JP2011125041A - Image processing apparatus, method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily manage an original image and an interpolation image when generating the interpolation image from the original image. <P>SOLUTION: An interpolation image generating section 5 generates the interpolation image from the original images L1 and R1 obtained by photographing a subject from different viewpoints. A file generating section 6 describes a viewpoint order of the original images L1, R1 and the interpolation image, and an identification code showing the type of the images to a header of each image, to generate one image file including the original images L1, R1 and the interpolation image. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and method for generating an interpolated image for interpolating the viewpoints of a plurality of original images when a subject is viewed from different viewpoints, and a program for causing a computer to execute the image processing method.

  A stereoscopic image that can be stereoscopically viewed using parallax is acquired by shooting with a compound-eye imaging device such as a stereo camera that acquires an image using at least a pair of imaging units arranged at different positions. be able to. In such a compound eye photographing apparatus, a convergence angle is given to the optical axes of a pair of photographing means, and the same subject is photographed from different viewpoints to obtain two left and right images, and parallax is used from these two images. A stereoscopic image that can be viewed stereoscopically is created.

  Here, a method of saving a plurality of images acquired in such a compound eye photographing apparatus as one image file has been proposed (see Patent Document 1). According to the technique described in Patent Document 1, since a plurality of images can be managed by one image file, it is possible to prevent the number of image files from increasing.

JP 2005-252754 A

  By the way, as described above, when the angle of convergence of the photographing unit for acquiring the stereoscopic image is large, the parallax between the acquired images increases. When the parallax between images increases, the amount of horizontal displacement between the left and right images on the stereoscopic display surface increases, resulting in an uncomfortable feeling. In this case, by predicting the viewpoint position between the photographing means and creating an interpolated image that interpolates two images, the parallax between adjacent images can be reduced, so that stereoscopic viewing can be performed well. It becomes possible. The interpolated image can be predicted and created not only between a plurality of images but also outside the plurality of images.

  However, when an interpolation image is created in this way, the original image and the interpolation image, which are images acquired by shooting, are managed as separate files. For this reason, the original image and the interpolated image are easily confused, and when the interpolated image is created again from the original image or only the interpolated image is deleted to reduce the file capacity, the original image is erroneously deleted. There is a fear.

  The present invention has been made in view of the above circumstances, and an object thereof is to easily manage an original image and an interpolation image.

An image processing apparatus according to the present invention includes an image acquisition unit that acquires a plurality of original images obtained by viewing a subject from different viewpoints;
Interpolated image generating means for generating at least one interpolated image for interpolating a viewpoint between the plurality of original images or outside the plurality of original images from the plurality of original images;
The plurality of original images and the interpolated images are provided with identification information indicating the types of the original images and the interpolated images, and viewpoint information indicating the viewpoint order, and the plurality of the identification information and the plurality of viewpoint information provided with the viewpoint information The image processing apparatus includes a storage unit that stores the original image and the interpolation image in association with each other.

  “Saving a plurality of original images and interpolated images in association with each other” means that the plurality of original images and the interpolated images are inseparable from each other for management purposes. Save in the same folder, give a common file name to a plurality of original images and interpolation images, save a plurality of original images and interpolation images in one file, and at least one of the plurality of original images and interpolation images A plurality of original images and interpolated images can be stored in association with each other by a method such as recording a file name of another image in the header area. Even when a plurality of original images and interpolated images are stored in association with each other, operations such as access to the original image or the interpolated image alone are permitted.

  In the image processing apparatus according to the present invention, the storage means may be means for storing the plurality of original images and the interpolated image as one image file.

  In the image processing apparatus according to the present invention, the storage unit may be a unit that stores the interpolated image in a region after the original image in the image file.

  “Save in the subsequent area” means that, for example, when the original image and the interpolated image are integrally recorded in a continuous address space on the recording medium, the area where the interpolated image is arranged is the area of the original image. It may be arranged to be in a lower area.

  In the image processing apparatus according to the present invention, the storage unit may be a unit that deletes only the interpolation image when an instruction to reduce the file capacity is given.

  “Instructions to reduce the file capacity” include, for example, instructions to increase the free capacity of the recording medium that records the image files of the original image and the interpolated image, instructions to delete the image file, and processing with a smaller file capacity is easier For example, an instruction to transfer an image file can be given.

