referred to as “pixel value”), and the maximum value and the minimum value of the pixel value of each pixel in the area are obtained (S5).
4). For each color, it is determined whether or not the difference between the maximum and minimum pixel values is within a predetermined range (S55). When the difference between the maximum value and the minimum value of the pixel value of each color is within a predetermined range, it is determined that the color of each pixel in the determination area is substantially uniform, and it is determined that those areas are part of the background. Because you can.

As a result of the above determination, if the difference between the maximum value and the minimum value of the pixel values in the U-shaped area is not within the predetermined range for all the colors of R, G, and B ("NO" in S55), the display is now performed. It is determined that the vertical direction of the image being performed is not correct, and the image is rotated clockwise by 90 ° to continue the inspection (S62). For example, in FIG. 11, the display image is rotated from (b) to (c). Thereafter, it is determined whether the image has rotated four times (ie, one rotation) (S6).
3) If it has not rotated four times, the process returns to step S54, and the above processing is repeated. When the image has been rotated four times, the image cannot be determined in the vertical direction, that is, it is determined that the image is invalid (NG) (S64), and the process ends.

On the other hand, as a result of judging the difference between the pixel values, if the difference between the maximum value and the minimum value of the pixel values in the U-shaped area is within a predetermined range ("YES" in S55), it is now displayed. It is determined that the vertical direction of the image is correct, that is, the orientation of the human image in the image is correct, and then the human image portion in the image is specified.

Here, a contour database used for specifying a human image portion in an image will be described. The contour database stores the shape of the face contour, face size, face position,
A plurality of pieces of contour data including information on the position of the tip of the head, information on the position of the chin, and information on the center position of the face are registered. This contour database is stored in a predetermined data storage unit (not shown) in the overall control unit 20. FIG. 10 shows a display example of the contour data. Each contour data has a region C corresponding to the region of the entire image and a region B ′ corresponding to the contour of the face (hereinafter referred to as “contour region”). The outline area B ′ has an area 1 and an area 2. Region 1 is the upper 3 in contour region B '
The area 2 is a lower area 2 in the outline area B ′.
This is a one-half area. Further, there is a region 3 outside the contour region B '. The area 3 is an upper area when the extension line L of the center line that separates the upper and lower parts in the outline area B ′ separates an area outside the outline area B ′. Region 1, region 2, and region 3 may be set to other ranges. As shown in each of (a) to (d) of FIG.
The size, shape, and positional relationship within the region C differ for each contour data. Although not shown, there are a plurality of types of the shape of the region C.

Returning to FIG. 7, in order to specify a human image portion in the image, first, one outline data for specifying an image area of a human in the image is read from the outline database (S56). Next, the entire area of the display image is made to correspond to the area C of the outline data, and the standard deviation of the pixel value is obtained in each of the areas 1, 2, and 3 specified by the outline database ( S57).
It is determined whether or not those standard deviations are within respective predetermined ranges (S58). When the standard deviations for the area 1, the area 2 and the area 3 are within the respective predetermined ranges, it is determined that the contour area B ′ specified by the contour database is the face area, that is, the valid “OK” It is determined that there is (S59). Thereafter, the process proceeds to step S60.

On the other hand, if at least one of the standard deviations with respect to the area 1, the area 2 and the area 3 is not within the predetermined range in step S58, it is determined whether or not the inspection has been performed using all the outline data in the outline database (step S58). S6
5) If all the contour data have been inspected, it is invalid (NG) because the human image portion cannot be specified (S6).
6), end the process. If there is any contour data that has not been inspected yet, the next contour data is read from the contour database and inspected in the same manner as described above using this (S56).
To S58).

In step 60, the frame is moved and enlarged / reduced so as to obtain the ID photo specified by the user. That is, based on the information on the head position, the position of the chin, and the information on the center of the face in the outline database, the outline area B ′ and the area A of the frame A are substantially coincident with each other.
Is moved (see FIG. 11D). Also, the frame and the image are appropriately enlarged or reduced so as to have a size corresponding to the type of the ID photo. Thereafter, the image in the frame of the ID photo is cut out, that is, extracted, and edited into image data for the ID photo (S61).

In the above description, an example in which a human image portion of an image is specified and processed has been described. However, by registering contour data corresponding to the target object in the contour database, other target objects other than the person can be registered. Can be automatically determined and extracted. Also, instead of using the contour database to determine the contour area of the image, the head of the head is determined from the change in color (pixel value) from the top to the bottom of the image, and the tip of the chin is estimated from the contour of the cheek. Is also good.

