Face region detection device, imaging apparatus, face region detection method, and face region detection program

1. A face region detection device that detects a face region from a captured image which is output from an imaging element, the device comprising:

a processor configured to

set the face region detected from a first captured image, as a reference image, and search for a similar region similar to a specific image which is a part of the reference image from a second captured image output after the first captured image,

in a case where the second captured image is determined to have the similar region similar to the specific image, set a first region as a face region, the first region includes the similar region in the second captured image,

determine a positional relationship between the reference image and the specific image in the first captured image; and

set the first region such that a size of the first region is same for a size of the reference image, and such that a position of the similar region in the first region coincide with a position of the specific image in the reference image, using the positional relationship between the reference image and the specific image in the first captured image,

wherein in a case where a difference between first distance information of a subject included in the reference image and second distance information of the subject included in the first region exceeds a predetermined distance threshold value,

the processor is configured to determine that the first region is a non-face region.

2. The face region detection device according to claim 1 ,

wherein the processor is configured to predict a movement of a face included in the face region of the first captured image, in an optical axis direction of an imaging optical system for making the imaging element form a subject image, on the basis of the first captured image and the captured image which is output from the imaging element prior to the first captured image, and

control the distance threshold value on the basis of the movement which is predicted.

3. The face region detection device according to claim 2 ,

wherein in a case where the movement is a first value, the processor is configured to reduce the distance threshold value as compared with a case where the movement is a second value larger than the first value.

4. The face region detection device according to claim 3 ,

wherein the processor is configured to perform focus control of an imaging optical system for making the imaging element form a subject image, on the basis of the face region detected from the captured image, and

in a case where the difference exceeds the distance threshold value, fix a focus state of the imaging optical system for a predetermined period of time.

5. The face region detection device according to claim 2 ,

wherein the processor is configured to perform focus control of an imaging optical system for making the imaging element form a subject image, on the basis of the face region detected from the captured image, and

in a case where the difference exceeds the distance threshold value, fix a focus state of the imaging optical system for a predetermined period of time.

6. The face region detection device according to claim 1 ,

wherein the processor is configured to perform focus control of an imaging optical system for making the imaging element form a subject image, on the basis of the face region detected from the captured image, and

in a case where the difference exceeds the distance threshold value, fix a focus state of the imaging optical system for a predetermined period of time.

7. The face region detection device according to claim 1 ,

wherein the processor is configured to decide the specific image on the basis of an analysis result of the reference image.

8. The face region detection device according to claim 7 ,

wherein the processor is configured to obtain a contrast value of the reference image, and decide a portion of the reference image in which the contrast value exceeds a predetermined threshold value, as the specific image.

9. An imaging apparatus comprising:

the face region detection device according to claim 1 ; and

the imaging element.

10. A face region detection method of detecting a face region from a captured image which is output from an imaging element, the method comprising:

a tracking processing step of setting the face region detected from a first captured image, as a reference image, searching for a similar region similar to a specific image which is a part of the reference image from a second captured image output after the first captured image, and setting a first region including the similar region in the second captured image as the face region in a case where the second captured image is determined to have the similar region similar to the specific image,

wherein, in the tracking processing step, determining a positional relationship between the reference image and the specific image in the first captured image, setting the first region such that a size of the first region is same for a size of the reference image, and such that a position of the similar region in the first region coincide with a position of the specific image in the reference image, using the positional relationship between the reference image and the specific image in the first captured image,

wherein in the tracking processing step, in a case where a difference between first distance information of a subject included in the reference image and second distance information of the subject included in the first region exceeds a predetermined distance threshold value, determination is made that the first region is a non-face region.

11. The face region detection method according to claim 10 , further comprising:

a prediction step of predicting a movement of a face included in the face region of the first captured image, in an optical axis direction of an imaging optical system for making the imaging element form a subject image, on the basis of the first captured image and the captured image which is output from the imaging element prior to the first captured image,

wherein in the tracking processing step, the distance threshold value is controlled on the basis of the movement predicted in the prediction step.

12. The face region detection method according to claim 11 ,

wherein in the tracking processing step, in a case where the movement is a first value, the distance threshold value is reduced as compared with a case where the movement is a second value larger than the first value.

13. The face region detection method according to claim 10 , further comprising:

a focus control step of performing focus control of an imaging optical system for making the imaging element form a subject image, on the basis of the face region detected from the captured image,

wherein in the focus control step, in a case where the difference exceeds the distance threshold value, a focus state of the imaging optical system is fixed for a predetermined period of time.

14. The face region detection method according to claim 10 ,

wherein in the tracking processing step, the specific image is decided on the basis of an analysis result of the reference image.

15. The face region detection method according to claim 14 ,

wherein in the tracking processing step, a contrast value of the reference image is obtained, and a portion of the reference image in which the contrast value exceeds a predetermined threshold value is decided as the specific image.

16. A non-transitory computer readable recording medium storing a face region detection program for detecting a face region from a captured image which is output from an imaging element, the program causing a computer to execute:

a tracking processing step of setting the face region detected from a first captured image, as a reference image, searching for a similar region similar to a specific image which is a part of the reference image from a second captured image output after the first captured image, and setting a first region including the similar region in the second captured image as the face region in a case where the second captured image is determined to have the similar region similar to the specific image,

wherein, in the tracking processing step, determining a positional relationship between the reference image and the specific image in the first captured image, setting the first region such that a size of the first region is same for a size of the reference image, and such that a position of the similar region in the first region coincide with a position of the specific image in the reference image, using the positional relationship between the reference image and the specific image in the first captured image,

wherein in the tracking processing step, in a case where a difference between first distance information of a subject included in the reference image and second distance information of the subject included in the first region exceeds a predetermined distance threshold value, determination is made that the first region is a non-face region.