JP2010021853A - Image processing device, image forming apparatus, image processing method, and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing device, image forming apparatus, image processing method and image processing program which more simplify work for synthesizing image data. <P>SOLUTION: A face image included in image data is detected by a face detection section 11. On the basis of a result of the detection, an image sorting section 12 sorts the image data from which the face image has been detected, as human-character image data, and sorts image data from which the face image has not been detected, as scenery image data. Thus, on the basis of the result of the detection performed by the face detection section 11, image data are sorted as the human-character image data or the scenery image data, so that work for synthesizing image data can be more simplified. Furthermore, by synthesizing the sorted human-character image data and scenery image data, the human-character image data can be synthesized not with a prepared template but with scenery image data photographed by a user himself/herself, for example, thereby forming synthesized image data rich in originality. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing apparatus that performs processing on image data, an image forming apparatus, an image processing method, and an image processing program.

  As an example of a service related to photographic printing, a service that provides image data read from each frame image of developed photographic film and image data obtained by taking a picture with a digital camera by combining them with a template prepared in advance It has been known. In recent years, a service is also known in which an album is created by selecting image data associated with each of a plurality of templates, binding the image data selected for each template and printing the book.

  Patent Document 1 below discloses an example of a method for combining a captured image with a background image that is an example of the template.

JP 2003-163803 A

  However, in a configuration in which image data to be synthesized is simply selected from a plurality of image data, the larger the number of the plurality of image data, the more complicated the operation of selecting desired image data. There's a problem.

  SUMMARY An advantage of some aspects of the invention is that it provides an image processing apparatus, an image forming apparatus, an image processing method, and an image processing program capable of further simplifying an operation of combining image data. And

  An image processing apparatus according to a first aspect of the present invention includes a face detection unit that detects a face image included in image data, and image data from which the face image is detected based on a detection result by the face detection unit. Image classification means for classifying image data in which a face image is not detected into landscape image data, and image composition means for synthesizing the person image data and the landscape image data classified by the image classification means. It is characterized by having.

  According to such a configuration, based on the detection result by the face detection means, the image data is classified into person image data or landscape image data, and the classified person image data and landscape image data are synthesized, thereby synthesizing. Since image data can be generated, the work of combining image data can be further simplified. In addition, since it is possible to synthesize person image data with, for example, landscape image data taken by itself instead of a template prepared in advance, it is possible to generate synthesized image data rich in originality.

  In the image processing apparatus according to the second aspect of the present invention, the image composition means performs processing on the person image data or the landscape image data so that the size of the landscape image data is larger than that of the person image data. And after the process, a part of the landscape image data is replaced with the person image data.

  According to such a configuration, it is possible to generate combined image data in which person image data is arranged on a part of the background image using the landscape image data as a background image. Therefore, if person image data and landscape image data related to each other, for example, person image data and landscape image data taken in the same region, are selected, more interesting composite image data can be generated.

  An image processing apparatus according to a third aspect of the present invention is configured to perform image correction on at least one of the person image data and the landscape image data classified by the image classification means, thereby performing the person image data and the landscape image. Image correction means for differentiating image characteristics from data is provided, and the image synthesis means synthesizes the person image data and the landscape image data after image correction by the image correction means.

  According to such a configuration, it is possible to generate composite image data in which one of the person image data or the landscape image data is emphasized by differentiating the image characteristics between the person image data and the landscape image data. For example, if composite image data in which person image data is emphasized with respect to landscape image data is generated, the landscape image data can be combined as a background image in a suitable mode.

  In the image processing apparatus according to the fourth aspect of the present invention, time information is associated with each of the person image data and the landscape image data, and the person synthesized by the image composition unit based on the time information. Image selection means for selecting image data and the landscape image data is provided.

  According to such a configuration, based on the time information respectively associated with the person image data and the landscape image data, the person image data and the landscape image data having higher relevance are automatically selected, and their synthesis is performed. Image data can be generated. For example, since human image data and landscape image data associated with relatively close time information are likely to have been shot in the same region, combining them automatically makes for a more elegant combination Image data can be easily generated.