[0042]

According to the image forming apparatus of the present invention, a portion including a target image can be automatically extracted from an input image and printed. As a result, an ID photograph or the like can be easily created from an image taken by a digital camera or the like. Further, according to the image vertical recognition apparatus and method of the present invention, the vertical direction of an image can be easily determined.

[Brief description of the drawings]

FIG. 1 is a perspective view of an image reproducing apparatus.

FIG. 2 is a plan view of a panel and a media loading unit.

FIG. 3 is a block diagram of an image reproducing apparatus.

FIG. 4 is a flowchart of a main process of the image reproducing apparatus.

FIG. 5 is a flowchart for selecting an ID photo print.

FIG. 6A is a diagram showing a service content display screen,
FIG. 7B shows a screen for selecting the type of ID photo.

FIG. 7 is a flowchart of an identification photograph image creation process.

FIG. 8A illustrates a vertical direction determination area (U-shaped area), and FIGS. 8B to 8E illustrate variations of the U-shaped area.

FIG. 9 is a diagram illustrating a cutout frame for an ID photograph.

FIG. 10 is a diagram illustrating an outline database (a to
d).

FIG. 11 is a view for explaining display of an image file (a), display of a cutout frame (b), rotation of an image file (c), and movement of a frame (d) in the identification photograph creation processing.

[Explanation of symbols]

 Reference Signs List 4 monitor 5, 7 operation unit 10-13 media insertion port 16 silver halide printer 18 seal printer 20 overall control unit

Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat II (Reference) G06F 15/70 350H F-term (Reference) 5B057 CC03 CD04 CE09 DA07 DA08 DB06 DC08 DC22 DC33 5C023 AA02 AA03 AA04 AA05 AA14 AA34 AA38 BA02 CA02 DA04 DA08 EA03 EA05 EA06 5C052 AA11 AB04 DD02 DD10 EE02 EE03 EE08 FA03 FA04 FA05 FA06 FB01 FC01 FC08 FD07 FD08 FD10 FE04 5C076 AA17 AA21 AA22 AA24 BA02 CA12 5L096 AA02 BA18 FA14 FA33 FA67 GA67 GA40

Claims (7)

[Claims] 1. An input means for inputting an image including a target image, and a determination area for determining the upper and lower sides of the input image has a substantially U-shape along three of four sides constituting an outer periphery of the image. Area setting means for setting an area; determining means for determining a maximum value and a minimum value of pixel values for pixels in the determination area, and determining the vertical direction of the image based on a difference between the maximum value and the minimum value; Rotating means for rotating the image based on the result so that the vertical direction of the image is a predetermined reference direction; identifying means for identifying a target image portion from the rotated image; and a predetermined image including a part of the identified target image. An image creating apparatus comprising: an extracting unit that extracts a portion; and a printing unit that prints the extracted image portion. 2. The method according to claim 1, wherein when the vertical direction of the image cannot be determined by the vertical direction determining unit, the rotating unit, the identifying unit, the extracting unit, and the printing unit do not perform processing. The image creation device according to claim 2. 3. A determination area for determining whether an image is up or down,
A determination region setting means for setting a substantially U-shaped region along three of the four sides constituting the outer periphery of the image, and determining a maximum value and a minimum value of pixel values for pixels in the determination region; An image up / down recognizing device, comprising: a determination unit configured to determine up / down of an image based on a difference between a maximum value and a minimum value of a pixel value. 4. The U-shaped area includes a first area provided along one of the four sides, a second area provided along a side that is in contact with one end of the one side, and The image vertical recognition apparatus according to claim 3, comprising a third area provided along a side contacting the other end of one side. 5. The apparatus according to claim 4, wherein the lengths of the second area and the third area are shorter than the length of a side provided along each of the areas. . 6. The apparatus according to claim 3, wherein at least one of the first area, the second area, and the third area is a discontinuous area. 7. A determination area for determining whether an image is up or down,
A substantially U-shaped area is set along three of the four sides constituting the outer periphery of the image, the maximum value and the minimum value of the pixel value are determined for the pixels in the determination area, and the maximum value of the pixel value is determined. And determining whether the image is up or down based on the difference between the image and the minimum value.