JP2013192828A - Apparatus and method for detecting uneven brightness - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately detect uneven brightness between images when connecting and synthesizing a plurality of images.SOLUTION: An overlapping part extracting section 12 extracts overlapping parts between two radiation images acquired by long-length photography. An offset correcting section 13 calculates the offset amounts r of the overlapping parts Cup and Cdown of the two radiation images, and an uneven brightness calculating section 14 calculates an uneven vector K(y) representing uneven brightness. An image processing section 15 carries out image processing to match brightness of the two radiation images on the basis of regions of parts without uneven brightness acquired from the uneven vector K(y). A synthesizing section 16 synthesizes the radiation images so as to connect the overlapping parts to acquire a long-sized radiation image GL.

Description

  The present invention provides a luminance unevenness for detecting luminance unevenness included in a plurality of images when a long image is obtained by combining a plurality of images obtained by panoramic photography or long photography in the medical field, for example. The present invention relates to a detection apparatus and method.

  2. Description of the Related Art Conventionally, in radiography in the medical field or the like, there is a case where long imaging is performed on a long region such as the entire spine (whole spine) or the entire foot (whole leg). For radiation imaging, various radiation detectors (so-called “Flat Panel Detectors”) that record radiation images related to the subject by irradiation of the radiation that has passed through the subject have been proposed and put to practical use. There may be a narrower range that can be taken. For this reason, in order to perform long imaging, the radiation detector is moved so as to partially overlap along a predetermined movement axis, and each time the position is changed, radiation is transmitted through the same subject. To. A reading operation is performed from the radiation detector for each radiation irradiation (recording of a radiation image), and a radiation image is acquired for each reading operation. Then, by combining the radiographic images so as to be joined later, a long radiographic image showing a long portion of the subject can be obtained.

  By the way, when performing such a long photographing, for example, when photographing all the lower limbs, the thigh, calf, and toes are photographed sequentially from the abdomen and pelvis. Therefore, the optimum values of imaging conditions such as tube voltage and tube current differ depending on the region to be imaged. However, if each part is photographed under the optimum photographing conditions for each part, there is a difference in brightness for each of a plurality of consecutively photographed images. When these images are combined with a long image, the brightness between the images is reduced. Due to the difference, an appropriate diagnosis may not be performed. For this reason, in many cases, each part is imaged under the same imaging conditions, and a plurality of images are acquired. This is a problem that occurs even when a panoramic image is taken using a digital camera, for example, and a plurality of images are acquired and combined to generate a panoramic image.

  Therefore, based on the overlap between images acquired by long shooting, a method of combining after matching the gradation between images, so that the index values of the overlap of multiple images acquired by long shooting match A method for performing gradation correction, a process for obtaining the minimum value of the output value of a pixel is performed for all the pixels in the overlapping portion of the image, and a long image is created using the obtained minimum value as the output value of the pixel, thereby producing unevenness. Various methods such as a method for obtaining a long image with little image are proposed (see Patent Documents 1 to 3). In addition, a method for reducing a density difference between partial images when combining a plurality of digital camera images has been proposed (see Patent Document 4).

JP 2011-110247 A JP 2004-057506 A JP 2005-46444 A Japanese Patent Laid-Open No. 11-055558

  On the other hand, when capturing a radiographic image, an irradiation field stop (collimator) may be used in order to prevent radiation from being applied to an unnecessary part of the subject. When a subject is photographed using the irradiation field stop, the radiation image includes an irradiation field inner region including image information of the subject and an irradiation field outer region not including the image information. As a result, the radiation image has an irradiation field area and an irradiation field area. However, due to the vignetting of the radiation by the irradiation field stop, white stripes are formed from the irradiation field area to the boundary between the irradiation field area and the irradiation field area. Brightness unevenness occurs. Further, due to the radiation heel effect, luminance unevenness such as shading in which the luminance changes in a gradation shape also occurs. When the above-described long imaging is performed, the irradiation field region and the boundary part between the irradiation field region and the irradiation field region occur in overlapping portions of the radiation images. For this reason, as described in Patent Documents 1 to 3, when a method of correcting a difference in density between images using gradations of overlapping portions is used, luminance unevenness caused by vignetting, shading, or the like is caused. Due to the presence, there is a possibility that the difference in density between images cannot be corrected with high accuracy. This is a problem that occurs not only in the case of a radiographic image but also when a panoramic image is created using a plurality of images acquired by performing panoramic photography using a digital camera.

