JP2014147034A - Image processing apparatus, image processing method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to calculate, in an image after distortion aberration correction processing, a rectangular area including a photographing area as much as possible, even in the case where a circumferential fisheye lens is used and the photographing area is not the entire area of the image.SOLUTION: An image processing apparatus 1 includes a correction unit 102 that performs distortion aberration correction processing to an image, and a calculation unit 103 that selects one of a plurality of predetermined processing methods according to photographing conditions of the image and correction conditions defined by the correction unit 102, and calculates the size of an area cut out from the image after the correction processing.

Description

  The present invention relates to an image processing apparatus, an image processing method, and a program. Specifically, the present invention relates to an image processing device, an image processing method, and a program for correcting an image captured by an imaging device, and more particularly, to an image clipping method at the time of aberration correction of the image processing device.

2. Description of the Related Art Conventionally, a technique for correcting optical distortion caused mainly by a lens called distortion is known for an image taken by an imaging apparatus. Further, when correcting distortion, the image after correction processing is roughly classified and transformed into one of two types of shapes called a pincushion type (see FIG. 9) and a barrel type (see FIG. 10). It is also known.
A digital image captured by the imaging device is generally rectangular. For this reason, for example, as in Patent Document 1 or Patent Document 2, a method of cutting out a rectangular region from an image after distortion correction processing has been studied. This method generally focused on cutting out the largest rectangular area possible from a deformed image after correction processing.
Incidentally, there is a fish-eye lens as one of the lenses that are generally used. Among the fish-eye lenses, particularly when using a lens called a circumferential fish-eye (or all-round fish-eye), the entire imaging surface such as an image sensor is not used as an imaging region, and has a radius from the center. The area within the circle is used as the imaging area. The area outside the circle is recorded as a black area that is an image but does not show the subject.
Even when correcting distortion of an image shot using a circumferential fisheye lens, it is necessary to correct the distortion and cut out the image thereafter. However, as in the conventional case, when the maximum rectangular area that can be cut out from the deformed image is cut out, a part of the circular shooting area may be cut out when the barrel is deformed. Yes (see FIG. 10). In a general lens other than the circumferential fisheye lens, since the entire imaging surface is a shooting area, there is no problem even in the process of cutting out the largest rectangular area inscribed in the barrel-shaped image. On the other hand, when the circumferential fisheye lens is used, the subject is reflected only in the circular area at the center of the imaging surface. For this reason, when the conventional method as described above is used for an image deformed into a barrel shape, there arises a problem that a part of a circular photographing region in which a subject is captured is cut off.

JP 2000-106623 A JP 2009-38646 A

  In view of the above situation, the problem to be solved by the present invention is that the shooting area is missing from the image after the distortion correction processing even when the shooting area is not the entire area of the image, such as when using a circumferential fisheye lens. Is to calculate a rectangular region that reduces the number of

In order to solve the above problems, an image processing apparatus according to the present invention includes a correction unit that performs distortion correction processing on an image, and one of a plurality of predetermined processing methods, the image capturing condition and the correction. And calculating means for calculating a size of a region to be cut out from the image after correction processing, which is selected according to the correction condition by the means.
The image processing method of the present invention includes a correction step for performing distortion correction processing on an image, and a step of calculating a region to be cut out from the image corrected in the correction step based on the image capturing condition and the correction condition in the correction step. It is characterized by having.
A program according to the present invention includes a correction step for performing distortion correction processing on an image on a computer, and a step of calculating a region to be cut out from the image corrected in the correction step according to the image capturing condition and the correction condition in the correction step. And making the computer execute.

  According to the present invention, it is possible to determine a rectangular region that includes a photographing region as much as possible from an image after correction processing, regardless of the lens used for photographing the image and the correction conditions. Therefore, it is possible to prevent or suppress the shooting area from being lost in the image of the clipped area.