  The image processing apparatus according to the fifth aspect of the present invention is the image processing apparatus according to the present invention, wherein the image selection means only when a time interval based on the time information associated with the person image data and the landscape image data is within a predetermined value. The image data is selected as image data to be combined by the image combining means.

  According to such a configuration, when the time interval based on the time information respectively associated with the person image data and the landscape image data exceeds a predetermined value, the image data is not synthesized. Image data with long time intervals are likely to have been shot in different areas. In such a case, image data with low relevance can be created by not creating composite image data. Combining each other can be prevented. Therefore, it is possible to easily generate only more interesting composite image data.

  An image forming apparatus according to a sixth aspect of the present invention includes a face detection unit that detects a face image included in image data, and image data in which the face image is detected based on a detection result by the face detection unit. And classifying the image data in which the face image is not detected into landscape image data, and combining the person image data and the landscape image data classified by the image classification unit, An image composition unit that generates composite image data using image data as a background image of an album board, and an image forming unit that creates an album print by generating a photo print based on the composite image data To do.

  According to such a configuration, by combining the classified person image data and landscape image data, composite image data using the landscape image data as the background image of the album mount is generated, and a photo print based on the composite image data is generated. An album print can be created by generating. Therefore, by binding the created album print, it is possible to produce an album with high originality.

  According to a seventh aspect of the present invention, there is provided a face detection step for detecting a face image included in image data, and image data from which the face image is detected based on a detection result of the face detection step. An image classification step for classifying image data in which a face image is not detected into landscape image data, and an image synthesis step for synthesizing the person image data and the landscape image data classified in the image classification step. It is characterized by having.

  According to such a configuration, it is possible to provide an image processing method having the same effects as those of the image processing apparatus according to the first aspect of the present invention.

  According to an eighth aspect of the present invention, there is provided an image processing program for causing a computer to detect a face image included in image data and image data from which the face image is detected based on a detection result by the face detection unit. Image classification means for classifying human image data and image data for which no face image has been detected into landscape image data, and image composition for combining the person image data and landscape image data classified by the image classification means It is made to function as a means.

  According to such a configuration, it is possible to provide an image processing program that has the same effect as the image processing apparatus according to the first aspect of the present invention.

  According to the present invention, based on the detection result by the face detection means, the image data is classified into person image data or landscape image data, and the classified person image data and landscape image data are combined, thereby combining the image data. Can be generated, so that the work of synthesizing the image data can be further simplified.

  FIG. 1 is a block diagram showing a configuration of an image processing apparatus 1 according to an embodiment of the present invention. The image processing apparatus 1 includes a control unit including, for example, a CPU (Central Processing Unit), and functions as various functional blocks when the control unit executes a computer program. Specifically, the control unit functions as a face detection unit 11, an image classification unit 12, an image correction unit 13, an image selection unit 14, an image synthesis unit 15, and the like.

  The image processing apparatus 1 includes a memory made up of, for example, a RAM (Random Access Memory), and various storage units are assigned to the storage area of the memory. Specifically, an initial image data storage unit 21, a person image data storage unit 22, a landscape image data storage unit 23, and the like are allocated to the memory. However, each said memory | storage part may be comprised by a separate memory. Each storage unit is not limited to a configuration built in the image processing apparatus 1 and may be configured by an external memory.

  The initial image data storage unit 21 can store a plurality of image data. Examples of the image data stored in the initial image data storage unit 21 include image data read from each frame image of a developed photographic film, image data obtained by photographing with a digital camera, and the like. For example, time information such as shooting date and time is stored in association with each image data. As long as the time information is information representing time, it may be automatically given by the function of the camera at the time of shooting, or may be given by a user operation.