  The present invention has been made in view of the above circumstances, and an object thereof is to accurately detect luminance unevenness between images when a plurality of images acquired by long shooting or panoramic shooting are connected and combined.

The brightness unevenness detection apparatus according to the present invention includes an image acquisition means for acquiring a plurality of images obtained by photographing the same subject a plurality of times with different exposure amounts, and at least some of the regions overlap.
Correction means for correcting the difference in brightness in the overlapping portion of the plurality of images;
It is characterized by comprising unevenness calculation means for calculating luminance unevenness in the overlapped portion based on the absolute value of the difference value of the corresponding position in the overlapped portion after correction between a plurality of images.

  In the luminance unevenness detection apparatus according to the present invention, the correction unit may be a unit that calculates an offset amount indicating a luminance difference of an overlapping portion between a plurality of images and corrects the luminance difference based on the offset amount. .

  In this case, the correction unit may be a unit that calculates the offset amount based on the representative value of the pixel value in the overlapping portion.

  In the luminance unevenness detection apparatus according to the present invention, the luminance unevenness may be one-dimensional information about the direction in which a plurality of images are arranged when a plurality of images are combined in an overlapping portion.

Further, in the luminance unevenness detection apparatus according to the present invention, based on image information of a portion other than the luminance unevenness in the overlapping portion, image processing means for performing image processing for matching the luminance between a plurality of images,
Combining means for generating a long image by combining a plurality of images after image processing may be further provided.

The luminance unevenness detection method according to the present invention obtains a plurality of images obtained by photographing the same subject a plurality of times with different exposure amounts, and at least some of the regions overlap.
In the overlapping part of multiple images, correct the difference in brightness,
Among the plurality of images, the luminance unevenness in the overlapping portion is calculated based on the absolute value of the difference value of the corresponding position in the overlapping portion after correction.

  According to the present invention, the luminance difference is corrected in the overlapping portion of the plurality of images, and the luminance unevenness in the overlapping portion is calculated based on the absolute value of the difference value of the corresponding position in the corrected overlapping portion. Is. For this reason, the brightness | luminance nonuniformity which exists in an overlap part can be detected accurately.

  Also, based on the image information of the part other than the luminance unevenness in the overlapping portion, image processing for matching the luminance between the plurality of images is performed, and the long image is synthesized by combining the plurality of images after the image processing. By generating, the luminance of a plurality of images can be matched without being affected by luminance unevenness, so that a high-quality long image or panoramic image can be generated. In particular, when the acquired image is a medical image, an appropriate diagnosis can be performed using a long image.

Schematic which shows the structure of the radiographic imaging system to which the brightness nonuniformity detection apparatus by the 1st Embodiment of this invention is applied. The figure for demonstrating the movement of the radiation detector at the time of long imaging | photography of a to-be-photographed object Schematic block diagram showing the configuration of the image processing apparatus Figure showing an example of a radiographic image Diagram showing overlapping parts of radiographic images The figure which shows the line average value of the overlapping part The figure which shows the value which added the line average Eup (y) and Edown (y) Diagram showing the value of the unevenness vector The figure which shows the histograms Hup and Hdown of the overlapping part Cup and Cdown The flowchart which shows the process performed in 1st Embodiment. The flowchart which shows the process performed in 2nd Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a schematic diagram showing a configuration of a radiographic image capturing system to which a luminance unevenness detecting apparatus according to a first embodiment of the present invention is applied. As shown in FIG. 1, a radiographic imaging system 1 according to the first embodiment includes a radiation source 2 that irradiates a subject S on an imaging table 5 with radiation, and a collimator that limits an irradiation range of radiation emitted from the radiation source 2. 3, a radiation detector 4 that detects radiation transmitted through the subject S, and an image processing device 10 that performs various processes on the radiation image acquired by the radiation detector 4.

  In the present embodiment, the collimator 3 limits the irradiation range of the radiation emitted from the radiation source 2 so as to be a rectangular irradiation region having sides in the body axis direction of the subject S and the direction perpendicular thereto. In the present embodiment, a coordinate system is defined in which the direction that coincides with the moving direction of the radiation detector 4 in the radiographic image is the y direction, and the direction orthogonal thereto is the x direction.

