Image processing apparatus, image processing method, and program

• US 8,860,784 B2
• Filed: 08/04/2009
• Issued: 10/14/2014
• Est. Priority Date: 08/06/2008
• Status: Expired due to Fees

• Current Assignee: Saturn Licensing LLC (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Kobayashi, Seiji
• Primary Examiner(s): Czekaj, Dave
• Assistant Examiner(s): Beck, Leron
• Application Number:
  US12/535,431
• Publication Number:
  US 20100033554A1
• Time in Patent Office: 1,897 Days
• Field of Search: 348/43, 345/419, 386/154
• US Class Current: 348/43
• CPC Class Codes: H04N 13/261 with monoscopic-to-stereoscopic image conversion
• CPC Class Codes: H04N 13/324 Colour aspects

Abstract

A image processing apparatus including: an image input section inputting a two-dimensional image signal; an image conversion section inputting an image signal output from the image input section, and generating and outputting images for right and left eyes for allowing binocular stereovision; and an image output section outputting the images for right and left eyes output from the image conversion section, wherein the image conversion section extracts a spatial feature quantity of the input image signal, and performs image conversion processing conducting enhancement processing applying the feature quantity to the input image signal so that either an image for a right eye or an image for a left eye is generated.

Claims

1. An image processing apparatus comprising:

an image conversion section inputting an image signal of an original signal, and generating and outputting images for right and left eyes for allowing binocular stereovision; and

an image output section outputting the images for right and left eyes output from the image conversion section,

wherein the image conversion section extracts a spatial feature quantity of the input image signal, and performs image conversion processing comprising applying the extracted spatial feature quantity to the input image signal so that either an image for a right eye or an image for a left eye is generated,

the extracted spatial feature quantity comprises one of;

a differential luminance signal of the input image signal, or a signal produced by non-linearly converting the differential luminance signal,

the image conversion section generates a signal by at least adding the extracted spatial feature quantity to the input image signal,

due to adding the extracted spatial feature quantity to the input image signal, a simulated local depth perceived as a result of local retinal disparity between the images for right and left eyes is a function of local spatial frequency of the original image, such that the higher the local spatial frequency, the smaller the local retinal disparity becomes, and the lower the local spatial frequency, the larger the local retinal disparity becomes, wherein areas of higher spatial frequencies are stereoscopically perceived to be in front of areas of lower spatial frequencies; and

wherein the local retinal disparity is a function of an annular frequency of the input image signal.

2. The image processing apparatus according to claim 1,

wherein the differential luminance signal is the extracted spatial feature quantity, and the signal generated by the image conversion section is a converted image signal for a right eye or a left eye, and

wherein the image conversion section outputs a non-converted signal as the direct input image signal as an image signal for a different one of the eyes from the eye for which the converted signal is produced.

3. The image processing apparatus according to claim 1,

wherein the differential luminance signal is the extracted spatial feature quantity, and the image conversion section further generates another signal by subtracting the extracted spatial feature quantity from the input image signal, thereby generating a pair of the two signals as a pair of images for a right eye and for a left eye.

4. The image processing apparatus according to claim 1,

wherein the signal produced by non linearly converting the differential luminance signal is the extracted spatial feature quantity, and the image conversion section produces the generated signal as an image signal for a right eye or an image signal for a left eye.

5. The image processing apparatus according to claim 1,

wherein the image conversion section generates an image for a right eye and an image for a left eye for each frame included in a moving image.

6. The image processing apparatus according to claim 5, further comprising an image output section outputting the images for right and left eyes, generated by the image conversion section, wherein the image output section outputs the image for a right eye and the image for a left eye, generated by the image conversion section, alternately at a speed two times an input-image frame rate.

7. The image processing apparatus according to claim 1, wherein the image conversion section alternately generates either an image for a right eye or an image for a left eye for each frame included in a moving image.

