Image processing device and image processing method
View Patent ↗A face detection processing unit performs a face detection process by rotating an image in increments of a predetermined angle and acquires a rotation angle at which a face is detected. An angle correction unit acquires an angle between the face and a shoulder by pattern matching and corrects the rotation angle of the image. A human-image orientation identification unit identifies a correct orientation of a human image based on the rotation angle. An image-characteristic analysis unit analyzes a frequency distribution and a brightness distribution of a non-human image. A non-human image orientation identification unit identifies the correct orientation of a non-human image based on distribution characteristics of a frequency of brightness with respect to an axis in a predetermined direction. An image-data updating unit incorporates information regarding the correct orientation in the image data.
1. An image processing device comprising:
an object detection processing unit configured to detect, by performing pattern matching between a preexisting image containing an object to be processed and a reference image of the object, the object in the image to be processed, while changing the relative angle of the images;
an object-image orientation identification unit configured to determine the top and the bottom of the preexisting image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected by the object detection processing unit, so as to identify a correct orientation of the preexisting image;
wherein the object detection processing unit performs a process for detecting a face as the object; and
an angle correction unit configured to obtain an angle between an axis of the face in a vertical direction and a horizontal axis of shoulders in the image by matching, against a face area of the preexisting image when the face is detected by the object detection processing unit, a plurality of reference images in which tilt of a head with respect to shoulders is changed in stages while showing contour of the head and the shoulders so as to correct the relative angle at which the face is detected; and
an image-data updating unit configured to update the preexisting image so that the image is displayed in the identified orientation.
2. The image processing device according to claim 1 ,
wherein the object detection processing unit records, for each of a plurality of objects captured in the preexisting image, the relative angle at which a corresponding object is detected, and
wherein the image orientation identification unit identifies a correct orientation of the preexisting image based on an average value of respective relative angles recorded for the plurality of objects.
3. The image processing device according to claim 1 , wherein, if it is determined that a nightscape was captured in the preexisting image based on the brightness of the image and on an image-capturing time of the preexisting image that is recorded along with image capturing, the non-object image orientation identification unit changes the sign of the inclination of the brightness in the evaluation value to be opposite from that in other cases.
4. The image processing device according to claim 1 , wherein the image-data updating unit asks the user for confirmation of orientation by displaying on a display device an image-orientation confirmation screen displaying the preexisting image to be processed in an original orientation side by side with the preexisting image to be processed in the orientation identified by the object-image orientation identification unit so as to update the preexisting image upon receiving confirmation from the user.
5. The image processing device according to claim 1 , wherein the preexisting image to be processed is a frame image included in a moving image.
6. An image processing method comprising:
reading out data of a specified image containing an object from a computer readable non-transitory memory device and then detecting, by performing pattern matching between the image and a reference image of the object, the object in the image, while changing a relative angle of the images;
determining the top and the bottom of the image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected, so as to identify a correct orientation of the image;
detecting a face as the object
obtaining an angle between an axis of the face in a vertical direction and a horizontal axis of shoulders in the image by matching, against a face area of the preexisting image when the face is detected, a plurality of reference images in which tilt of a head with respect to shoulders is changed in stages while showing contour of the head and the shoulders so as to correct the relative angle at which the face is detected; and
updating the data of the image so that the image is displayed in the identified orientation.
7. A computer program embedded in a non-transitory computer readable recording medium, comprising:
an object detection processing module configured to read out data of a specified image containing an object from a memory device and then to detect, by performing pattern matching between the image and a reference image of the object, the object in the image, while changing a relative angle of the images;
an object-image orientation identification module configured to determine the top and the bottom of the image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected, so as to identify a correct orientation of the image;
wherein the object detection processing module performs a process for detecting a face as the object; and
an angle correction module configured to obtain an angle between an axis of the face in a vertical direction and a horizontal axis of shoulders in the image by matching, against a face area of the preexisting image when the face is detected by the object detection processing module, a plurality of reference images in which tilt of a head with respect to shoulders is changed in stages while showing contour of the head and the shoulders so as to correct the relative angle at which the face is detected; and
an image-data updating module configured to update the data of the image so that the image is displayed in the identified orientation.
8. A non-transitory computer-readable recording medium having embodied thereon a computer program product comprising:
an object detection processing module configured to read out data of a specified image containing an object from a memory device and then to detect, by performing pattern matching between the image and a reference image of the object, the object in the image, while changing a relative angle of the images;
an object-image orientation identification module configured to determine the top and the bottom of the image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected, so as to identify a correct orientation of the image;
wherein the object detection processing module performs a process for detecting a face as the object; and
an angle correction module configured to obtain an angle between an axis of the face in a vertical direction and a horizontal axis of shoulders in the image by matching, against a face area of the preexisting image when the face is detected by the object detection processing module, a plurality of reference images in which tilt of a head with respect to shoulders is changed in stages while showing contour of the head and the shoulders so as to correct the relative angle at which the face is detected; and
an image updating module configured to update the data of the image so that the image is displayed in the identified orientation.
