IMAGE PROCESSOR, IMAGE PROCESSING METHOD, AND DIGITAL CAMERA

• US 20140085498A1
• Filed: 11/25/2013
• Published: 03/27/2014
• Est. Priority Date: 05/31/2011
• Status: Active Grant

• Current Assignee: Panasonic Semiconductor Solutions Co., Ltd. (Winbond Electronics Corporation)
• Original Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Inventors: HATANO, Toshinobu
• Granted Patent: US 8,995,792 B2
• Time in Patent Office: Days
• US Class Current: 348/221.1
• CPC Class Codes: H04N 19/423 characterised by memory arrangements H04N19/433 takes precedence
• CPC Class Codes: H04N 19/436 using parallelised computational arrangements
• CPC Class Codes: H04N 19/59 involving spatial sub-sampling or interpolation, e.g. alteration of picture size or resolution
• CPC Class Codes: H04N 19/85 using pre-processing or post-processing specially adapted for video compression
• CPC Class Codes: H04N 23/12 with one sensor only
• CPC Class Codes: H04N 23/61 based on recognised objects
• CPC Class Codes: H04N 23/611 where the recognised objects include parts of the human body
• CPC Class Codes: H04N 23/67 Focus control based on electronic image sensor signals
• CPC Class Codes: H04N 23/69 Control of means for changing angle of the field of view, e.g. optical zoom objectives or electronic zooming
• CPC Class Codes: H04N 23/843 Demosaicing, e.g. interpolating colour pixel values
• CPC Class Codes: H04N 25/133 including elements passing panchromatic light, e.g. filters passing white light
• CPC Class Codes: H04N 25/134 based on three different wavelength filter elements
• CPC Class Codes: H04N 25/136 using complementary colours
• CPC Class Codes: H04N 9/70 for colour killing

Abstract

A digital camera includes: a horizontal reduction (resizing) processor 301 for reducing a RAW image from single-sensor color imaging device to an image corresponding to a video recording size in an input line direction; a memory section 303 storing horizontally resized image data; a plurality of vertical reduction (resizing) processors 304, 306, and 308 reducing (resizing), in a vertical direction orthogonal to the input line direction, a plurality of pieces of reduced line data read out from the memory section 303; and horizontal reduction (resizing) processors 310 and 312 reducing a plurality of images reduced (resized) in the horizontal and vertical directions back into images of a display size and a face detection size in the input line direction.

Claims

1. An image processor for use in an imaging system that generates at least one of a still image or a video image from an original image output from an imaging device and having pixels of a plurality of colors arranged in a color array with a periodicity and records the at least one of the still image or the video image, the image processor comprising:

a pretreatment section configured to generate a plurality of reduced-size images with respect to the original image;

a memory section configured to hold the reduced-size images temporarily via a memory control section;

a face information detector configured to detect face information via the memory control section with respect to one of the reduced-size images read out from the memory section; and

a central processing unit configured to perform at least one of focus control or exposure control for a specific area based on the face information detected by the face information detector, wherein

the pretreatment section includes

a first horizontal reduction resizing processor configured to reduce, in an input line direction, the original image to a reduced-size image of an arbitrary size, and to store the reduced-size image of the arbitrary size temporarily,

a plurality of vertical reduction resizing processors each configured to read the reduced-size image reduced in the input line direction from the memory section by the first horizontal reduction resizing processor, and to reduce the reduced-size image to an image of an arbitrary size in a vertical direction orthogonal to the input line direction, and

a plurality of second horizontal reduction resizing processors each configured to reduce reduced-size images reduced in size in the horizontal and vertical directions and obtained by the vertical reduction resizing processors again to images of an arbitrary size in the input line direction.

2. The image processor of claim 1, wherein

when the pretreatment section generates n types, where n is an integer of four or more, of reduced-size images,

the first horizontal reduction resizing processor reduces the original image to an image of a horizontally reduced size of one of the n-types of reduced-size images having a largest horizontal image size, with a ratio K1,

one of the vertical reduction resizing processors reduces the reduced-size image reduced with the ratio K1 in the input line direction, with the ratio K1 in the vertical direction,

another of the vertical reduction resizing processors reduces the reduced-size image reduced with the ratio K1 in the input line direction, with a ratio K2 in the vertical direction,

yet another of the vertical reduction resizing processors reduces the reduced-size image reduced with the ratio K1 in the input line direction, with a ratio K3 in the vertical direction,

each of the other (n−

3) vertical reduction resizing processors reduces the reduced-size image reduced with the ratio K1 in the input line direction, with a ratio Kn in the vertical direction,

the number of the second horizontal reduction resizing processors is n−

1,

one of the second horizontal reduction resizing processors reduces the reduced-size image reduced with the ratio K2 in the vertical direction, with a coefficient M2 satisfying a relationship of K1*M2=K2 in the input line direction,

another one of the second horizontal reduction resizing processors reduces the reduced-size image reduced with the ratio K3 in the vertical direction, with a coefficient M3 satisfying a relationship of K1*M3=K3 in the input line direction, and

each of the other second horizontal reduction resizing processors reduces the reduced-size image reduced with the ratio Kn in the vertical direction, with a coefficient Mn satisfying a relationship of K1*Mn=Kn in the input line direction.