  The face detection unit 11 is a face detection unit that sequentially reads out each image data stored in the initial image data storage unit 21 and detects a face image included in the image data. The face detection unit 11 detects a face image included in the image data by using a well-known face detection technique. For example, the face detection unit 11 detects image feature data such as eyes, nose, mouth, or contour. It is possible to determine whether or not a face image is included.

  The image classification unit 12 is an image classification unit that classifies each image data stored in the initial image data storage unit 21 into person image data or landscape image data based on the detection result by the face detection unit 11. The person image data is image data from which a face image has been detected by the face detection unit 11, and since a face image has been detected, it can be determined that a person image is included. On the other hand, the landscape image data is image data in which no face image is detected by the face detection unit 11, and since no face image is detected, it is determined that the image is a landscape image that does not include a person image. Can do.

  Of the image data classified by the image classification unit 12, the person image data is stored in the person image data storage unit 22, and the landscape image data is stored in the landscape image data storage unit 23. The time information is associated with each person image data stored in the person image data storage unit 22 and each landscape image data stored in the landscape image data storage unit 23.

  The person image data and the landscape image data are appropriately selected by an image selection unit 14 as an image selection unit, read out from the person image data storage unit 22 or the landscape image data storage unit 23, respectively, and image synthesis as an image synthesis unit. Synthesized by the unit 15. At this time, the image selection unit 14 selects the person image data and the landscape image data to be combined by the image combining unit 15 based on the time information respectively associated with the person image data and the landscape image data. More specifically, the image selection unit 14 selects at least one person image data in association with one landscape image data when a predetermined condition is satisfied.

  Thus, in this embodiment, based on the detection result by the face detection unit 11, the image data is classified into person image data or landscape image data, and the classified person image data and landscape image data are synthesized. Since composite image data can be generated, the work of combining image data can be further simplified. In addition, since it is possible to synthesize person image data with, for example, landscape image data taken by itself instead of a template prepared in advance, it is possible to generate synthesized image data rich in originality.

  When synthesizing the person image data and the landscape image data selected by the image selection unit 14, the image composition unit 15 performs an enlargement / reduction process on the person image data or the landscape image data, Is replaced with human image data. Accordingly, it is possible to generate composite image data in which person image data is arranged on a part of the background image using the landscape image data as a background image. Therefore, if person image data and landscape image data related to each other, for example, person image data and landscape image data taken in the same region, are selected, more interesting composite image data can be generated. If the image composition unit 15 is configured to perform processing so that the size of landscape image data is larger than that of human image data, the image composition unit 15 performs processing of reducing human image data or expanding landscape image data. A configuration may be used, or a configuration in which processing for reducing or enlarging person image data and landscape image data, respectively, may be performed.

  The image correction unit 13 is an image correction unit that differentiates the image characteristics of the person image data and the landscape image data by performing image correction on at least one of the person image data and the landscape image data. In this example, the image correction unit 13 performs image correction only on the landscape image data. In this case, the image that reduces the contrast of the landscape image data or increases or decreases the brightness. By performing the correction, it is possible to emphasize the person image data to be combined with the landscape image data.

  However, the configuration is not limited to the above, and for example, a configuration in which image correction is performed only on human image data may be used. In this case, the person image data can be emphasized with respect to the landscape image data by performing image correction such as making the contrast of the person image data higher than that of the landscape image data. Further, the image correction may be performed on the person image data and the landscape image data. In this case, a configuration in which image correction for each image data as described above is performed in combination may be used. The image correction by the image correction unit 13 is not limited to an aspect in which the person image data is emphasized with respect to the landscape image data, but may be an aspect in which the landscape image data is emphasized with respect to the person image data. .

  In this way, by distinguishing the image characteristics between the person image data and the landscape image data, it is possible to generate composite image data in which one of the person image data or the landscape image data is emphasized. For example, as in the present embodiment, if the composite image data in which the person image data is emphasized with respect to the landscape image data is generated, the landscape image data can be combined as a background image.

