Method, apparatus, and program for discriminating faces
A face discrimnating process judges whether a discrimination target image is an image of a face, based on characteristic amounts of the discrimination target image. A first and a second brightness gradation converting process are administered as preliminary processes. The first brightness gradation converting process is administered on regions in which the degrees of variance of pixel values are greater than or equal to a first predetermined level. The second brightness gradation converting process is administered on regions in which the degrees of variance are less than the first predetermined level. The first brightness gradation converting process causes the degree of variance to approach a second predetermined level. The second brightness gradation converting process suppresses the degree of variance to be less than the second predetermined level.
1 . A face discrimnating method, comprising:
a normalizing step for suppressing fluctuations in contrast within a discrimination target image, which is a target of judgment regarding whether the image is a facial image; and
a face discrimnating step for calculating at least one characteristic amount related to the brightness distribution of the discrimination target image, on which the normalizing step has been administered, and discrimnating whether the discrimination target image is a facial image by employing the characteristic amount;
the normalizing step comprising:
a first brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are greater than or equal to a first predetermined level, to approach a second predetermined level, which is higher than the first predetermined level, for each of a plurality of local regions within the discrimination target image; and
a second brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are less than the first predetermined level, to be suppressed such that it is lower than the second predetermined level, for each of a plurality of local regions within the discrimination target image.
2 . A face discrimnating method, comprising:
a normalizing step for suppressing fluctuations in contrast within a discrimination target image, which is a target of judgment regarding whether the image is a facial image; and
a face discrimnating step for calculating at least one characteristic amount related to the brightness distribution of the discrimination target image, on which the normalizing step has been administered, and discrimnating whether the discrimination target image is a facial image by employing the characteristic amount;
the normalizing step comprising:
a first brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are greater than or equal to a first predetermined level, to approach a second predetermined level, which is higher than the first predetermined level, for each of a plurality of local regions within the discrimination target image.
3 . A face discrimnating apparatus, comprising:
normalizing means for administering a normalizing process that suppresses fluctuations in contrast within a discrimination target image, which is a target of judgment regarding whether the image is a facial image; and
face discrimnating means for calculating at least one characteristic amount related to the brightness distribution of the discrimination target image, on which the normalizing process has been administered, and discrimnating whether the discrimination target image is a facial image by employing the characteristic amount;
the normalizing process comprising:
a first brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are greater than or equal to a first predetermined level, to approach a second predetermined level, which is higher than the first predetermined level, for each of a plurality of local regions within the discrimination target image; and
a second brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are less than the first predetermined level, to be suppressed such that it is lower than the second predetermined level, which is lower than the first predetermined level, for each of a plurality of local regions within the discrimination target image.
4 . A face discrimnating apparatus, comprising:
normalizing means for administering a normalizing process to suppress fluctuations in contrast within a discrimination target image, which is a target of judgment regarding whether the image is a facial image; and
face discrimnating means for calculating at least one characteristic amount related to the brightness distribution of the discrimination target image,-on which the normalizing process has been administered, and discrimnating whether the discrimination target image is a facial image by employing the characteristic amount;
the normalizing process comprising:
a first brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are greater than or equal to a first predetermined level, to approach a second predetermined level, which is higher than the first predetermined level, for each of a plurality of local regions within the discrimination target image.
5 . A face discrimnating apparatus as defined in claim 3 , wherein the normalizing process comprises the steps of:
sequentially setting each pixel within the discrimination target image as a pixel of interest;
calculating degrees of variance within local regions of a predetermined size, of which the pixels of interest are representative pixels; and
causing the differences between the pixel values of the pixels of interest and predetermined statistical representative pixel values of the local regions to become smaller, as the difference between the degrees of variance and a reference value corresponding to the second predetermined level become greater when the degrees of variance are greater than the reference value, and causing the differences between the pixel values of the pixels of interest and the predetermined statistical representative pixel values of the local regions to become greater, as the difference between the degrees of variance and the reference value become greater, when the degrees of variance are less than the reference value, at least in cases that the degrees of variance are greater than or equal to a threshold value corresponding to the first predetermined level, as the first brightness gradation converting process.
6 . A face discrimnating apparatus as defined in claim 4 , wherein the normalizing process comprises the steps of:
sequentially setting each pixel within the discrimination target image as a pixel of interest;
calculating degrees of variance within local regions of a predetermined size, of which the pixels of interest are representative pixel:s; and
causing the differences between the pixel values of the pixels of interest and predetermined statistical representative pixel values of the local regions to become smaller, as the difference between the degrees of variance and a reference value corresponding to the second predetermined level become greater when the degrees of variance are greater than the reference value, and causing the differences between the pixel values of the pixels of interest and the predetermined statistical representative pixel values of the local regions to become greater, as the difference between the degrees of variance and the reference value become greater, when the degrees of variance are less than the reference value, at least in cases that the degrees of variance are greater than or equal to a threshold value corresponding to the first predetermined level, as the first brightness gradation converting process.