  Further, the “instruction to reduce the file capacity” may be performed in units of image files of the associated original image and interpolated image. For example, in the case of reducing the file capacity by deleting, as the deletion menu, an instruction to delete all of the original image and the interpolated image for the selected image, and an instruction to reduce the capacity of the selected image If the latter is instructed, the process of deleting only the interpolated image may be performed. Further, for example, when the user instructs deletion of a part of the original image and the interpolated image (for example, one interpolated image), the user is informed that there is an image related to the image. In addition to notifying, the user may be notified that the above-described deletion menu (or at least a capacity reduction instruction from the deletion menu) is prepared.

  In the image processing apparatus according to the present invention, when the original image is acquired by photographing, the interpolated image generating means determines the number of the interpolated images based on the positional relationship of the photographing means that acquired the original image. It is good also as a means to determine.

  For example, at least one of a convergence angle and a baseline length in a plurality of imaging units can be used as the positional relationship of the imaging units that acquire the original image.

In addition, the imaging device according to the present invention includes a plurality of imaging units arranged at different positions to acquire the original image,
The image processing apparatus according to the present invention is provided.

  The image processing method according to the present invention may be provided as a program for causing a computer to execute the image processing method.

  According to the present invention, identification information indicating the type of the original image and the interpolation image and viewpoint information indicating the viewpoint order are assigned to the plurality of original images and interpolation images, and the plurality of original images to which the identification information and viewpoint information are assigned. In addition, the interpolation image is stored in association with each other. For this reason, management of an original image and an interpolation image becomes easy. In particular, since the identification information indicating the type of the original image and the interpolation image is given, it is possible to prevent the original image from being accidentally deleted at the time of deletion. In addition, since viewpoint information representing the viewpoint order is also given, it is possible to easily generate a stereoscopic image and regenerate an interpolated image.

  Further, by storing a plurality of original images and interpolation images as one image file, the number of files is not increased, and as a result, management of the plurality of original images and interpolation images can be made easier. it can.

  In this case, by storing the interpolation image in a region after the original image, it is not necessary to regenerate the image file when the interpolation image is deleted, so that the interpolation image can be easily deleted.

  Further, by deleting only the interpolated image when an instruction to reduce the file capacity is given, the file capacity can be easily reduced.

  In addition, when the original image is acquired by photographing, it is possible to create the interpolated image so as to obtain an appropriate number of viewpoints by determining the number of the interpolated images based on the positional relationship of the photographing unit that acquired the original image. it can.

1 is a schematic block diagram showing a configuration of a compound eye photographing apparatus to which an image processing apparatus according to an embodiment of the present invention is applied. Diagram showing the configuration of the shooting unit The figure for demonstrating arrangement | positioning of an imaging | photography part Figure showing the original image Diagram for explaining generation of interpolation image The figure which shows the relationship between a solid shape model and the image projection surface which acquired the original image The figure which shows the file structure of the original image file in this embodiment The figure which shows the file structure of the image file containing the interpolation image in this embodiment The flowchart which shows the process performed at the time of the interpolation performed in this embodiment The flowchart which shows the process performed when the process to reduce the file capacity is instructed

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic block diagram showing a configuration of a compound eye photographing apparatus to which an image processing apparatus according to an embodiment of the present invention is applied. As shown in FIG. 1, the compound eye photographing apparatus 1 according to the present embodiment includes two photographing units 2A and 2B, a photographing control unit 3, an image processing unit 4, an interpolated image generating unit 5, a file generating unit 6, and a stereoscopic image generating unit 7. , A frame buffer 8, a media control unit 9, an input unit 10 including operation buttons for performing various inputs, a display unit 11 such as a liquid crystal monitor for performing various displays, and a control unit 12. Yes.

  FIG. 2 is a diagram illustrating the configuration of the imaging units 2A and 2B. As shown in FIG. 2, the photographing units 2A and 2B include lenses 20A and 20B, diaphragms 21A and 21B, shutters 22A and 22B, CCDs 23A and 23B, analog front ends (AFE) 24A and 24B, and A / D conversion units 25A, 25B is provided. As shown in FIG. 3, the photographing units 2A and 2B are arranged so as to have a predetermined baseline length K with a convergence angle α that allows the subject to be seen. Information about the convergence angle α and the baseline length K is stored in the ROM 12C described later.