  The radiation source 2, the collimator 3, and the radiation detector 4 are configured to move in synchronization with the body axis direction of the subject S by a moving mechanism (not shown). As a result, the radiographic image capturing system 1 sequentially captures a plurality of adjacent areas in the subject S, combines them so that the plurality of radiographic images acquired thereby are joined together, and is a long image that indicates most of the subject S. It is comprised so that long imaging | photography which acquires the radiographic image of can be performed.

  Alternatively, only the radiation detector 4 may be moved in the body axis direction of the subject S, and the radiation source 2 and the collimator 3 may be swung so as to irradiate the radiation toward the moved radiation detector 4. Good.

  Here, the radiation detector 4 detects the radiation transmitted through the subject S, converts it into an electrical signal, and outputs image data representing the radiation image of the subject S. The radiation detector 4 uses either a direct type detector that directly converts radiation into electric charges, or an indirect type detector that converts radiation once into light and further converts the converted light into electric signals. Is possible.

  FIG. 2 is a diagram for explaining the movement of the radiation detector 4 when the subject S is photographed long. As shown in FIG. 2, the radiation detector 4 moves to positions P1, P2, and P3 so that regions near the edge of the imaging region overlap each other. In the radiographic imaging system 1, imaging is performed by irradiating the subject S with radiation from the radiation source 2 at each of the positions P1 to P3. And a long radiographic image is acquirable by connecting a some radiographic image so that the overlapping part of the some radiographic image acquired by this may correspond.

  When such long shooting is performed, the body thickness varies depending on the portion of the subject S. Specifically, the body thickness of the abdomen and pelvis is relatively large, and the body thickness gradually decreases from the thigh and calf toward the toes. Also, the body thickness differs between the chest and abdomen. For this reason, it is necessary to change the radiation dose by changing the tube current corresponding to these imaging regions, but the luminance differs between the radiographic images due to the change of the radiation dose. In order to correct such a difference in luminance between radiographic images, image processing such as gradation conversion is performed in the image processing apparatus 10.

  The image processing apparatus 10 is a computer having a high-definition liquid crystal display that displays images and the like, a keyboard and a mouse that accept input from a user, and a main body that includes a CPU, a memory, a hard disk, a communication interface, and the like. Various functions are performed on a plurality of radiographic images acquired by long imaging, and a function of acquiring a long radiographic image by combining the plurality of radiographic images is provided.

  FIG. 3 is a schematic block diagram showing the configuration of the image processing apparatus 10. As shown in FIG. 2, the image processing apparatus 10 includes an image acquisition unit 11, an overlapping part extraction unit 12, an offset correction unit 13, a luminance unevenness calculation unit 14, an image processing unit 15, and a synthesis unit 16.

  The image acquisition unit 11 acquires, as digital data, a plurality of radiation images Gi (i = 1 to n: n is the number of images taken by long imaging) detected by the radiation detector 2 by long imaging. FIG. 4 is a diagram illustrating an example of a radiographic image. Here, it is assumed that three radiation images G1 to G3 of the subject S have been acquired. As shown in FIG. 4, the radiographic images G <b> 1 to G <b> 3 include image information of the entire spinal column and all the lower limbs of the subject S in which regions within a predetermined range from the edge overlap each other.

  The overlapping part extracting unit 12 extracts overlapping parts (hereinafter referred to as overlapping parts) in the radiation image Gi from each of the radiation images Gi. In the radiographic image shown in FIG. 4, the overlapping part C1 is extracted from the radiographic image G1, the overlapping parts C2-1 and C2-2 are extracted from both ends of the radiographic image G2, and the overlapping part C3 is extracted from the radiographic image G3. . FIG. 5 is a diagram illustrating the overlapping portions C1 and C2-1 of the radiographic image. As shown in FIG. 5, the overlapping portions C1 and C2-1 include image information of the same part, but due to the difference in the radiation dose at the time of imaging, the luminance is different even in the same part. . In FIG. 5, the difference in luminance is indicated by the difference in hatching density. Further, the edge of the overlapping portions C1 and C2-1 (that is, the edge of the original radiation image) includes white stripe-like luminance unevenness M0 extending in the x direction when the radiation is applied to the collimator 3. Yes.

The offset correction unit 13 corrects the luminance difference in the overlapping portion of the radiation image Gi. Hereinafter, processing performed by the offset correction unit 13 will be described. The offset correction unit 13 first calculates the line average of pixels extending in the x direction of two overlapping portions extracted from two radiographic images acquired at adjacent imaging positions by the following formulas (1) and (2). Are calculated respectively.