FIG. 1 is a block diagram showing an example of the configuration of the image processing apparatus according to the first embodiment of the present invention. FIG. 2 is a flowchart showing a series of flows of determination of imaging conditions, distortion correction processing, and clipping processing. FIG. 3 is a flowchart showing the distortion correction processing in steps S202 and S203 of FIG. FIG. 4 is a flowchart showing a process for calculating the size of the cutout area of the image after the correction process. FIG. 5 is a flowchart showing the contents of sub-process A in FIG. FIG. 6 is a diagram schematically illustrating the relationship among the original image, the imaging region, and the cutout region. FIG. 7A is a diagram showing an example of a conversion table when a pincushion type deformation occurs, and FIG. 7B is a graph showing the conversion table shown in FIG. 7A. FIG. 8A is a diagram illustrating an example of a conversion table when barrel-shaped deformation occurs, and FIG. 8B is a graph of the conversion table of FIG. 8A. It is a figure which shows typically the example of a deformation | transformation of a pincushion type | mold. It is a figure which shows typically the example of a barrel-shaped deformation | transformation.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

(First embodiment)
First, a schematic configuration of an image processing apparatus 1 according to the first embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram showing an example of the configuration of the image processing apparatus 1 according to the first embodiment of the present invention. The image processing apparatus 1 includes an input unit 101, a correction unit 102, a calculation unit 103, and an output unit 104.
An image (image data) is input to the input unit 101 from the outside.
The correction unit 102 performs distortion correction processing on the image input from the input unit 101. For convenience of explanation, an image before distortion correction processing is performed by the correction unit 102 may be referred to as an “original image”, and a corrected image may be referred to as an “image after correction processing”.
The calculation unit 103 calculates the size of a rectangular area cut out from the image after correction processing. For convenience of explanation, a rectangular area cut out from an image after correction processing is referred to as a “cutout area”. In order to calculate the size of the cut-out area, a plurality of processing methods corresponding to shooting conditions are determined in advance. Then, the calculation unit 103 selects one processing method from the plurality of processing methods, and calculates the size of the cutout region using the selected one processing method. Furthermore, the calculation unit 103 uses the calculation result of the cutout region to cut out the cutout region from the corrected image. For convenience of explanation, the process of cutting out an image is referred to as “cutout process”.
The output unit 104 outputs an image of the area cut out by the calculation unit 103.

  Next, a processing flow according to the present embodiment will be described. In the present embodiment, photographing using a circumferential fisheye lens is given as an unusual photographing condition. FIG. 2 is a flowchart showing a series of flows of determination of imaging conditions, distortion correction processing, and clipping processing.

  In step S201, the calculation unit 103 checks the shooting conditions of the image (original image) input to the input unit 101, and determines whether this image is an image shot using a circumferential fisheye lens. If this image is an image taken using a circumferential fisheye lens, the process proceeds to step S202; otherwise, the process proceeds to step S203.

  In step S <b> 203, the correction unit 102 performs normal distortion correction processing on the image input to the input unit 101. Here, a conventional general distortion correction process of converting the pixel coordinates of the image before correction using the conversion table can be applied.

In step S <b> 202, the correction unit 102 performs distortion correction processing for an image captured using a circumferential fisheye lens (hereinafter referred to as “distortion aberration for circumferential fisheye” on the image input to the input unit 101. (Referred to as “correction processing”). Note that the content of the distortion correction processing for the circular fisheye may be the same as the normal distortion correction processing. However, special attention may be required in the distortion correction processing for the circumferential fisheye.
The specific is as follows. The circular exposure area (sometimes referred to as an image circle) of the circumferential fisheye lens is smaller than the rectangular imaging surface of the sensor of the imaging device (referred to as a surface that can convert the formed optical image into image data). For this reason, an image photographed using a circumferential fisheye lens has a circular area in which the subject appears in the center of the rectangular image. This area is referred to as a “photographing area”. The imaging area is an area where the exposure area of the lens is projected. The area outside the shooting area is a black area.
As described above, the image captured using the all-around fisheye lens does not have the entire imaging region, and the image includes a region that is not the imaging region.
Since the black area is usually outside the imaging area, there may be no conversion table by design. However, if no correction is performed, continuity within the image may not be maintained. In addition, there are cases where information that the user is not conscious of, such as normal noise, also exists in the black area. As described above, in the distortion correction processing, special attention is required for handling the black region.
Therefore, in the present embodiment, the distortion correction process for the circumferential fisheye is a process different from the normal distortion correction process. For example, the correction unit 102 virtually uses the correction amount of the conversion table used from the center of the imaging region to the outer edge of the circular region while maintaining the tendency.

