Image recognition apparatus and image recognition method
View Patent ↗An image recognition apparatus according to one aspect of the present invention has a measurement unit measuring a blur level of an image, a comparison unit comparing the blur level measured in the measurement unit with a threshold, an image processing unit applying to the image a blurring filter which increases the blur level when the blur level measured in the measurement unit is smaller than the threshold, and applying to the image a deblurring filter which decreases the blur level when the blur level measured in the measurement unit is larger than the threshold, and a recognition unit recognizing the image from features of the image processed in the image processing unit.
1. An image recognition apparatus, comprising:
a storage unit storing a threshold calculated based on a blur level of a normalized image, the normalized image being obtained by normalizing an image in a predetermined size;
a measurement unit measuring a blur level of an image;
a comparison unit comparing the blur level measured in the measurement unit with the threshold;
an image processing unit applying to the image a blurring filter which increases the blur level when the blur level measured in the measurement unit is smaller than the threshold, and applying to the image a deblurring filter which decreases the blur level when the blur level measured in the measurement unit is larger than the threshold; and
a recognition unit recognizing the image from features of the image processed in the image processing unit.
2. The apparatus according to claim 1 ,
wherein the threshold includes a first value and a second value different from each other; and
wherein the image processing unit applies the blurring filter or the deblurring filter to the image depending on a comparison result in the comparison unit until the blur level measured in the measurement unit becomes larger than the first value and smaller than the second value.
3. The apparatus according to claim 2 , wherein the image processing unit decreases an amount of change of the blur level by the blurring filter or the deblurring filter in stages.
4. The apparatus according to claim 1 , wherein the image processing unit uses a Gaussian filter as the blurring filter, and uses a Laplacian filter as a part of the deblurring filter.
5. An image recognition method, comprising:
normalizing an image in a predetermined size;
measuring a blur level of the normalized image;
calculating a threshold based on the blur level;
measuring a blur level of an input image;
comparing the blur level with the threshold;
applying to the input image a blurring filter which increases the blur level when the blur level measured in the measurement is smaller than the threshold, and applying to the input image a deblurring filter which decreases the blur level when the blur level measured in the measurement is larger than the threshold; and
recognizing the input image from features of the input image after the filter is applied.
6. The method according to claim 5 ,
wherein the threshold includes a first value and a second value different from each other; and
wherein in the application, the blurring filter or the deblurring filter is applied to the input image depending on a comparison result in the comparison until the blur level measured in the measurement becomes equal to or larger than the first value and equal to or smaller than the second value.
7. The method according to claim 5 , wherein in the application, an amount of change of the blur level by the blurring filter or the deblurring filter is decreased in stages.
8. The method according to claim 5 , wherein in the application, a Gaussian filter is used as the blurring filter, and a Laplacian filter is used as the deblurring filter.