  The image synthesis unit 15 synthesizes the person image data and the landscape image data after the image correction by the image correction unit 13. At the time of image composition, the image selection unit 14 selects the image data to be composed only when the time interval based on the time information respectively associated with the person image data and the landscape image data is within a predetermined value. The For example, when it is determined that the person image data and the landscape image data are photographed at a time interval greater than or equal to a predetermined value based on time information including the photographing date and time, the image data is combined with the image data. Is not selected as. The predetermined value can be set to an arbitrary value such as a day unit such as “1 day” or a time unit such as “1 hour”.

  In the present embodiment, the image composition unit 15 synthesizes the classified person image data and the landscape image data to generate composite image data using the landscape image data as the background image of the album mount. That is, the image based on the landscape image data combined with the person image data by the image combining unit 15 becomes the background image of the mount when the album is created. The composite image data generated in the image composition unit 15 is sent to an image formation unit 16 as an image forming unit, and an album print is created by generating a photo print based on the composite image data in the image formation unit 16. Is done. The image forming unit 16 can employ a well-known configuration that performs image printing exposure on photographic print paper. The image forming unit 16 and the image processing apparatus 1 create a photographic print. An image forming apparatus is configured.

  According to the above configuration, by combining the classified person image data and the landscape image data, the composite image data having the landscape image data as the background image of the album mount is generated, and the photograph based on the composite image data An album print can be created by generating a print. Therefore, by binding the created album print, it is possible to produce an album with high originality.

  FIG. 2 is a diagram showing an aspect when selecting and combining person image data and landscape image data. In this example, a case where time information corresponding to this order is associated with the six pieces of image data 31 to 36 shown in FIG. That is, among the six pieces of image data 31 to 36, the time information associated with the leftmost image data 31 is the oldest, and the time information associated with the image data is sequentially updated toward the right side. The time information associated with the rightmost image data 36 is the newest.

  Of the six image data 31 to 36, the image data 31 and 34 is landscape image data that does not include a face image, and the other image data 32, 33, 35, and 36 are person image data that includes a face image. It is. Accordingly, image correction for emphasizing the person image data to be synthesized is performed on the landscape image data 31 and 34. At this time, it is preferable that image correction is performed on the landscape image data 31 and 34 in accordance with the image characteristics of the person image data 32, 33, 35, and 36.

  In this example, the time interval based on the time information associated with the landscape image data 31 and the person image data 32 and the time based on the time information associated with the landscape image data 34 and the person image data 35, respectively. The intervals are all within the predetermined value and are combined with each other. In particular, in the present embodiment, when other person image data is adjacent to the person image data to be combined with the landscape image data, the person image data is also combined with the landscape image data. In this example, two person image data 32 and 33 are combined with the landscape image data 31, and two person image data 35 and 36 are combined with the landscape image data 34.

  Landscape image data is often taken as landscape image data. On the other hand, the person image data may be photographed as vertically long image data or may be photographed as horizontally long image data. When photographed as vertically long image data such as the person image data 32 and 33, these person image data 32 and 33 are horizontally arranged so as to be arranged side by side as shown in FIG. By combining with the landscape image data 31, composite image data 41 is generated. In this case, the person image data 32 and 33 may be arranged in parallel to the horizontal direction as in the example of FIG. 2B, or may be arranged in a direction inclined with respect to the horizontal direction. However, the portrait image data 32 and 33 can be combined with the landscape image data 31 so as to be arranged vertically.

  On the other hand, when photographed as landscape image data such as person image data 35 and 36, these person image data 35 and 36 are arranged side by side as shown in FIG. In this way, by combining the landscape image data 34 with landscape, the composite image data 42 is generated. In this case, the human image data 35 and 36 may be arranged in parallel to the vertical direction, or may be arranged in a direction inclined with respect to the vertical direction as in the example of FIG. However, the horizontally long person image data 35 and 36 can be combined with the horizontally long landscape image data 34 so as to be arranged side by side.