  In step S204, the calculation unit 103 determines whether the image is deformed into a barrel shape by the distortion correction process. As described above, the image after the correction processing is deformed into a barrel shape or a pincushion shape depending on the distortion aberration correction processing conditions (correction conditions). If the correction condition is such that the image is deformed into a barrel shape, the process proceeds to step S206; otherwise, the process proceeds to step S205.

  In step S205, the calculation unit 103 selects, from a plurality of processing methods, a clipping process that cuts out the largest rectangular area that is inscribed in the corrected image. That is, a clipping process is selected in which a part of the image at the time of shooting is cut off. Then, the calculation unit 103 executes the clipping process using the selected processing method.

  In step S <b> 206, the calculation unit 103 selects a clipping process for cutting out a rectangular area so as to include the entire imaging area from a plurality of processing methods. Then, the calculation unit 103 executes the clipping process using the selected processing method.

  In step S <b> 208, the calculation unit 103 executes a process of blackening the area outside the shooting area of the cut image.

  In step S207, the output unit 104 outputs an image of the clipped area.

  As described above, a plurality of processing methods for the cutting process are determined in advance. The plurality of processing methods include at least the following two. (1) A processing method of cutting out the maximum rectangular area inscribed in the corrected image. (2) A processing method of cutting out a rectangular area so as to include the entire photographing area included in the corrected image. The processing method (2) further includes a plurality of processing methods according to the correction conditions. Then, the calculation unit 103 selects one of a plurality of processing methods according to the image capturing condition and the correction condition, and executes a clipping process using the selected processing method. A conventionally known processing method can be applied to the processing method (1). Details of the processing method (2) will be described later.

Next, distortion correction processing will be described with reference to FIG. FIG. 3 is a flowchart showing the distortion correction processing in steps S202 and S203 of FIG.
As described above, the distortion correction processing differs between the normal case and the case of the circumferential fisheye lens, but here, common processing will be described.

In step S301, the correction unit 102 selects the first pixel of the original image.
In step S302, the correction unit 102 calculates the distance from the center of the shooting area to the selected pixel.
In step S303, the correction unit 102 calculates the distance from the center of the imaging region after the correction process of the selected pixel.
In step S304, the correction unit 102 calculates coordinates after correction processing of the selected pixel from the calculated distance.
In step S305, the correction unit 102 records the calculated coordinates as coordinates after the correction processing of the selected pixel.
In step S306, the correction unit 102 determines whether the pixel on which the processing in steps S302 to S305 has been performed is the last pixel. If it is not the last pixel, the process proceeds to step S307, and if it is the last pixel, the process proceeds to step S308.
In step S307, the correction unit 102 selects the next pixel as a correction target pixel. Then, the process proceeds to step S302, and the processes after step S302 are repeated. As a result, the correction unit 102 executes the processes of steps S302 to S305 for all the pixels included in the original image.
In step S <b> 308, since the above processing has been completed for all pixels, the correction unit 102 outputs the corrected image to the calculation unit 103.

  Next, processing for calculating the size of the cutout area will be described with reference to FIGS. FIG. 4 is a flowchart showing a process for calculating the size of the cutout area of the image after the correction process. FIG. 5 is a flowchart showing the contents of sub-process A in FIG. FIG. 6 is a diagram schematically illustrating the relationship among the original image, the imaging region, and the cutout region. The calculation unit 103 calculates the size of the cutout area using the original image and the conversion table.

  In steps S <b> 401 to S <b> 406, the calculation unit 103 selects the upper left pixel, upper middle pixel, and left middle pixel of the original image, and executes the sub-process A for each of the selected pixels. Note that the selected pixel is an example, and another pixel may be selected. For example, pixels located at four corners other than the upper left of the original image may be used instead of the upper left pixel. Further, it may be a pixel at the middle point of the lower side instead of a pixel at the middle point of the upper side. Further, it may be a pixel at the midpoint of the right side instead of a pixel at the midpoint of the left side.