In Expressions (1) and (2), I (x, y) is the luminance value of each pixel in the overlapped portion, E (y) is the line average of the luminance values, and the subscript up is a combination of the radiation images Gi. In this case, the upper radiographic image and the subscript down represent the lower radiographic image. In addition, one edge coordinate value in the x direction in the overlapping portion is x = 0, and W-1 is the other edge coordinate value (W is the total number of pixels). The line average value is shown in FIG. As shown in FIG. 6, the line averages Eup (y) and Edown (y) are one-dimensional information about the y direction. In FIG. 6, the origin of the coordinates is the upper left corner of the overlapping portions Cup and Cdown. H is the number of pixels in the y direction of the overlapping portions Cup and Cdown.

  Since the upper overlapping portion Cup includes a luminance unevenness M0 at the lower end and the lower overlapping portion Cdown includes an upper end, as shown in FIG. 6, the value of the line average Eup of the upper overlapping portion Cup is y = H, i.e., higher in the vicinity of the edge, and the value of the line average Edown of the lower overlapping portion Cdown is higher in y = 0, that is, in the vicinity of the edge.

Next, the offset correction unit 13 adds the line averages Eup (y) and Edown (y) of the corresponding y coordinates in the overlapping portions Cup and Cdown as shown in the following formula (3), and the added value is the minimum. The value of y is calculated as ybase. This calculation is to obtain the coordinates with the largest radiation dose in the overlapping portion. FIG. 7 shows the value of Eup (y) + Edown (y).

Further, the offset correction unit 13 calculates a difference value between Edown (ybase) and Eup (ybase) as an offset amount r of the overlapping portions Cup and Cdown as shown in the following equation (4). As shown in FIG. 6, the offset amount r represents the difference between the line averages Eup (y) and Edown (y) of the overlapping portion Cup and the overlapping portion Cdown.

Then, the offset correction unit 13 performs offset correction of the line averages Eup (y) and Edown (y) of the overlapping portions Cup and Cdown using the offset amount r as shown in the following equation (5). As a result, the difference in luminance between the overlapping portion Cup and the overlapping portion Cdown based on the difference in radiation dose during imaging is corrected. Equation (5) performs offset correction so that the value of Edown (y) matches Eup (y). Note that when the offset correction is performed according to the equation (5), the value of the line average Eup (y) is not changed, but the value of the line average Eup (y) is also changed depending on the offset correction method. The line average Eup (y) to be used later is represented as E′up (y).

As shown in the following equation (6), the luminance unevenness calculation unit 14 calculates the absolute value K of the difference value between the line averages E′up (y) and E′down (y) of the overlapping portions Cup and Cdown after the offset correction. (Y) is calculated as a non-uniformity vector representing luminance non-uniformity in the y direction at the overlapping portion Cup and the overlapping portion Cdown. FIG. 8 shows the value of the unevenness vector K (y). As shown in FIG. 8, the value of the unevenness vector K (y) is one-dimensional information about the y direction. The value becomes large at the position where the luminance unevenness M0 exists in the overlapping portions Cup and Cdown, and becomes substantially zero at other positions.

  The image processing unit 15 uses the luminance value of the pixel in the region having the coordinate value in the y direction where the value of the unevenness vector K (y) is equal to or less than the threshold value Th1 in the overlapping portions Cup and Cdown, and uses the radiation image Gup , Gdown brightness conversion processing is performed. Specifically, as shown in the range of the arrow A in FIG. 8, the pixels in the region where the y coordinate is such that the value of the unevenness vector K (y) in the overlapping portion Cup and the overlapping portion Cdown is equal to or less than the threshold value Th1. Luminance value histograms Hup and Hdown are calculated. FIG. 9 is a diagram showing histograms Hup and Hdown of overlapping portions Cup and Cdown. As shown in FIG. 9, the histograms Hup and Hdown have substantially the same shape because the image information of the same part of the subject S is included, but the histogram Hdown is shifted by ΔE toward the high luminance side.

  For this reason, the image processing unit 15 performs a process of matching the luminance values Iup (x, y) and Idown (x, y) of the radiation images Gup and Gdown based on ΔE. Specifically, an operation such as adding ΔE to the luminance value Iup (x, y) of each pixel of the radiation image Gup or subtracting ΔE from the luminance value Idown (x, y) of each pixel of the radiation image Gdown. Thus, the luminance values Iup (x, y) and Idown (x, y) of the respective pixels of the radiation images Gup and Gdown are matched.