7 . A face discrimnating apparatus as defined in claim 5 , wherein:
the threshold value varies according to the pixel value of the pixel of interest.
8 . A face discrimnating apparatus as defined in claim 6 , wherein:
the threshold value varies according to the pixel value of the pixel of interest.
9 . A face discrimnating apparatus as defined in claim 3 , wherein:
the face discrimnating means comprises a plurality of different weak classifiers for discrimnating whether the discrimination target image is a facial image, which are linearly linked in order of reliability thereof.
10 . A face discrimnating apparatus as defined in claim 4 , wherein:
the face discrimnating means comprises a plurality of different weak classifiers for discrimnating whether the discrimination target image is a facial image, which are linearly linked in order of reliability thereof.
11 . A face discrimnating apparatus as defined in claim 5 , wherein:
the face discrimnating means comprises a plurality of different weak classifiers for discrimnating whether the discrimination target image is a facial image, which are linearly linked in order of reliability thereof.
12 . A face discrimnating apparatus as defined in claim 6 , wherein:
the face discrimnating means comprises a plurality of different weak classifiers for discrimnating whether the discrimination target image is a facial image, which are linearly linked in order of reliability thereof.
13 . A face discrimnating apparatus as defined in claim 7 , wherein:
the face discrimnating means comprises a plurality of different weak classifiers for discrimnating whether the discrimination target image is a facial image, which are linearly linked in order of reliability thereof.
14 . A face discrimnating apparatus as defined in claim 8 , wherein:
the face discrimnating means comprises a plurality of different weak classifiers for discrimnating whether the discrimination target image is a facial image, which are linearly linked in order of reliability thereof.
15 . A face discrimnating apparatus as defined in claim 9 , wherein:
the types of weak classifiers to be employed and the order in which the weak classifiers are linked is determined by learning, employing sample facial images, in which the directions that the faces pictured therein are facing and the vertical orientations thereof are the same.
16 . A face discrimnating apparatus as defined in claim 10 , wherein:
the types of weak classifiers to be employed and the order in which the weak classifiers are linked is determined by learning, employing sample facial images, in which the directions that the faces pictured therein are facing and the vertical orientations thereof are the same.
17 . A face discrimnating apparatus as defined in claim 11 , wherein:
the types of weak classifiers to be employed and the order in which the weak classifiers are linked is determined by learning, employing sample facial images, in which the directions that the faces pictured therein are facing and the vertical orientations thereof are the same.
18 . A face discrimnating apparatus as defined in claim 12 , wherein:
the types of weak classifiers to be employed and the order in which the weak classifiers are linked is determined by learning, employing sample facial images, in which the directions that the faces pictured therein are facing and the vertical orientations thereof are the same.
19 . A face discrimnating apparatus as defined in claim 13 , wherein:
the types of weak classifiers to be employed and the order in which the weak classifiers are linked is determined by learning, employing sample facial images, in which the directions that the faces pictured therein are facing and the vertical orientations thereof are the same.
20 . A face discrimnating apparatus as defined in claim 14 , wherein:
the types of weak classifiers to be employed and the order in which the weak classifiers are linked is determined by learning, employing sample facial images, in which the directions that the faces pictured therein are facing and the vertical orientations thereof are the same.
21 . A computer readable medium having a program stored therein, the program causing a computer to execute:
a normalizing step for suppressing fluctuations in contrast within a discrimination target image, which is a target of judgment regarding whether the image is a facial image; and
a face discrimnating step for calculating at least one characteristic amount related to the brightness distribution of the discrimination target image, on which the normalizing step has been administered, and discrimnating whether the discrimination target image is a facial image by employing the characteristic amount;
the normalizing step comprising:
a first brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are greater than or equal to a first predetermined level, to approach a second predetermined level, which is higher than the first predetermined level, for each of a plurality of local regions within the discrimination target image; and
a second brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are less than the first predetermined level, to be suppressed such that it is lower than the second predetermined level, for each of a plurality of local regions within the discrimination target image.
22 . A computer readable medium having a program stored therein, the program causing a computer to execute:
a normalizing step for suppressing fluctuations in contrast within a discrimination target image, which is a target of judgment regarding whether the image is a facial image; and
a face discrimnating step for calculating at least one characteristic amount related to the brightness distribution of the discrimination target image, on which the normalizing step has been administered, and discrimnating whether the discrimination target image is a facial image by employing the characteristic amount;
the normalizing step comprising:
a first brightness gradation converting process, for causing the degree of variance of pixel values representing brightness, which are greater than or equal to a first predetermined level, to approach a second predetermined level, which is higher than the first predetermined level, for each of a plurality of local regions within the discrimination target image.