Here, the sub-process A will be described with reference to FIG.
In step S501, the calculation unit 103 calculates the distance from the center of the shooting area of each selected pixel.
In step S502, the calculation unit 103 uses the conversion table to calculate coordinates after correction processing of each selected pixel.
In step S <b> 503, the calculation unit 103 calculates the vertical and horizontal distances from the center of the imaging region after the correction processing of each selected pixel using the calculated coordinates.
In step S504, the calculation unit 103 returns the calculated distance between the vertical direction and the horizontal direction.
Then, the process proceeds to step S407 in FIG.

  In step S407, the calculation unit 103 determines whether the left side of the image is a short side. When the left side is a short side, the image is horizontally long. In this case, the process proceeds to step S408. On the other hand, when the left side is not a short side, the image is vertically long. In this case, the process proceeds to step S412.

In step S <b> 408, the calculation unit 103 sets a size that is twice the vertical distance from the center of the imaging region to the upper-side midpoint pixel after the correction processing as the vertical size of the cut-out region. This distance is calculated in sub-process A of step S404.
As shown in FIG. 10, in the conventional method, when the largest rectangular area P that is inscribed is cut out from the corrected image Q transformed into a barrel shape, the upper and lower portions S (hatching is performed on the circular imaging area R). Area) may be lost.
On the other hand, in the embodiment of the present invention, as shown in FIG. 6, the vertical size of the cutout area D is the same as or larger than the circular imaging area C. Accordingly, it is possible to prevent a defect from occurring in the upper and lower portions of the circular imaging region C.

In step S409, the calculation unit 103 compares the horizontal distance from the center of the imaging region with respect to the pixel on the upper left after the correction processing and the pixel at the midpoint on the left side after the correction processing. These distances are calculated in sub-process A in steps S402 and S406.
If the lateral distance from the center of the imaging region of the upper left pixel after the correction process is larger than the distance of the pixel at the midpoint of the left side after the correction process, the process proceeds to step S410. On the other hand, when the lateral distance from the center of the imaging region of the upper left pixel after the correction process is smaller than the distance of the pixel at the midpoint of the left side after the correction process, the process proceeds to step S411.

In step S410, the calculation unit 103 sets a size that is twice the horizontal distance from the center of the shooting area of the pixel at the left-side midpoint after the correction process to the horizontal size of the cut-out area.
In step S411, the calculation unit 103 sets a size that is twice the horizontal distance from the center of the imaging region of the upper left pixel after the correction process to the horizontal size of the cutout region.

On the other hand, if the left side is not a short side and the image is vertically long in step S407, a defect may occur only on the left and right of the circular shooting area. That is, FIG. 10 is rotated 90 degrees left or right. In this case, the process proceeds to step S412. The processing of steps S412 to S415 is processing of the content in which the vertical direction and the horizontal direction of the image are interchanged in the processing of steps S408 to S411. Specifically, it is as follows.
In step S412, the calculation unit 103 sets a size that is twice the horizontal distance from the center of the shooting area to the pixel at the midpoint of the left side after the correction process as the horizontal size of the cut-out area. As described above, when the image is vertically long, a defect may occur only on the left and right in the circular imaging region. Therefore, by setting the horizontal size of the cutout region as described above, it is possible to prevent a defect from being generated on the left and right of the circular imaging region.
In step S413, the calculation unit 103 compares the vertical distance from the center of the imaging region for the upper left pixel after the correction process and the pixel of the upper side midpoint after the correction process. If the vertical distance from the center of the shooting area of the upper left pixel after the correction process is greater than the distance of the upper middle point pixel after the correction process, the process proceeds to step S414. On the other hand, if the vertical distance from the center of the shooting area of the upper left pixel after the correction process is smaller than the distance of the pixel at the upper middle point after the correction process, the process proceeds to step S415.
In step S414, the calculation unit 103 sets a size that is twice the vertical distance from the center of the imaging region of the upper-side midpoint pixel after the correction processing to the vertical size of the cut-out region.
In step S411, the calculation unit 103 sets a size that is twice the vertical distance from the center of the imaging region of the upper left pixel after the correction process to the vertical size of the cutout region.