8. The image processing apparatus according to claim 1,

wherein the image conversion section generates an image for a right eye and an image for a left eye for each frame included in a moving image, and performs generation processing of a binocular disparity image alternately including line data included in the generated images for right and left eyes.

9. The image processing apparatus according to claim 1,

wherein the image conversion section performs generation processing of an image for a right eye and an image for a left eye so as to set a sum signal of the image for a right eye and the image for a left eye equal to the input signal, or approximately equal to the input signal such that viewing of the image for a right eye and the image for a left eye, without stereoscopic glasses, does not visually differ from viewing the original image.

10. The image processing apparatus according to any one of claims 1 to 9, further comprising an image display section displaying the image generated by the image conversion section.

11. The image processing apparatus according to claim 10,

wherein the image display section performs time division stereoscopic display processing alternately outputting the image for a right eye and the image for a left eye.

12. The image processing apparatus according to claim 11,

wherein, when the image display section performs time division stereoscopic display processing alternately outputting the image for a right eye and the image for a left eye, the image display section switches the display such that switching timing of outputting the image for a right eye and the image for a left eye is synchronized with switching of shutters of right and left glasses worn by an observer of the image.

13. The image processing apparatus according to claim 10,

wherein the image display section has a configuration in which a polarizing filter whose polarization direction is set different for each horizontal line is attached on a front surface of the display section, and displays a binocular disparity image including line data alternately forming the image for a right eye and the image for a left eye generated by the image conversion section.

14. A method of processing an image in an image processing apparatus, comprising the steps of:

inputting a two-dimensional image signal of an original image;

converting the input image signal to generate images for right and left eyes for allowing binocular stereovision; and

outputting the images for right and left eyes, wherein the step of converting the input image signal comprises;

extracting a spatial feature quantity of the input image signal, and

performing image conversion processing comprising applying the extracted spatial feature quantity to the input image signal so that either the image for a right eye or the image for a left eye is generated,

wherein the extracted spatial feature quantity comprises one of;

a differential luminance signal of the input image signal, or a signal produced by non-linearly converting the differential luminance signal,

the image conversion section generates a signal by at least adding the extracted spatial feature quantity to the input image signal, and

due to adding the extracted spatial feature quantity to the input image signal, a simulated local depth perceived as a result of local retinal disparity between the images for right and left eyes is a function of local spatial frequency of the original image, such that the higher the local spatial frequency, the smaller the local retinal disparity becomes, and the lower the local spatial frequency, the smaller the local retinal disparity becomes, wherein areas of higher spatial frequencies are stereoscopically perceived to be in front of areas of lower spatial frequencies; and

wherein the local retinal disparity is a function of an annular frequency of the input image signal.

15. A non-transitory computer readable medium comprising program code for causing a computer to perform steps comprising:

inputting a two-dimensional image signal of an original image;

converting the input image to generate images for right and left eyes for allowing binocular stereovision; and

outputting the images for right and left eyes, wherein the step of converting the input image comprises;

extracting a spatial feature quantity of the input image signal, and

performing image conversion processing comprising applying the extracted spatial feature quantity to the input image signal so that either the image for a right eye or the image for a left eye is generated,

wherein the extracted spatial feature quantity comprises one of;

a differential luminance signal of the input image signal, or a signal produced by non-linearly converting the differential luminance signal,

the image conversion section generates a signal by at least adding the extracted spatial feature quantity to the input image signal, and

due to adding the extracted spatial feature quantity to the input image signal, a stimulated local depth perceived as a result of local retinal disparity between the images for right and left eyes is a function of local spatial frequency of the original image, such that the higher the local spatial frequency, the smaller the local retinal disparity becomes, and the lower the local spatial frequency, the larger the local retinal disparity becomes, wherein areas of higher spatial frequencies are stereoscopically perceived to be in front of areas of lower spatial frequencies; and

wherein the local retinal disparity is a function of an annular frequency of the input image signal.