  In the present embodiment, based on the time information respectively associated with the person image data and the landscape image data, the person image data and the landscape image data having higher relevance are automatically selected, and the combined image data is selected. Can be generated. For example, since human image data and landscape image data associated with relatively close time information are likely to have been shot in the same region, combining them automatically makes for a more elegant combination Image data can be easily generated.

  Further, in the present embodiment, when the time interval based on the time information respectively associated with the person image data and the landscape image data exceeds a predetermined value, those image data are not combined. Image data with long time intervals are likely to have been shot in different areas. In such a case, image data with low relevance can be created by not creating composite image data. Combining each other can be prevented. Therefore, it is possible to easily generate only more interesting composite image data.

  If the configuration is such that the two person image data 32 and 33 are combined side by side with the landscape image data 31 that is horizontally long as shown in FIG. When printing as a two-page spread image, person image data can be arranged on each of the left and right pages. Therefore, it is suitable for a service for producing an album by binding a printed composite image data.

  However, the number of person image data to be combined with the landscape image data is not limited to two and may be three or more. The landscape image data is not limited to horizontally long image data, and may be vertically long image data. In this case, the configuration may be such that portrait image data is vertically arranged and combined with vertically long landscape image data, or may be configured to be combined horizontally.

  FIG. 3 is a flowchart showing an example of processing when classifying each image data stored in the initial image data storage unit 21 into person image data or landscape image data. As shown in FIG. 3, when classifying image data stored in the initial image data storage unit 21, face detection is sequentially performed on each image data (step S101: face detection step).

  If a face image is detected by the face detection (Yes in step S102), the image data is classified as human image data (step S103), and if no face image is detected (in step S102). No), the image data is classified into landscape image data (step S104). When such an image classification step (steps S102 to S104) is performed on all image data stored in the initial image data storage unit 21 (Yes in step S105), the image data is classified. This process ends.

  FIG. 4 is a flowchart showing an example of processing when selecting and combining classified person image data and landscape image data. In the example illustrated in FIG. 4, out of the classified person image data and landscape image data, the person image data is sequentially read from the person image data storage unit 22 and is combined with the landscape image data only when a predetermined condition is satisfied. It has come to be. In this example, the case where the human image data is reduced and combined with the landscape image data will be described. However, as described above, the configuration may be such that the landscape image data is enlarged. The configuration may be such that data is reduced or enlarged, respectively.

  The person image data is read from the person image data storage unit 22 (step S201), and landscape image data whose time interval based on the associated time information is within a predetermined value for the person image data is the landscape image data. If it is stored in the storage unit 23 (Yes in step S202), the landscape image data is selected as image data to be synthesized (step S203: image selection step). At this time, if there is only one corresponding landscape image data, the landscape image data is selected as image data to be synthesized. On the other hand, if there are a plurality of corresponding landscape image data, the landscape image data with the shortest time interval based on the time information associated with the person image data, that is, the landscape image data closest in time is synthesized. Selected as image data. However, the present invention is not limited to the configuration in which the landscape image data closest in time is selected from the plurality of landscape image data. For example, when the user operates an operation unit (not shown), the plurality of landscape image data is selected. The configuration may be such that desired landscape image data can be selected from the list.

  The landscape image data selected in this way is subjected to image correction for emphasizing the person image data with respect to the landscape image data (step S204: image correction step). If there is human image data adjacent to the read human image data (Yes in step S205), that is, when the human image data and the landscape image data are arranged in time series, the landscape image data is interposed between them. If there is person image data adjacent without being sandwiched, the two person image data including the person image data is reduced (step S206), and a part of the landscape image data after image correction is reduced 2 By being replaced with one person image data, composite image data is generated (step S208).

  On the other hand, when there is no person image data adjacent to the read person image data (No in step S205), only the read person image data is reduced (step S207), and the image after image correction is corrected. By replacing a part of the landscape image data with the reduced human image data, composite image data is generated (step S208).