  In step S416, the calculation unit 103 cuts out the cutout region D from the image B after the correction processing with the set vertical and horizontal sizes. In FIG. 6, a hatched region E is a region that does not exist in the original image. Thus, the calculation unit 103 adds an area that does not exist in the original image to the cutout area D. Note that the area outside the shooting area C in the original image is black if the image is shot using a circumferential fisheye lens. Therefore, the calculation unit 103 performs a process of filling the added area E with black. Thereby, it is possible to prevent the cut image from becoming unnatural.

The following method can also be applied to calculate the size of the cutout area.
The calculation unit 103 acquires information on the sensor and lens used to capture the image. The information regarding the sensor includes information regarding the size of the imaging surface. The information about the lens includes the size of the shooting area (image circle). Note that the image processing apparatus 1 may be configured to hold information related to sensors of the imaging apparatus and information related to imaging areas of various lenses. In this case, the calculation unit 103 acquires information such as the model (model number) of the imaging device and the model (model number) of the lens used to capture the image, and reads information corresponding to the imaging device and the lens. Used for cutting out. When the lens used for imaging is a zoom lens whose magnification can be changed, the size of the exposure region may change depending on the magnification. Therefore, the calculation unit 103 also acquires information on the magnification of the lens.
Next, the calculation unit 103 determines the position and size of the shooting area included in the shot image (original image) from the size of the imaging surface of the sensor and the size of the exposure area of the lens.
Next, the calculation unit calculates the position of the outer edge of the shooting area in the image after the correction process, and further calculates a rectangular area including the entire shooting area in the image after the correction process based on the calculation result. The rectangular area calculated in this way becomes a cutout area.
Next, the calculation unit 103 cuts out the calculated cutout region from the image after the correction process.

Furthermore, the image processing apparatus 1 may use the processing shown in FIGS. 4 and 5 together with the processing using the above-described calculation result of the size of the imaging region.
For example, the calculation unit 103 calculates the size of the cutout region by the processing illustrated in FIGS. 4 and 5 and the processing using the above-described calculation result of the size of the imaging region. Then, the calculation unit 103 compares the size of the cut-out area calculated by the processes shown in FIGS. 4 and 5 with the size of the cut-out area calculated using the above-described shooting area size. Then, the calculation unit 103 determines the larger cutout size as the final cutout size. Thereafter, the calculation unit 103 performs a clipping process on the corrected image using the determined size of the clipping region.
Alternatively, as a result of the comparison, if the size of the cutout area calculated by the processing shown in FIGS. 4 and 5 is smaller than the size of the cutout area calculated using the above-described shooting area size, 4 and the size of the cutout area calculated by the processing shown in FIG. Next, the calculation unit 103 compares the sizes again. When the size of the cutout area calculated by the processing shown in FIGS. 4 and 5 is equal to or larger than the size of the cutout area calculated using the above-described shooting area size, the calculation unit 103 enlarges the cutout area. The size of the area is determined as the size of the final cut-out area.

Next, a conversion table used in the distortion correction process will be described with reference to FIGS.
FIG. 7A shows an example of a conversion table when a pincushion type deformation occurs. As shown in FIG. 7A, the conversion table has discrete projections of distances from the center of the imaging region as the size on the sensor. In practice, the correction unit 102 interpolates and uses the values defined in this table. In order to increase the accuracy of correction, it is preferable that the conversion table has more projections. FIG. 7B is a graph showing the conversion table shown in FIG. As shown in FIG. 7B, the amount of change increases as the area approaches the periphery of the imaging region. This is a feature of the conversion table when the image is transformed into a pincushion type by correction.
FIG. 8A shows an example of a conversion table when barrel-shaped deformation occurs. FIG. 8B is a graph showing the conversion table of FIG. As shown in FIG. 8B, the conversion table in the case of barrel-shaped deformation has a tendency for the amount of change to decrease as it approaches the periphery of the imaging region.
As described above, image deformation by distortion correction processing is roughly classified into two types: pincushion type and barrel type. For this reason, the conversion table can also be broadly classified into the above two types.

