Image processing unit, image processing method, and image processing program to correct blurs and noise in an image
An image processing unit includes a memory unit in which continuously captured images including a reference image and a comparative image are stored, an image dividing unit to divide the reference image and the comparative image into image blocks of a predetermined size, a mean value calculator unit to calculate a mean value of pixel outputs in each image block of each of the reference and comparative images, a threshold determining unit to determine a threshold according to a mean value of pixel outputs of an image block of the reference image, and a determiner unit to compare the threshold with a difference value of the mean values of the pixel outputs in the image blocks of the reference and comparative images to be synthesized and determine whether the image blocks of the reference and comparative images are suitable for image synthesis based on a result of the comparison.
1. An image processing unit comprising:
an image divider configured to divide each of a reference image and a comparative image, different from the reference image, into image blocks of a predetermined size;
a mean value calculator configured to calculate a first mean value of pixel outputs in each image block of the reference image and a second mean value of pixel outputs in each image block of the comparative image;
a threshold determiner configured to determine a threshold according to the first mean value of pixel outputs of an image block of the reference image, the threshold being for determining whether the image blocks of the reference image and the comparative image are suitable for image synthesis; and
a determiner configured to compare the threshold determined by the threshold determiner with a difference value that is a difference between the first mean value of pixel outputs in the image blocks of the reference image and the second mean value of pixel outputs in the image blocks of the comparative image and to determine whether the image blocks of the reference image and the comparative image are suitable for image synthesis based on a result of comparison.
2. The image processing unit according to claim 1 , further comprising:
a motion data calculator configured to calculate data on a motion between the reference image and the comparative image, wherein
the mean value calculator is configured to calculate the first mean value of pixel outputs for each image block of the reference image and the second mean value of pixel outputs for each image block of the comparative image whose positions are associated with each other based on the calculated motion data, and
the determiner is configured to determine whether the image blocks of the comparative image and reference image whose positions are associated with each other are suitable for image synthesis.
3. The image processing unit according to claim 1 , wherein
the first mean value of pixel outputs and the second mean value of pixel outputs calculated by the mean value calculator are mean values of each of R pixel outputs, G pixel outputs, and B pixel outputs in each image block, and
the threshold determiner is configured to determine the threshold according to the mean values of each of the R pixel outputs, G pixel outputs, and B pixel outputs.
4. The image processing unit according to claim 1 , wherein
the first mean value of pixel outputs and the second mean value of pixel outputs calculated by the mean value calculator is a mean value of each of R pixel outputs, G pixel outputs, and B pixel outputs in each image block, and
the threshold determiner is configured to determine the threshold according to a maximal value of respective mean values of the R pixel outputs, G pixel outputs, and B pixel outputs.
5. The image processing unit according to claim 1 , wherein
the first mean value of pixel outputs and the second mean value of pixel outputs calculated by the mean value calculator is a mean value of brightness of outputs of pixels included in each image block.
6. The image processing unit according to claim 1 , wherein
the threshold determiner is configured to determine different thresholds according to ISO sensitivity.
7. The image processing unit according to claim 1 , further comprising
an image synthesizer configured to synthesize the image block of the reference image and the image block of the comparative image determined as suitable for image synthesis by the determiner.
8. An image processing method, comprising:
dividing each of a reference image and a comparative image, different from the reference image, into image blocks of a predetermined size;
calculating a first mean value of pixel outputs in each image block of each of the reference image and a second mean value of pixel outputs in each image block of the comparative image;
determining a threshold according to the first mean value of pixel outputs of an image block of the reference image, the threshold being for determining whether the image blocks of the reference image and the comparative image are suitable for image synthesis; and
comparing the threshold with a difference value that is a difference between the first mean values of pixel outputs in the image blocks of the reference image and the second mean of pixel outputs in the image blocks of the comparative image and determining whether the image blocks of the reference image and the comparative image are suitable for image synthesis based on a result of said comparing.
9. The image processing method according to claim 8 , further comprising:
calculating data on a motion between the reference image and the comparative image, wherein
the first mean value of pixel outputs is calculated for each image block of the reference image and the second mean value of pixel outputs is calculated for each image block of the comparative image whose positions are associated with each other based on the calculated motion data, and
a determination is made on whether the image blocks of the comparative image and reference image whose positions are associated with each other are suitable for image synthesis.
10. The image processing method according to claim 8 , wherein:
the first mean value of pixel outputs and the second mean value of pixel outputs are mean values of each of R pixel outputs, G pixel outputs, and B pixel outputs in each image block, and
the threshold is determined according to the mean values of each of the R pixel outputs, G pixel outputs, and B pixel outputs.
11. The image processing method according to claim 8 , wherein
the first mean value of pixel outputs and the second mean value of pixel outputs are mean values of each of R pixel outputs, G pixel outputs, and B pixel outputs in each image block, and
the threshold is determined according to a maximal value of respective mean values of the R pixel outputs, G pixel outputs, and B pixel outputs.
12. The image processing method according to claim 8 , wherein
the first mean value of pixel outputs and the second mean value of pixel outputs are mean values of brightness of outputs of pixels included in each image block.
13. The image processing method according to claim 8 , wherein
different thresholds are determined according to ISO sensitivity.
14. The image processing method according to claim 8 , further comprising:
synthesizing the image block of the reference image and the image block of the comparative image determined as suitable for image synthesis.
15. An image processing program stored on a non-transitory medium which causes a computer to function as the image processing unit according to claim 1 .
16. The image processing unit according to claim 1 , wherein the threshold determiner is configured to determine a different threshold for each image block of the reference image.
17. The image processing unit according to claim 16 , wherein the different threshold for each image block of the reference image is determined in accordance with the first mean value of pixel outputs of each image block of the reference image.
18. The image processing unit according to claim 1 , wherein a value of the threshold of each image block of the reference image is directly proportional to the first mean value of pixel outputs of each image block of the reference image.