  The synthesizing unit 16 synthesizes the overlapping portions of the radiographic images Gi whose luminance values have been corrected so as to join together to generate a long radiographic image GL.

  Next, processing performed in the first embodiment will be described. FIG. 10 is a flowchart showing processing performed in the first embodiment. It is assumed that a plurality of radiation images Gi have already been acquired and taken into the image processing apparatus 10. First, the image processing apparatus 10 selects a set of radiographic images composed of two radiographic images to be processed first (step ST1), and the overlapping portion extraction unit 12 selects two radiographic images included in the selected set. An overlapping part is extracted (step ST2). Then, the offset correction unit 13 calculates the offset amount r of the overlapping portions Cup and Cdown in the two radiographic images (step ST3), and the luminance unevenness calculation unit 14 calculates the unevenness vector K (y) (step ST4). .

  Then, the image processing unit 15 performs image processing for matching the luminances of the two radiographic images based on the unevenness vector K (y) (step ST5). Then, the image processing apparatus 10 determines whether or not processing has been performed for all sets of radiographic images (step ST6). If step ST6 is negative, the processing target is changed to the next set of radiographic images. (Step ST7), the process returns to Step ST2. The radiographic image set is selected by selecting one position where the radiographic image is acquired, such as radiographic images G2 and G3 next to the radiographic images G1 and G2, and radiographic images G3 and G4 next to the radiographic images G2 and G3. It is assumed that the set of radiation images is changed by shifting each step. At this time, the image processing unit 15 is based on the brightness value of the area other than the brightness unevenness obtained from the unevenness vector K (y) acquired in each set so that the brightness of all the sets of radiographic images matches. Image processing.

  If step ST6 is affirmed, the combining unit 16 combines the overlapping portions of the radiation images Gi to join together to obtain a long radiation image GL (step ST8), and ends the process.

  As described above, in this embodiment, the difference in luminance is corrected by the offset amount r in the overlapping portion of the plurality of radiation images Gi, and the line averages E′up (y), E′down ( Based on the absolute value of the difference value of y), the unevenness vector K (y) representing the brightness unevenness in the overlapping portion is calculated. For this reason, the brightness | luminance nonuniformity which exists in an overlap part can be detected accurately.

  Also, based on the brightness value of the area other than the brightness unevenness in the overlapping part, image processing for matching the brightness between the radiographic images is performed, and a plurality of radiographic images after the image processing are combined and combined to be long. Since the radiographic image GL is generated, the luminance of the plurality of radiographic images can be matched without being affected by the luminance unevenness, whereby a high-quality long image can be generated. In particular, in the case of a medical image, an appropriate diagnosis can be performed using a long radiation image GL.

  In the first embodiment, the processing in the case of using three or more radiographic images has been described. However, the present invention is also applied to the case where a long image is generated using only two radiographic images. be able to. Hereinafter, this will be described as a second embodiment. FIG. 11 is a flowchart showing processing performed in the second embodiment.

  First, the overlapping part extraction unit 12 extracts an overlapping part of two radiographic images included in the selected set (step ST11). Then, the offset correction unit 13 calculates the offset amounts r of the overlapping portions Cup and Cdown in the two radiographic images (step ST12), and the luminance unevenness calculation unit 14 calculates the unevenness vector K (y) (step ST13). .

  Then, the image processing unit 15 performs image processing for matching the luminances of the two radiation images based on the unevenness vector K (y) (step ST14). Next, the synthesizing unit 16 synthesizes the overlapping portions of the two radiographic images together to obtain a long radiographic image GL (step ST15), and ends the process.

In the first and second embodiments, the offset amount r is calculated using the line average of the overlapping portions Cup and Cdown. However, as shown in the following equation (7), the overlapping portions Cup, The offset amount r may be calculated using the Cdown surface average.

In the first and second embodiments, the offset amount r is calculated using the line average or the surface average of the overlapping portions Cup and Cdown. However, the overlapping portion Cup, Any representative value can be used as long as it represents a Cdown line or surface. For example, not only the line average or the surface average but also the offset amount r is calculated using the luminance value at an arbitrary predetermined pixel position (x0, y0) as a representative value as shown in the following equation (8). You may do it.

