Image processing device, method, and computer-readable medium for executing pixel value correction in a synthesized image

1. An image processing device comprising:

an image correction unit which receives a synthesized image generated by synthesizing a plurality of images and pseudo color pixel information, the pseudo color pixel information indicating a pseudo color pixel position included in the synthesized image, and executes pixel value correction using a pseudo color pixel in the synthesized image as a pixel to be corrected,

wherein the image correction unit:

selects non-pseudo color pixels, which are not determined as a pseudo color based on the pseudo color pixel information, as reference pixels, from a reference region set in a region surrounding the pixel to be corrected,

sets a similarity weight coefficient in which a large weight is set as a pixel value of the reference pixel that is close to a pixel value of the pixel to be corrected, and a pixel position weight coefficient in which a large weight is set as the pixel value of the reference pixel that is near a pixel position of the pixel to be corrected, with respect to the reference pixels, and

determines a correction pixel value of the pixel to be corrected by a weighted summation process of summing values obtained by multiplying the pixel values of the reference pixels by the similarity weight coefficient and the pixel position weight coefficient.

2. The image processing device according to claim 1 , wherein the image correction unit calculates the similarity weight coefficient according to a difference in pixel value configuration information of the pixels of the synthesized image.

3. The image processing device according to claim 2 , wherein:

pixel value information corresponding to a Lab color space is set in each of the pixels of the synthesized image, and

the image correction unit calculates the similarity weight coefficient according to the difference in each of luminance L and a color difference ab, both of which are pixel value configuration information of the pixels of the synthesized image.

4. The image processing device according to any one of claims 1 to 3 , further comprising a reduced image generation unit which executes a process of reducing the synthesized image,

wherein the image correction unit sets the reference region in the reduced image generated by the reduced image generation unit, and applies a reference pixel selected from the reference region set in the reduced image co as to calculate the similarity weight coefficient and the pixel position weight coefficient.

5. The image processing device according to any one of claims 1 to 3 , wherein

a second synthesized image based on an image photographed at a timing different from a first synthesized image including the pixel to be corrected or a reduced image of the second synthesized image, is received as a second reference image,

a second reference region is set with respect to the second reference image in the vicinity of the pixel position corresponding to the pixel to be corrected of the first synthesized image,

non-pseudo color pixels included in the second reference region are selected as reference pixels,

the similarity weight coefficient and the pixel position weight coefficient are set with respect to the selected reference pixels, and

both the reference pixels selected from the synthesized image including the pixel to be corrected and the reference pixels selected from the second reference region are applied so as to determine the correction pixel value of the pixel to be corrected by a weighted summation process of summing values obtained by multiplying the pixel values of the reference pixels by the similarity weight coefficient and the pixel position weight coefficient.

6. The image processing device according to any one of claims 1 to 3 , wherein the image correction unit calculates the similarity weight coefficient and the pixel position weight coefficient by applying a Gaussian function.

7. The image processing device according to any one of claims 1 to 3 , further comprising an image analysis unit which receives the synthesized image and executes detection of the pseudo color pixel included in the synthesized image so as to generate the pseudo color pixel information,

wherein the image correction unit applies the pseudo color pixel information generated by the image analysis unit so as to perform the selection of the non-pseudo color pixels in the reference region.

8. The image processing device according to any one of claims 1 to 3 , further comprising a synthesized image generation unit which synthesizes the plurality of images to generate the synthesized image,

wherein the image correction unit executes the correction of the synthesized image generated by the synthesized image generation unit.

9. The image processing device according to any one of claims 1 to 3 , wherein the synthesized image is generated by synthesizing the plurality of images photographed by setting different exposure times.

10. An imaging apparatus comprising:

an imaging device which photographs captured images of different exposure times; and

the image processing unit which executes the image processing according to any one of claims 1 to 3 .

11. An image processing method which executes pixel value correction in an image processing device, the image processing method comprising the steps of:

at an image correction unit:

receiving a synthesized image generated by synthesizing a plurality of images and pseudo color pixel information indicating a pseudo color pixel position included in the synthesized image; and

executing the pixel value correction using a pseudo color pixel in the synthesized image as a pixel to be corrected,

wherein the correction comprises:

selecting non-pseudo color pixels, which are not determined as a pseudo color based on the pseudo color pixel information, as reference pixels, from a reference region set in a surrounding region of the pixel to be corrected,

setting a similarity weight coefficient in which a large weight is set as a pixel value of the reference pixel that is close to a pixel value of the pixel to be corrected, and a pixel position weight coefficient in which a large weight is set as the pixel value of the reference pixel that is near a pixel position of the pixel to be corrected, with respect to the reference pixels, and

determining a correction pixel value of the pixel to be corrected by a weighted summation process of summing values obtained by multiplying the pixel values of the reference pixels by the similarity weight coefficient and the pixel position weight coefficient.

12. A non-transitory computer readable medium comprising program code for causing an image processing device to perform steps comprising:

at an image correction unit:

receiving a synthesized image generated by synthesizing a plurality of images and pseudo color pixel information indicating a pseudo color pixel position included in the synthesized image; and

executing pixel value correction using a pseudo color pixel in the synthesized image as a pixel to be corrected,

wherein correction comprises:

selecting non-pseudo color pixels, which are not determined as a pseudo color based on the pseudo color pixel information, as reference pixels, from a reference region set in a surrounding region of the pixel to be corrected,

setting a similarity weight coefficient in which a large weight is set as a pixel value of the reference pixel that is close to a pixel value of the pixel to be corrected, and a pixel position weight coefficient in which a large weight is set as the pixel value of the reference pixel that is near a pixel position of the pixel to be corrected, with respect to the reference pixels, and

determining a correction pixel value of the pixel to be corrected by a weighted summation process of summing values obtained by multiplying the pixel values of the reference pixels by the similarity weight coefficient and the pixel position weight coefficient.