  As described above, according to the embodiment of the present invention, the imaging region is as much as possible regardless of whether or not distortion correction is performed on an image captured using a circumferential fisheye lens. The size of the cutout area can be calculated so as to be included in the cutout area.

(Second Embodiment)
In the first embodiment, the processing for an image photographed using the circumferential fisheye lens has been described as an example, but the present invention is not limited to such a configuration. The present invention can also be applied to, for example, an image captured by mounting an imaging device lens having a sensor different from the predetermined size on an imaging device having a sensor of a predetermined size.
For example, an image pickup apparatus having a sensor having an image pickup surface size of 36 mm × 24 mm (hereinafter referred to as “36 mm sensor”) may be photographed with a sensor lens having a smaller size attached thereto. In such a combination, the exposure area of the lens may be smaller than the imaging surface of the 36 mm sensor. In this case, if the size of the exposure area of the lens and the size of the imaging surface of the sensor are known in advance, the calculation unit 103 can calculate the size of the cutout area. For example, in order to expose the entire area of the 36 mm sensor, a lens having a circular area with a radius of 21.635 mm or more is necessary. On the other hand, in a lens for a sensor having an imaging surface size of 24 mm × 16 mm (hereinafter referred to as “24 mm sensor”), it can be assumed that the radius of the exposure region is 14.42 mm.
Therefore, the calculation unit 103 performs the following process when it can be determined from the shooting conditions that the lens for the 24 mm sensor is attached to the imaging apparatus having the 36 mm sensor. That is, the calculation unit 103 determines that a region inside the circle having a radius of 14.42 mm or a rectangular region inscribed in the circle is a photographing region on the photographing surface of the sensor. Then, the calculation unit 103 can calculate a region inside this circle or a rectangular region inscribed in this circle as a cutout region by the same processing as in the first embodiment. As described above, although different from the example of the circumferential fisheye lens shown in the first embodiment, the image processing apparatus 1 can switch the content of the calculation process even when the sensor size is different from that of the lens.
These are uniquely determined when the shooting conditions are determined, but a plurality of processing methods (algorithms for calculating the size of the cutout area) should be selectively used according to the shooting conditions.

Here, a hardware configuration of the image processing apparatus 1 according to the embodiment of the present invention will be briefly described. An image processing apparatus 1 according to an embodiment of the present invention is a computer having a CPU, a RAM, a ROM, and an interface for transmitting and receiving data to and from the outside. Furthermore, the image processing apparatus 1 according to the embodiment of the present invention may include a storage device that can store an image input from the outside or an image that has been subjected to correction processing, and an interface that can connect such a storage device. Good. As the storage device, for example, an HDD or an SSD can be applied. A computer program (computer software) for executing the distortion correction process and the cutting process is stored in the ROM in advance. Further, the ROM stores a computer program for executing a process for inputting an image from the outside and a program for executing a process for outputting an image of the clipped area to the outside.
The CPU reads a computer program stored in the ROM and executes it using the RAM as a work area. As a result, the computer functions as each unit and the processes are executed.
The conversion table used for the distortion correction process is stored in the ROM as a computer-readable table. The CPU reads this conversion table from the ROM and uses it in the distortion correction process.
Similarly, a plurality of processing methods for the cutting process are stored in advance in the ROM as computer programs for executing the plurality of processes. Various information necessary for distortion correction processing is also stored in the ROM as computer-readable data. Then, the CPU selects, reads out and executes one of a plurality of processing methods according to the image capturing condition and the correction condition. Thereby, a plurality of processing methods are selectively executed.

  Further, the image processing apparatus 1 according to the embodiment of the present invention may be configured to be incorporated in an imaging apparatus having a sensor and a recording medium capable of storing a captured image. For example, an image processing apparatus according to an embodiment of the present invention can be configured to be incorporated in a digital single lens reflex camera.