  In each of the above embodiments, the luminance unevenness K (y) is one-dimensional information, but may be two-dimensional information. In this case, the luminance unevenness K (x, y) consisting of two-dimensional information may be calculated by subtracting the offset amount r from each pixel in the overlapping portion.

  In each of the above embodiments, streaky luminance unevenness is removed. However, due to the radiation heel effect, luminance unevenness such as shading in which the luminance changes in a gradation shape also occurs in the radiation image. The luminance unevenness due to the heel effect is in a state where the streaky unevenness is blurred. For this reason, as for the luminance unevenness due to the heel effect, as in the first and second embodiments, the offset amount and the unevenness vector are calculated, and image processing for matching the brightness of the two radiation images based on the unevenness vector is performed. It is possible.

  Further, in each of the above embodiments, the processing in the case of creating a long radiographic image from a plurality of radiographic images that have undergone long radiography has been described, but also when performing panoramic radiography using a digital camera, Photographing is performed so that a part of a subject such as a landscape overlaps, and a plurality of images partially overlapping are acquired. Here, when a square lens hood that does not match the size is attached to the digital camera, or when the orientation of the flower lens hood is attached incorrectly, streaky luminance unevenness occurs at the edge of the acquired image. To do. Even when the flash application range is narrower than the angle of view, streaky luminance unevenness may occur at the edge of the image. For this reason, as in the first and second embodiments, the offset correction amount r of the overlapping portion is calculated for a plurality of images that are partially overlapped by the digital camera, and luminance unevenness is extracted. By performing image processing so that the luminance values of a plurality of images are matched, unevenness between the images can be accurately corrected and a high-quality panoramic image can be acquired, as in the case of a radiographic image.

  Note that the ybase calculated by the offset correction unit 13 described above corresponds to the coordinates with the largest radiation dose in the overlapping portion, but in the case of an image acquired by a digital camera, the ybase corresponds to the coordinates with the largest amount of light. Here, in the radiographic image, the luminance value decreases as the radiation dose increases. Conversely, in the image acquired by the digital camera, the luminance value increases as the light amount increases. For this reason, in the image acquired by the digital camera, the y value that maximizes the value obtained by adding the line averages Eup (y) and Edown (y) of the corresponding y coordinates in the overlapping portion is calculated as ybase.

DESCRIPTION OF SYMBOLS 1 Radiographic imaging system 2 Radiation source 3 Collimator 4 Radiation detector 10 Image processing apparatus 11 Image acquisition part 12 Duplicate part extraction part 13 Offset correction part 14 Brightness nonuniformity calculation part 15 Image processing part 16 Synthesis | combination part

Claims (6)

Image acquisition means for acquiring a plurality of images obtained by photographing the same subject a plurality of times with different exposure amounts, at least a part of which overlaps;
Correction means for correcting a difference in luminance in the overlapping portion of the plurality of images;
A luminance calculating unit that calculates luminance unevenness in the overlapping portion based on an absolute value of a difference value of a corresponding position in the overlapping portion after the correction between the plurality of images; Unevenness detection device.   2. The correction unit according to claim 1, wherein the correction unit is a unit that calculates an offset amount representing a difference in luminance of the overlapping portion between the plurality of images and corrects the difference in luminance based on the offset amount. Brightness unevenness detection device.   The luminance unevenness detection apparatus according to claim 2, wherein the correction unit is a unit that calculates the offset amount based on a representative value of pixel values in the overlapping portion.   The brightness unevenness detection apparatus according to claim 2 or 3, wherein the brightness unevenness is one-dimensional information about a direction in which the plurality of images are arranged when the plurality of images are combined in the overlapping portion. . Image processing means for performing image processing for matching the luminance between the plurality of images based on image information of a portion other than the luminance unevenness in the overlapping portion;
The brightness nonuniformity detection apparatus according to claim 1, further comprising: a combining unit that combines the plurality of images after the image processing so as to join together to generate a long image. . Acquire a plurality of images obtained by photographing the same subject multiple times with different exposure amounts, at least part of which overlaps,
In the overlapping part of the plurality of images, correct the difference in brightness,
A luminance unevenness detection method, comprising: calculating luminance unevenness in the overlapping portion based on an absolute value of a difference value of corresponding positions in the overlapped portion after the correction between the plurality of images.