  The lenses 20A and 20B are composed of a plurality of functional lenses such as a focus lens for focusing on a subject and a zoom lens for realizing a zoom function, and focus data obtained by AF processing performed by the imaging control unit 3. Based on zoom data obtained by operating a zoom lever (not shown), the position is adjusted by a lens driving unit (not shown).

  The apertures 21A and 21B are adjusted in aperture diameter by an aperture drive unit (not shown) based on aperture value data obtained by AE processing performed by the imaging control unit 3.

  The shutters 22A and 22B are mechanical shutters and are driven by a shutter driving unit (not shown) according to the shutter speed obtained by the AE process.

  The CCDs 23A and 23B have a photoelectric surface in which a large number of light receiving elements are two-dimensionally arranged, and subject light is imaged on the photoelectric surface and subjected to photoelectric conversion to acquire an analog imaging signal. In addition, color filters in which R, G, and B color filters are regularly arranged are arranged on the front surfaces of the CCDs 23A and 23B.

  The AFEs 24A and 24B perform processing for removing noise of the analog imaging signal and processing for adjusting the gain of the analog imaging signal (hereinafter referred to as analog processing) for the analog imaging signals output from the CCDs 23A and 23B.

  The A / D converters 25A and 25B convert the analog imaging signals subjected to analog processing by the AFEs 24A and 24B into digital signals. Note that images L1 and R1 are images represented by digital image data acquired by the photographing units 2A and 2B, respectively.

  The imaging control unit 3 includes an AF processing unit and an AE processing unit (not shown). The AF processing unit determines the focal lengths of the lenses 20A and 20B based on the pre-images acquired by the imaging units 2A and 2B by half-pressing the release button included in the input unit 10, and outputs the focal lengths to the imaging units 2A and 2B. The AE processing unit determines the aperture value and the shutter speed based on the pre-image, and outputs them to the photographing units 2A and 2B. Note that it is also possible to perform shooting using a predetermined focus position, aperture value, and shutter speed without performing AF processing and AE processing.

  The photographing control unit 3 instructs the photographing units 2A and 2B to acquire the main images of the stereoscopic image generation images L1 and R1 by pressing the release button fully. Note that before the release button is operated, the photographing control unit 3 instructs the photographing units 2A and 2B to acquire a through image having a smaller number of pixels than the main image for confirming the photographing range.

  Here, the images L1 and R1 are obtained by photographing the subject at two different photographing positions as shown in FIG. 3, and the subjects included in the images L1 and R1 as shown in FIG. Has a parallax corresponding to a difference in photographing position. The images L1 and R1 are acquired at the photographing positions of the photographing units 2A and 2B, that is, the left and right photographing positions toward the subject, and generate a stereoscopic image with the images L1 and R1 as a reference. In the description, the images L1 and R1 are referred to as original images L1 and R1. Note that an image displayed on the left side when generating a stereoscopic image is an image L1, and an image displayed on the right side is an image R1.

  In the present embodiment, a stereoscopic image is generated from the two original images L1 and R1, but three or more acquired by performing photographing at three or more different photographing positions with three or more photographing units. A plurality of images may be generated from the original image.

  The image processing unit 4 uses the digital image data of the original images L1 and R1 acquired by the imaging units 2A and 2B to shift the angle of view of the imaging units 2A and 2B, the difference in zoom magnification, and the image generated by the rotation of the CCD. Misalignment, correction processing for correcting trapezoidal distortion and the like caused by shooting the subject with the convergence angle α, and processing for adjusting white balance, gradation correction, sharpness correction, color correction, etc. Apply image quality correction processing. Note that the reference symbols L1 and R1 before processing are also used for the original images L1 and R1 after processing in the image processing unit 4.

  The interpolated image generation unit 5 generates at least one interpolated image from the original images L1 and R1 acquired by shooting or the original images L1 and R1 acquired in advance and recorded on the medium 9A. In the present embodiment, the number of interpolated images corresponding to the size of the convergence angle α of the photographing units 2A and 2B is created.

