Electronic apparatus, blur image sorting method, and program

1. An electronic apparatus, comprising:

an extraction means for extracting, from an image, a subject area having a predetermined feature in the image;

a first calculation means for calculating a first blur degree that indicates a blur degree of the extracted subject area;

a second calculation means for calculating, when the number of subject areas in the image from which the first blur degree is calculated is one, a second blur degree that indicates a blur degree of the entire image based on the first blur degree, and calculating, when the number of subject areas in the image from which the first blur degree is calculated is plural, the second blur degree based on a value obtained by performing weighted averaging on the plurality of first blur degrees in accordance with dimensions of the plurality of subject areas; and

a sorting means for sorting an image having the calculated second blur degree that is equal to or larger than a predetermined threshold value, as a blur image, from a plurality of images.

2. The electronic apparatus according to claim 1 , further comprising

an optimization means for optimizing the extracted subject area so that the extracted subject area will have a predetermined size adequate to a calculation of the first blur degree.

3. The electronic apparatus according to claim 2 ,

wherein the extraction means calculates a score that indicates a certainty of extraction of the subject area, and

wherein the second calculation means calculates, when the number of subject areas in the image from which the first blur degree is calculated is plural, the second blur degree based on a value obtained by performing weighted averaging on the plurality of first blur degrees in accordance with the dimensions of the plurality of subject areas and the calculated score.

4. The electronic apparatus according to claim 3 ,

wherein the extraction means includes

a face recognition means for recognizing a face area of a human face as the subject area and calculating a first score that indicates the score of the recognized face area, and

a feature recognition means for recognizing a feature area that is visually salient as the subject area and calculating a second score that indicates the score of the recognized feature area, and

wherein, in a case where the number of subject areas in the image from which the first blur degree is calculated is plural, the second calculation means calculates the second blur degree without using the first score in the weighted averaging when the face area is recognized as the subject area by the face recognition means, and calculates the second blur degree using the second score in the weighted averaging when the feature area is recognized as the subject area by the feature recognition means.

5. The electronic apparatus according to claim 2 ,

wherein the first calculation means calculates the first blur degree with the entire image as the subject area when no subject area is extracted.

6. The electronic apparatus according to claim 2 , further comprising:

an operation reception means for receiving an operation of a user; and

a display means for displaying the plurality of images,

wherein the sorting means sorts out the blur image in accordance with a predetermined operation of the user, and

wherein the display means displays only the sorted blur image out of the plurality of displayed images when the predetermined operation is received.

7. A blur image sorting method, comprising:

extracting, from an image, a subject area having a predetermined feature in the image;

calculating a first blur degree that indicates a blur degree of the extracted subject area;

calculating, when the number of subject areas in the image from which the first blur degree is calculated is one, a second blur degree that indicates a blur degree of the entire image based on the first blur degree, and calculating, when the number of subject areas in the image from which the first blur degree is calculated is plural, the second blur degree based on a value obtained by performing weighted averaging on the plurality of first blur degrees in accordance with dimensions of the plurality of subject areas; and

sorting an image having the calculated second blur degree that is equal to or larger than a predetermined threshold value, as a blur image, from a plurality of images.

8. A blur image sorting method, comprising:

optimizing a subject area having a predetermined feature, that is extracted from an image so that the subject area will have a predetermined size;

acquiring a first blur degree that is calculated from the optimized subject area and indicates a blur degree of the subject area;

calculating, when the number of subject areas in the image from which the first blur degree is calculated is one, a second blur degree that indicates a blur degree of the entire image based on the first blur degree, and calculating, when the number of subject areas in the image from which the first blur degree is calculated is plural, the second blur degree based on a value obtained by performing weighted averaging on the plurality of first blur degrees in accordance with dimensions of the plurality of subject areas; and

sorting an image having the calculated second blur degree that is equal to or larger than a predetermined threshold value, as a blur image, from a plurality of images.

9. A non-transitory computer readable media having computer readable instructions stored thereon that when executed by a processing circuit causes an electronic apparatus to perform a method comprising:

extracting, from an image, a subject area having a predetermined feature in the image;

calculating a first blur degree that indicates a blur degree of the extracted subject area;

calculating, when the number of subject areas in the image from which the first blur degree is calculated is one, a second blur degree that indicates a blur degree of the entire image based on the first blur degree, and calculating, when the number of subject areas in the image from which the first blur degree is calculated is plural, the second blur degree based on a value obtained by performing weighted averaging on the plurality of first blur degrees in accordance with dimensions of the plurality of subject areas; and

sorting an image having the calculated second blur degree that is equal to or larger than a predetermined threshold value, as a blur image, from a plurality of images.

10. A non-transitory computer readable media having computer readable instructions stored thereon that when executed by a processing circuit causes an electronic apparatus to perform a method comprising:

optimizing a subject area having a predetermined feature, that is extracted from an image so that the subject area will have a predetermined size;

acquiring a first blur degree that is calculated from the optimized subject area and indicates a blur degree of the subject area;

calculating, when the number of subject areas in the image from which the first blur degree is calculated is one, a second blur degree that indicates a blur degree of the entire image based on the first blur degree, and calculating, when the number of subject areas in the image from which the first blur degree is calculated is plural, the second blur degree based on a value obtained by performing weighted averaging on the plurality of first blur degrees in accordance with dimensions of the plurality of subject areas; and

sorting an image having the calculated second blur degree that is equal to or larger than a predetermined threshold value, as a blur image, from a plurality of images.

11. An electronic apparatus, comprising:

an extraction section to extract, from an image, a subject area having a predetermined feature in the image;

a processing circuit that implements a first calculation section to calculate a first blur degree that indicates a blur degree of the extracted subject area;

a second calculation section to calculate, when the number of subject areas in the image from which the first blur degree is calculated is one, a second blur degree that indicates a blur degree of the entire image based on the first blur degree, and calculate, when the number of subject areas in the image from which the first blur degree is calculated is plural, the second blur degree based on a value obtained by performing weighted averaging on the plurality of first blur degrees in accordance with dimensions of the plurality of subject areas; and

a sorting section to sort an image having the calculated second blur degree that is equal to or larger than a predetermined threshold value, as a blur image, from a plurality of images.