3. The image processor of claim 1, wherein

when the pretreatment section generates a reduced-size image for face detection from the original image, the pretreatment section generates a reduced image of a luminance signal component having no color carrier components by performing:

a color carrier removal process in the input line direction; and

a color carrier removal process in the vertical direction.

4. The image processor of claim 1, wherein

among the plurality of reduced-size images generated by the pretreatment section, an image having a largest horizontal image size is an image for video recording conforming to a high-definition video standard, and

each of the other reduced-size image(s) is an image for display or image recognition such as face recognition.

5. The image processor of claim 1, wherein

the original image is an image obtained by using a single-sensor color imaging device and has a plurality of pixels of colors arranged in a color array with a periodicity through a primary color filter.

6. The image processor of claim 1, wherein

the pretreatment section generates a plurality of reduced-size images in a first frame period in which the original image is input to the image processor, and

the face information detector detects face information in a second frame period subsequent to the first frame period.

13. A digital camera comprising the image processor of claim 1.

7. An image processing method for use in an imaging system that generates at least one of a still image or a video image from an original image output from an imaging device and having pixels of a plurality of colors arranged in a color array with a periodicity and records the at least one of the still image or the video image, the method comprising:

a pretreatment step of generating a plurality of reduced-size images with respect to the original image;

a face information detection step of detecting face information with respect to one of the reduced-size images; and

an imaging control step of performing at least one of focus control or exposure control for a specific area based on the face information detected in the face information detection step, wherein

the pretreatment step include

a first horizontal reduction resizing process of reducing, in an input line direction, the original image to an image of an arbitrary size,

a plurality of vertical reduction resizing processes of each reducing the reduced-size image that has been reduced in the input line direction in the first horizontal reduction resizing process to an image of an arbitrary size in a vertical direction orthogonal to the input line direction, and

a plurality of second horizontal reduction resizing processes of each reducing reduced-size images that have been reduced in size in the horizontal and vertical directions and obtained in the vertical reduction resizing processes again to images of an arbitrary size in the input line direction.

8. The image processing method of claim 7, wherein

when n-types, where n is an integer of four or more, of reduced-size images are generated in the pretreatment step,

in the first horizontal reduction resizing process, the original image is reduced to an image of a horizontally reduced size of one of the n-types of reduced-size images having a largest horizontal image size, with a ratio K1,

in one of the vertical reduction resizing processes, the reduced-size image that has been reduced with the ratio K1 in the input line direction, is reduced with the ratio K1 in the vertical direction,

in another of the vertical reduction resizing processes, the reduced-size image that has been reduced with the ratio K1 in the input line direction is reduced with a ratio K2 in the vertical direction,

in yet another of the vertical reduction resizing processes, the reduced-size image that has been reduced with the ratio K1 in the input line direction is reduced with a ratio K3 in the vertical direction,

in each of the other (n−

3) vertical reduction resizing processes, the reduced-size image that has been reduced with the ratio K1 in the input line direction is reduced with a ratio Kn in the vertical direction,

the number of the second horizontal reduction resizing processes is n−

1,

in one of the second horizontal reduction resizing processes, the reduced-size image that has been reduced with the ratio K2 in the vertical direction is reduced in the input line direction with a coefficient M2 satisfying a relationship of K1*M2=K2,

in another of the second horizontal reduction resizing processes, the reduced-size image that has been reduced with the ratio K3 in the vertical direction is reduced in the input line direction with a coefficient M3 satisfying a relationship of K1*M3=K3,

in each of the other (n−

3) second horizontal reduction resizing processes, the reduced-size image that has been reduced with the ratio Kn in the vertical direction is reduced in the input line direction with a coefficient Mn satisfying a relationship of K1*Mn=Kn.

9. The image processing method of claim 7, wherein

in the pretreatment step,

when generating a reduced-size image for face detection from the original image, a reduced-size image of a luminance signal component having no color carrier components is generated by performing:

a color carrier removal process concurrently with the first horizontal reduction resizing process; and

a color carrier removal process concurrently with the vertical reduction resizing process.

10. The image processing method of claim 7, wherein

among the plurality of reduced-size images generated in the pretreatment step, an image having a largest horizontal image size is an image for video recording conforming to a high-definition video standard, and

each of the other reduced-size image(s) is an image for display or image recognition such as face recognition.

11. The image processing method of claim 7, wherein

the original image is an image obtained by using a single-sensor color imaging device and has a plurality of pixels of colors arranged in a color array with a periodicity through a primary color filter.

12. The image processing method of claim 7, wherein

in the pretreatment step, a plurality of reduced-size images are generated in a first frame period in which the original image is input in the pretreatment step, and

in the face information detection step, face information is detected in a second frame period subsequent to the first frame period.