Here, when a stereoscopic image is created and stereoscopically viewed, if the convergence angle of the imaging units 2A and 2B is increased to, for example, 1 degree or more, the parallax between the original images L1 and R1 increases. The amount of horizontal displacement between the left and right images on the stereoscopic display surface increases, resulting in a feeling of strangeness. Therefore, the interpolated image generation unit 5 determines whether or not the convergence angle α of the imaging units 2A and 2B described in the ROM 12C or described as header information of the image file described later is less than 1 degree. If not, the maximum N value is set as the number of interpolations so that the value obtained by dividing the convergence angle α by N + 1 is less than 1. That is, the interpolation number N is set by obtaining the maximum N that satisfies the relationship of the following equation (1). Then, an interpolation image of the set interpolation number N is generated.
1 degree <α / (N + 1) (1)
Note that an interpolated image may be generated in the compound eye photographing apparatus 1 according to the present embodiment from an image acquired by a photographing apparatus having a different convergence angle from the photographing units 2A and 2B of the compound eye photographing apparatus 1 according to the present embodiment. In this case, the convergence angle only needs to be described in the header of the image file of the image, but the convergence angle may not be described. In this case, the interpolated image generation unit 5 obtains corresponding points corresponding to each other between the original images, and calculates a convergence angle based on the corresponding points.

  FIG. 5 is a diagram for explaining generation of an interpolation image. Here, the number of interpolations is 2. As shown in FIG. 5, the interpolated image generation unit 5 generates two interpolated images H1 and H2 from the original images L1 and R1. Interpolated images H1 and H2 are generated so that the convergence angles in adjacent original image L1 and interpolated image H1, interpolated image H1 and interpolated image H2, and interpolated image H2 and original image R1 are equal.

  Here, the interpolated image generation unit 5 processes the original images L1 and R1 by morphing so that the parallax between corresponding pixels included in the original images L1 and R1 is gradually reduced. Generate. Specifically, corresponding points corresponding to each other in the left and right images are detected, the corresponding points are connected by a straight line or a curve, a pseudo corresponding point is calculated by dividing the straight line or the curve, and the original images L1 and R1 are converted into pseudo corresponding points. The interpolation images H1 and H2 are generated by deforming to correspond to. Also, any known method described in, for example, Japanese Patent Laid-Open No. 2002-190020 can be used.

  Here, the interpolation image is generated between the viewpoints of viewing the original images L1 and R1, but the interpolation image H3 is predicted by predicting the viewpoint outside the original image L1 as in the interpolation image H3 shown in FIG. You may make it produce | generate.

  Further, instead of morphing, a solid shape model of a subject included in the original images L1 and R1 may be created, and an interpolation image may be generated from the solid shape model. The three-dimensional shape model is created by obtaining corresponding points between the original images L1 and R1 and calculating the distance from the shooting position to the corresponding point on the subject based on the principle of triangulation based on the parallax between the corresponding points. be able to. Here, at the time of shooting, a straight line connecting each point on the subject and the focal points of the shooting units 2A and 2B corresponds to an image projection plane on which an image is projected (the imaging planes of the CCDs 23A and 23B of the shooting units 2A and 2B). A set of points intersecting with the plane to be the projection image, that is, the original images L1 and R1. FIG. 6 is a diagram showing the relationship between the three-dimensional shape model and the image projection plane from which the original images L1 and R1 are obtained. Here, the three-dimensional model for explanation is a cube. Therefore, as shown in FIG. 6, a virtual image projection plane T corresponding to the number of interpolations is set between the image projection planes of the original images L1 and R1 (the number of interpolations is 1 in FIG. 6). An interpolation image may be generated by obtaining a projection image on the image projection plane T.

  The file generation unit 6 performs a compression process on the image data of the original images L1 and R1 in a compression format such as JPEG to generate an original image file F0. When the interpolated images H1 and H2 are generated, the image file F1 is generated from the image data of the original images L1 and R1 and the interpolated images H1 and H2. Based on the Exif format or the like, a header storing additional information such as the shooting date and time is added to the image file.

  FIG. 7 is a diagram showing the file structure of the original image file in the present embodiment. As shown in FIG. 7, the image file F0 includes the header area (header 1) 40 of the original image L1, the image data area 41 of the original image L1, the header area (header 2) 42 of the original image R1, and the image of the original image R1. The data area 43 is configured by arranging in this order.