  Such image composition steps (steps S206 to S208) are performed only for the person image data that satisfies the predetermined condition as described above, and the predetermined conditions for all the person image data are determined and combined. At the time (Yes in step S209), the process for synthesizing the image data ends. The image data selected as the adjacent person image data in step S204 is preferably configured to be excluded from the determination target of the predetermined condition.

  In the above embodiment, the configuration in which the classified person image data and landscape image data are automatically combined has been described. However, the present invention is not limited to such a configuration, and a configuration in which subsequent composition is performed manually is acceptable as long as each image data is classified into person image data or landscape image data. In this case, the person image data is not limited to the composition that is synthesized with the landscape image data, but may be the composition that is synthesized with the template, for example. Here, as image data to be combined with a template such as a background image, human image data including a human face image is often selected. Therefore, by classifying image data into person image data or landscape image data as described above, person image data is extracted from a plurality of image data, and desired image data can be easily extracted from these person image data. Therefore, the work of synthesizing image data can be further simplified.

1 is a block diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention. It is the figure which showed the aspect at the time of selecting and synthesize | combining person image data and landscape image data. It is the flowchart which showed an example of the process at the time of classifying each image data memorize | stored in the initial image data memory | storage part into person image data or landscape image data. It is the flowchart which showed an example of the process at the time of selecting and synthesize | combining the classified person image data and landscape image data.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 11 Face detection part 12 Image classification part 13 Image correction part 14 Image selection part 15 Image composition part 16 Image formation part 21 Initial image data storage part 22 Person image data storage part 23 Landscape image data storage part

Claims (8)

Face detection means for detecting a face image included in the image data;
Image classification means for classifying image data in which a face image is detected into person image data based on a detection result by the face detection means, and classifying image data in which no face image is detected into landscape image data;
An image processing apparatus comprising: an image synthesis unit that synthesizes the person image data and the landscape image data classified by the image classification unit.   The image synthesizing unit performs processing on the person image data or the landscape image data so that the size of the landscape image data is larger than that of the person image data, and after the processing, a part of the landscape image data The image processing apparatus according to claim 1, wherein the person image data is replaced. Image correction means for differentiating image characteristics between the person image data and the landscape image data by performing image correction on at least one of the person image data and the landscape image data classified by the image classification means. With
The image processing apparatus according to claim 1, wherein the image composition unit synthesizes the person image data and the landscape image data after the image correction by the image correction unit. The person image data and the landscape image data are respectively associated with time information,
The image processing according to any one of claims 1 to 3, further comprising image selection means for selecting the person image data and the landscape image data to be synthesized by the image synthesis means based on the time information. apparatus.   The image selecting means synthesizes the image data by the image synthesizing means only when the time interval based on the time information respectively associated with the person image data and the landscape image data is within a predetermined value. The image processing apparatus according to claim 4, wherein the image processing apparatus is selected as image data to be processed. Face detection means for detecting a face image included in the image data;
Image classification means for classifying image data in which a face image is detected into person image data based on a detection result by the face detection means, and classifying image data in which no face image is detected into landscape image data;
Image synthesizing means for synthesizing the person image data and the landscape image data classified by the image classifying means to generate synthesized image data using the landscape image data as a background image of an album mount;
An image forming apparatus comprising: image forming means for creating an album print by generating a photographic print based on the composite image data. A face detection step for detecting a face image included in the image data;
An image classification step for classifying image data in which a face image is detected into person image data and classifying image data in which no face image is detected into landscape image data based on the detection result in the face detection step;
An image processing method comprising: an image synthesis step for synthesizing the person image data and the landscape image data classified by the image classification step. Computer
Face detection means for detecting a face image included in the image data;
Image classification means for classifying image data in which a face image is detected into person image data and classifying image data in which no face image is detected into landscape image data based on the detection result by the face detection means;
An image processing program that functions as an image combining unit that combines the person image data and the landscape image data classified by the image classification unit.

Images