Determining image quality based on distribution of representative autocorrelation coefficients
View Patent ↗An image evaluation device includes: a partial area extracting section extracting plural partial areas from an original image; an extracted image generating section generating an extracted image corresponding to each of the partial areas and having pixels whose pixel values correspond to a gradient of pixel values in the image; an autocorrelation calculating section calculating plural autocorrelation coefficients for each extracted images; a representative coefficient value calculating section calculating a representative coefficient value for each of the autocorrelation coefficients among the partial areas; and a checking section checking the quality of the image based on a distribution of the representative coefficient values.
1. An image evaluation device, comprising:
a partial area extracting section that extracts a plurality of partial areas from an image;
an extracted image generating section that extracts plural extracted images corresponding to plural partial areas from the image;
an autocorrelation calculating section that calculates autocorrelation coefficients in two dimensional displacement for each extracted image;
a degradation function estimating section that calculates a value of a degradation function for each two dimensional displacement, by calculating a representative value of the autocorrelation coefficients at the corresponding two dimensional displacement among the partial areas, the degradation function representing a degree of degradation of the image from the autocorrelation coefficients for each two dimensional displacement; and
a checking section that checks the quality of the image based on the degradation function.
2. The image evaluation device according to claim 1 ,
wherein the degradation function estimating section calculates the degradation function by calculating an extent width of the distribution of the representative coefficient values.
3. The image evaluation device according to claim 2 ,
wherein the degradation function estimating section calculates the degradation function by comparing the extent width with a predetermined threshold value.
4. The image evaluation device according to claim 1 ,
wherein the representative coefficient value of the plurality of autocorrelation coefficients is the minimum value of the plurality of autocorrelation coefficients or a value of autocorrelation coefficients in predetermined order from the minimum value.
5. The image evaluation device according to claim 4 ,
wherein the representative coefficient value of the plurality of autocorrelation coefficients is replaced with a value equal to or greater than 0 if the value is less than 0.
6. The image evaluation device according to claim 1 ,
wherein the extracted images are extracted by extracting an edge image for each partial area.
7. The image evaluation device according to claim 6 ,
wherein pixel value of each pixel in the edge image represents the gradient of pixel value in a predetermined direction or a pair of directions on the pixel.
8. The image evaluation device according to claim 1 ,
wherein the extracted image generating section comprises,
an image generating section that generates an image having pixels whose pixel values correspond to the gradient of pixel values in the original image; and
wherein an extracted image extracting section extracts extracted images for each of the partial areas, from the generated image.
9. The image evaluation device according to claim 1 ,
wherein the partial area extracting section comprises,
a dividing section that divides the image into plural partial areas;
wherein a first decision section computes a representative pixel value for each of the divided partial areas;
wherein a second decision section computes a range of pixel values based on the computed representative pixel value; and
wherein an extracting section extracts a partial area having a predetermined number or more of pixels whose pixel values are outside of the computed range of the pixel values, from the divided partial areas.
10. The image evaluation device according to claim 1 , further comprising,
a gray-scale section gray-scaling an image,
wherein the extracted image generating section generates an extracted image from the gray-scale image.
11. An image evaluation method, comprising:
extracting plural partial areas from an image;
generating plural extracted images corresponding to the plural partial areas;
calculating plural autocorrelation coefficients in two dimensional displacement for each extracted image;
calculating a value of a degradation function for each two dimensional displacement, by calculating a representative value of the autocorrelation coefficients at the corresponding two dimensional displacement among the partial areas, the degradation function representing a degree of degradation of the image from the autocorrelation coefficients for each two dimensional displacement; and
checking the quality of the image based on the degradation function.
12. The image evaluation method according to claim 11 ,
wherein the degradation function is calculated by calculating an extent width of the distribution of the representative coefficient values.
13. The image evaluation method according to claim 12 ,
wherein the degradation function is calculated by comparing the extent width with a predetermined threshold value.
14. The image evaluation method according to claim 11 ,
wherein the representative coefficient value of the autocorrelation coefficients is the minimum value of the autocorrelation coefficients or a value of autocorrelation coefficient in predetermined order from the minimum value.
15. The image evaluation method according to claim 14 ,
wherein the representative coefficient value of the plural autocorrelation coefficients is replaced with a value equal to or greater than 0 when the value is less than 0.
16. The image evaluation method according to claim 11 ,
wherein the extracted image is extracted by extracting an edge image in a partial area.
17. The image evaluation method according to claim 16 ,
wherein the pixel value for each pixel in the edge image represents a gradient of pixel value in a predetermined direction or a pair of directions at the position.
18. The image evaluation method according to claim 11 ,
wherein the step of generating the extracted images comprises:
generating an image having pixels whose pixel values correspond to the gradient of pixel values in the original image; and
extracting extracted images from each of the partial areas in the generated image.
19. The image evaluation method according to claim 11 ,
wherein the step of generating the extracted images comprises:
dividing the image into a plurality of partial areas;
computing a representative pixel value for each of the plurality of divided partial areas;
computing a range of pixel values based on the computed representative pixel value; and
extracting, from the divided partial areas, a partial area having a predetermined number or more of pixels whose pixel values are outside of the computed range of the pixel values.
20. The image evaluation method according to claim 11 , further comprising,
gray-scaling an image,
wherein in the step of generating the extracted images, an extracted image is generated from the gray-scale image.