  In the header area 40 of the original image L1, an address start position, attribute information, and incidental information are described as header information. The address start position is the start position of the address of the header area 42 of the original image R1, and is described as a list. The attribute information includes a viewpoint order, an image identification code, and shooting condition information. Here, the viewpoint order is the viewpoint order when the original images L1 and R1 are viewed from the left, and the viewpoint order is 1 and 2 for each of the original image L1 and the original image R1. The image identification code is information indicating whether the image stored in the image data area corresponding to the header area is an original image or an interpolated image, and is 1 for an original image and 2 for an interpolated image. Further, the imaging condition information is information on the convergence angle α and the baseline length K of the imaging units 2A and 2B, and information stored in the ROM 12C is used. Note that the viewpoint order of the original image L1 is 1, and the identification code of the image is 1 representing the original image. Further, the incidental information is information such as the shooting date and time.

  In the header area 42 of the original image R1, the attribute information of the original image R1, that is, the viewpoint order (= 2) and the image identification code (= 1) are described as header information.

  FIG. 8 is a diagram showing a file structure of an image file including an interpolation image in the present embodiment. As shown in FIG. 8, the image file F1 includes a header area (header 1) 50 of the original image L1, an image data area 51 of the original image L1, a header area (header 2) 52 of the original image R1, and image data of the original image R1. The area 53, the header area (header 3) 54 of the interpolation image H1, the image data area 55 of the interpolation image H1, the header area (header 4) 56 of the interpolation image H2, and the image data area 57 of the interpolation image H2 are arranged in this order. It is constituted by being done.

  In the header area 50 of the original image L1, an address start position, attribute information, and incidental information are described as header information. The address start position is the address start position of the header areas 52, 54, and 56 (that is, the headers 2, 3, and 4) of the original image R1 and the interpolation images H1 and H2, and is described as a list. The attribute information includes a viewpoint order, an image identification code, and shooting condition information. Here, the viewpoint order is the viewpoint order when the original images L1, R1 and the interpolated images H1, H2 are viewed from the left, and the viewpoint order is determined for each of the original image L1, the interpolated image H1, the interpolated image H2, and the original image R1. Are 1, 2, 3, and 4. The image identification code is information indicating whether the image stored in the image data area corresponding to the header area is an original image or an interpolated image, and is 1 for an original image and 2 for an interpolated image. The imaging condition information is information on the convergence angle α and the baseline length K of the imaging units 2A and 2B. The viewpoint order of the original image L1 is 1, and the image identification code is 1 representing the original image. Further, the incidental information is information such as the shooting date and time.

  In the header area 52 of the original image R1, attribute information of the original image R1, that is, the viewpoint order (= 4) and the image identification code (= 1) are described as header information.

  In the header area 54 of the interpolated image H1, attribute information of the interpolated image H1, that is, viewpoint order (= 2), image identification code (= 2), and interpolation method information are described as header information. The interpolation method is a method for creating an interpolation image, and describes information on a method used when the interpolation image generation unit 5 generates the interpolation images H1 and H2.

  In the header area 56 of the interpolation image H2, attribute information of the interpolation image H2, that is, viewpoint order (= 3), image identification code (= 2), and interpolation method information are described as header information.

  The stereoscopic image generating unit 7 performs a stereoscopic display process on the original images L1 and R1 and the interpolated images H1 and H2 so that stereoscopic display is possible on the display unit 11, and generates a stereoscopic image. Note that the stereoscopic display conversion depends on the way of stereoscopic viewing and whether or not the display unit 11 is a 3D liquid crystal, but the colors of the paired images (that is, the images in which the viewpoint order is adjacent) are, for example, red and blue. Can be superimposed (anaglyph method), and the polarization directions of the paired images can be superimposed (polarized filter method), or the paired images can be alternately combined for each line. Yes (Parallax barrier method and lenticular method).

  The frame buffer 8 is a storage unit that temporarily stores an image to be displayed on the display unit 11.

  The media control unit 9 controls the writing and reading of the image file F0 by accessing the recording medium 9A.

  The control unit 12 is a work area when the apparatus 1 performs various processes such as a CPU 12A, when generating an interpolation image as described later, when generating a stereoscopic image, when generating an image file, and when displaying a stereoscopic image. And a ROM 12C that stores the operation program of the apparatus 1 and various constants, etc., and controls the operation of each part of the apparatus 1.