9. An image processing device comprising:
an object detection processing unit configured to detect, by performing pattern matching between a preexisting image containing an object to be processed and a reference image of the object, the object in the image to be processed, while changing the relative angle of the images;
an object-image orientation identification unit configured to determine the top and the bottom of the preexisting image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected by the object detection processing unit, so as to identify a correct orientation of the preexisting image;
an image-characteristic analysis unit configured to acquire a frequency distribution and a brightness distribution of the preexisting image, if an object is not detected at any relative angle by the object detection processing unit, so as to acquire respective inclinations obtained when distribution characteristics with respect to a vertical axis and a horizontal axis of the preexisting image after approximating the respective distributions by respective straight lines; and
a non-object image orientation identification unit configured to determine the top and the bottom of the preexisting image, based on an evaluation value for each axis that is obtained by weighted summation of the respective inclinations of the frequency distribution characteristics and the brightness distribution characteristics both obtained by the image-characteristic analysis unit and based on the inclinations, so as to identify a correct orientation of the preexisting image; and
an image-data updating unit configured to update the preexisting image so that the image is displayed in the identified orientation.
10. An image processing method comprising:
reading out data of a specified image containing an object from a computer readable non-transitory memory device and then detecting, by performing pattern matching between the image and a reference image of the object, the object in the image, while changing a relative angle of the images;
determining the top and the bottom of the image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected, so as to identify a correct orientation of the image;
acquiring a frequency distribution and a brightness distribution of the preexisting image, if an object is not detected at any relative angle, so as to acquire respective inclinations obtained when distribution characteristics with respect to a vertical axis and a horizontal axis of the preexisting image after approximating the respective distributions by respective straight lines;
determining the top and the bottom of the preexisting image, based on an evaluation value for each axis that is obtained by weighted summation of the respective inclinations of the frequency distribution characteristics and the brightness distribution characteristics and based on the inclinations, so as to identify a correct orientation of the preexisting image; and
updating the data of the image so that the image is displayed in the identified orientation.
11. A computer program embedded in a non-transitory computer readable recording medium, comprising:
an object detection processing module configured to read out data of a specified image containing an object from a memory device and then to detect, by performing pattern matching between the image and a reference image of the object, the object in the image, while changing a relative angle of the images;
an object-image orientation identification module configured to determine the top and the bottom of the image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected, so as to identify a correct orientation of the image;
an image-characteristic analysis module configured to acquire a frequency distribution and a brightness distribution of the preexisting image, if an object is not detected at any relative angle by the object detection processing module, so as to acquire respective inclinations obtained when distribution characteristics with respect to a vertical axis and a horizontal axis of the preexisting image after approximating the respective distributions by respective straight lines;
a non-object image orientation identification module configured to determine the top and the bottom of the preexisting image, based on an evaluation value for each axis that is obtained by weighted summation of the respective inclinations of the frequency distribution characteristics and the brightness distribution characteristics both obtained by the image-characteristic analysis module and based on the inclinations, so as to identify a correct orientation of the preexisting image; and
an image-data updating module configured to update the data of the image so that the image is displayed in the identified orientation.
12. A non-transitory computer-readable recording medium having embodied thereon a computer program product comprising:
an object detection processing module configured to read out data of a specified image containing an object from a memory device and then to detect, by performing pattern matching between the image and a reference image of the object, the object in the image, while changing a relative angle of the images;
an object-image orientation identification module configured to determine the top and the bottom of the image corresponding to the top and the bottom of the object, based on the relative angle at which the object is detected, so as to identify a correct orientation of the image;
an image-characteristic analysis module configured to acquire a frequency distribution and a brightness distribution of the preexisting image, if an object is not detected at any relative angle by the object detection processing module, so as to acquire respective inclinations obtained when distribution characteristics with respect to a vertical axis and a horizontal axis of the preexisting image after approximating the respective distributions by respective straight lines;
a non-object image orientation identification module configured to determine the top and the bottom of the preexisting image, based on an evaluation value for each axis that is obtained by weighted summation of the respective inclinations of the frequency distribution characteristics and the brightness distribution characteristics both obtained by the image-characteristic analysis module and based on the inclinations, so as to identify a correct orientation of the preexisting image; and
an image updating module configured to update the data of the image so that the image is displayed in the identified orientation.