  Further, the present invention includes a configuration in which a program that realizes the functions of the above-described embodiments is supplied to a system or apparatus having a computer that can execute the program directly from a recording medium or using wired / wireless communication and executed. It is. Accordingly, the program code itself supplied and installed in the computer in order to implement the functional processing of the present invention by the computer also realizes the present invention. That is, the program itself for realizing the functional processing of the present invention is also included in the present invention. In this case, the program may be in any form as long as it has a program function, such as an object code, a program executed by an interpreter, or script data supplied to the OS. As a recording medium for supplying the program, for example, a magnetic recording medium such as a flexible disk, a hard disk, a magnetic tape, MO, CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-R, DVD- RW etc. are applicable. Further, an optical / magneto-optical storage medium and a non-volatile semiconductor memory can also be applied.

  As mentioned above, although embodiment (Example) of this invention was described in detail, the said embodiment (Example) only showed the specific example in implementing this invention. The technical scope of the present invention is not limited to the above-described embodiment (example). The present invention can be variously modified without departing from the gist thereof, and these are also included in the technical scope of the present invention.

  The present invention relates to an image processing apparatus, an image processing method, and a program that can cut out a cutout region from an image deformed by correction. According to the present invention, it is possible to calculate a rectangular area that includes the shooting area as much as possible from the image after the correction process, regardless of the lens used for shooting and the type of geometric transformation.

1: Image processing apparatus 101: Input unit 102: Correction unit 103: Calculation unit 104: Output unit

Claims (11)

Correction means for performing distortion correction processing on the image;
Calculating means for selecting one of a plurality of predetermined processing methods according to the photographing condition of the image and a correction condition by the correction means, and calculating a size of a region to be cut out from the image after the correction processing;
An image processing apparatus comprising:   2. The image processing according to claim 1, wherein, when the image is an image shot with a fisheye lens, the calculation unit calculates the area to be cut out by a different calculation method from an image shot with a lens that is not a fisheye lens. apparatus.   The calculation unit according to claim 1, wherein when the image includes a region that is not a shooting region, the calculation unit calculates the size of the cut-out region by a method different from that when the entire image is a shooting region. The image processing apparatus described.   In the case where the calculation means is an image shot with a fisheye lens, the image after the distortion correction processing is included in the image after the distortion correction processing is included. The image processing apparatus according to claim 2, wherein the size of the cutout area is calculated.   The calculation means is an image taken with a fisheye lens, and when the image after the distortion aberration correction processing is barrel-shaped, in the image after the distortion aberration correction processing, The image processing apparatus according to claim 2, wherein the size of the cut-out area is calculated so that the entire shooting area before the distortion aberration correction process is performed is included.   When the calculation means is not an image photographed with a fisheye lens, the image after the distortion correction processing is not included in a part of the photographing region before the distortion correction processing is performed. The image processing apparatus according to claim 4, wherein the size of the cutout area is calculated.   When the image includes a region that is not a shooting region, the calculation unit includes all of the shooting region before the distortion correction processing is performed in the image after the distortion correction processing is performed. The image processing apparatus according to claim 3, wherein the size of the cutout area is calculated as described above.   When the image includes a region that is not a shooting region and the image after the distortion correction processing is barrel-shaped, the calculation means The image processing apparatus according to claim 3, wherein the size of the cutout area is calculated so that all of the imaging area before the distortion aberration correction process is performed is included.   When the image does not include a region that is not a shooting region, the calculation unit includes a portion of the shooting region before the distortion aberration correction processing is performed in the image after the distortion aberration correction processing is performed. The image processing apparatus according to claim 7, wherein the size of the cutout area is calculated so as not to be included. A correction step for performing distortion correction processing on the image;
Calculating a region to be cut out from the image corrected in the correction step according to the imaging condition of the image and the correction condition in the correction step;
An image processing method comprising: On the computer,
A correction step for performing distortion correction processing on the image;
Calculating a region to be cut out from the image corrected in the correction step according to the imaging condition of the image and the correction condition in the correction step;
A program characterized by having executed.

Images