  Next, processing performed in the present embodiment will be described. FIG. 9 is a flowchart showing processing performed at the time of interpolation in the present embodiment. Here, it is assumed that the original images L1 and R1 are acquired by photographing and recorded on the medium 9A. When an instruction for generating an interpolated image is issued from the input unit 10, the control unit 12 starts processing, and the media control unit 9 reads the original image file F0 from the media 9A (step ST1). The original image file F0 is decompressed to the RAM 12B, and the interpolated image generation unit 5 determines whether or not the convergence angle α is described as the shooting condition information of the header information of the original image file F0 (step ST2). If step ST2 is negative, a convergence angle α is calculated from the distribution of corresponding points between the original images L1 and R1 (step ST3). If the convergence angle α is always described in the original image file F0, the determination in step ST2 may not be performed.

  When step ST2 is affirmed and following step ST3, it is determined whether or not the convergence angle α is less than 1 degree (step ST4). If step ST4 is affirmed, the generation of an interpolated image is not necessary, and the process ends. At that time, a warning may be displayed on the display unit 12 indicating that the generation of the interpolation image is not necessary. If step ST4 is negative, the interpolated image generation unit 5 sets the number of interpolations (step ST5). The number of interpolations is set by calculating the maximum value of N that satisfies the above equation (1).

  Then, the interpolation image generation unit 5 generates an interpolation image with the number of interpolations N from the original images L1 and R1 (step ST6). Next, the file generation unit 6 sets the viewpoint order of the original images L1 and R1 and the interpolation image (step ST7), and generates addresses of the data areas of the original images L1 and R1 and the interpolation image in the image file to be generated (step ST7). ST8). Then, header information for the interpolated image is generated (step ST9), and the header information is updated for the original images L1 and R1 (step ST10). Subsequently, the image file F1 is obtained from the original images L1 and R1 and the interpolated image. Is generated (step ST11). Then, the media control unit 9 records the image file F1 on the media 9A (step ST12), and the process ends.

  Next, a process when a process for reducing the file capacity is instructed with respect to the generated image file F1 will be described. FIG. 10 is a flowchart showing a process performed when a process for reducing the file capacity is instructed. The “instruction to reduce the file capacity” is, for example, an instruction to increase the free capacity of the recording medium for recording the image file, and the processing can be performed more easily when the file capacity is smaller. Instructions are included.

  The control unit 12 starts processing when an instruction for processing to reduce the file capacity is issued from the input unit 10, and reads the identification code from the header information included in the image file F1 (step ST21). Then, the header area and the image data area in which the read identification code is an interpolation image are deleted from the image file F1 (interpolation image deletion, step ST22). Further, the header information is updated for the remaining original images L1 and R1 (step ST23). Here, the viewpoint order is changed in the header information of the original images L1 and R1. That is, when the original images L1 and R1 and the two interpolated images H1 and H2 are included in the image file F1, the viewpoint order of the original images L1 and R1 is changed from 1 to 4 to 1 and 2, respectively. Is done. In addition, information about the interpolation image is deleted from the offset list included in the header information of the original image L1. Then, the media control unit 9 records the original image file F0 on the media 9A (step ST24) and ends the process.

  As described above, in the first embodiment, as the header information of the original images L1 and R1 and the interpolation image, the identification code indicating the type of the original image and the interpolation image and the viewpoint information indicating the viewpoint position are described, and the identification code The original images L1, R1 and the interpolated image to which the header describing the viewpoint information is added are stored and saved in one image file F1. This facilitates management of the original images L1 and R1 and the interpolation image. In particular, since the identification codes representing the types of the original images L1 and R1 and the interpolation image are given, it is possible to prevent the original image from being accidentally deleted at the time of deletion. Further, since viewpoint information representing the viewpoint position is also given, it is possible to easily generate a stereoscopic image and regenerate an interpolated image.

  Further, by saving the original images L1, R1 and the interpolation image as one image file F1, the number of files is not increased, and as a result, the management of the original images L1, R1 and the interpolation image is made easier. can do.

  Further, in the image file F1, since the interpolation image is stored in a region after the original images L1 and R1, it is not necessary to regenerate the file when the interpolation image is deleted. As a result, the interpolation image is deleted. It becomes easy.

  Further, by deleting only the interpolated image when an instruction to reduce the file capacity is given, the file capacity can be easily reduced.

  In the above embodiment, an interpolation image is created by an instruction to create an interpolation image. However, an interpolation image may be created following acquisition of the original images L1 and R1 by shooting.

  In the above embodiment, when processing for reducing the file capacity is instructed, the area of the interpolated image is deleted, but in order to facilitate recombination from the original images L1 and R1 later. It is preferable to leave information on the number of interpolations as header information.

  In the above embodiment, the image processing apparatus according to the present invention is applied to the compound eye photographing apparatus including the photographing units 2A and 2B. However, the image processing apparatus according to the present invention may be provided alone. In this case, a plurality of images acquired by photographing the same subject at a plurality of different positions are input to the image processing apparatus, and an image file F1 is created as in the above embodiment. In this case, the image file F1 is generated not only in the original images L1 and R1 acquired by shooting, but also in the case of generating an interpolated image using an original image created by computer graphics, for example, as in the above embodiment. can do.

  In the above embodiment, the original images L1, R1 and the interpolated image are associated with each other by storing the original images L1, R1 and the interpolated image in one file. However, the original images L1, R1 and the interpolated image are integrated. The original image L1, R1 and the interpolated image may be stored in the same folder, a common file name may be assigned to the original image L1, R1, and the interpolated image, or the original image L1, R1 and the interpolated image may be inseparable. A plurality of original images and interpolated images may be associated by a method such as recording the file name of another image in at least one header of the image.

  Although the apparatus 1 according to the embodiment of the present invention has been described above, the computer is caused to function as a unit corresponding to the interpolation image generation unit 5 and the file generation unit 6 described above, and processing as illustrated in FIGS. 9 and 10 is performed. The program to be executed is also one embodiment of the present invention. A computer-readable recording medium in which such a program is recorded is also one embodiment of the present invention.

DESCRIPTION OF SYMBOLS 1 Compound eye imaging device 2A, 2B Image pick-up part 3 Image pick-up control part 4 Image processing part 5 Interpolation image generation part 6 File generation part 7 Stereo image generation part 8 Frame buffer 9 Media control part 10 Input part 11 Display part 12 Control part

Claims (8)

Image acquisition means for acquiring a plurality of original images obtained by viewing a subject from different viewpoints, acquired by a plurality of imaging means;
An interpolated image that determines the number of interpolated images for interpolating viewpoints between the plurality of original images based on the positional relationship of the plurality of photographing means, and generates the determined number of interpolated images from the plurality of original images Generating means;
The identification information indicating the type of the original image and the interpolation image and the viewpoint information indicating the viewpoint order are assigned to the plurality of original images and the interpolation image, and the plurality of the identification information and the viewpoint information are added to the plurality of original images and the interpolation image. An image processing apparatus comprising: a storage unit that stores an original image and the interpolated image in association with each other.   The image processing apparatus according to claim 1, wherein the storage unit is a unit that stores the plurality of original images and the interpolated image as one image file.   The image processing apparatus according to claim 2, wherein the storage unit is a unit that stores the interpolated image in a region after the original image in the image file.   The image processing apparatus according to claim 1, wherein the storage unit is a unit that deletes only the interpolated image when an instruction to reduce a file capacity is given.   5. The interpolated image generating means is means for predicting a viewpoint outside the plurality of original images, and further generating an interpolated image of the predicted viewpoint. The image processing apparatus described. A plurality of photographing means disposed at different positions to obtain the original image;
An imaging apparatus comprising: the image processing apparatus according to claim 1. Acquire a plurality of original images obtained by a plurality of photographing means and viewed from different viewpoints,
Determining the number of interpolated images for interpolating viewpoints between the plurality of original images based on the positional relationship of the plurality of imaging means;
Generating the determined number of interpolated images from the plurality of original images;
The identification information indicating the type of the original image and the interpolation image and the viewpoint information indicating the viewpoint order are assigned to the plurality of original images and the interpolation image, and the plurality of the identification information and the viewpoint information are added to the plurality of original images and the interpolation image. An image processing method comprising: storing an original image and the interpolated image in association with each other. A procedure for acquiring a plurality of original images obtained by a plurality of photographing means and viewing a subject from different viewpoints;
A procedure for determining the number of interpolated images for interpolating viewpoints between the plurality of original images based on the positional relationship of the plurality of photographing means;
Generating the determined number of interpolated images from the plurality of original images;
The plurality of original images and the interpolated images are provided with identification information indicating the types of the original images and the interpolated images, and viewpoint information indicating the viewpoint order, and the plurality of the identification information and the plurality of viewpoint information provided with the viewpoint information A program for causing a computer to execute an image processing method, comprising: storing an original image and the interpolated image in association with each other.

Images