Traitement en cours

Veuillez attendre...

Paramétrages

Paramétrages

Aller à Demande

1. WO2020117379 - ANTI-DÉDOUBLEMENT D'IMAGE ET FUSION À PLAGE DYNAMIQUE ÉLEVÉE

Note: Texte fondé sur des processus automatiques de reconnaissance optique de caractères. Seule la version PDF a une valeur juridique

[ EN ]

What is claimed is:

1. A method comprising:

receiving input images from image sensors in a linear domain, each input image having a different exposure time;

aligning, in the linear domain, in luminance by applying a gain factor to appropriate input images so same objects with different exposure times have a same grey level;

blending, in the linear domain, luminance values at each pixel from each of the aligned images to generate a blended image;

determining, in the linear domain, a useful image by selecting which of the aligned images has useful information for a pixel;

calculating, in the linear domain, a differential luminance value from the aligned images for the pixel;

determining, in the linear domain, an anti-ghosting weight based on the useful image and the differential luminance value for the pixel;

applying, in the linear domain, proportionally the determined anti-ghosting weight to the pixel for each of the aligned images to generate an output image; and

storing, displaying, or transmitting the output image based on at least the anti ghosting weight.

2. The method of claim 1, the method further comprising:

regularizing, in the linear domain, the anti -ghosting weight for the pixel.

3. The method of any of claims 1-2, the method further comprising:

computing the luminance values for each pixel in each aligned image.

4. The method of any of claims 1-3, wherein the determining a useful image further comprising:

using a threshold based on the gain factor and the blended image to determine the useful image.

5. The method of any of claims 1-4, wherein the useful image represents the aligned image that has a largest dynamic range for the pixel.

6. The method of any of claims 1-5, wherein when the useful image is multiple aligned images, the method further comprising:

averaging luminance values from each of the aligned images.

7. The method of any of claims 1-6, wherein a luminance value for the pixel is from the useful image.

8. The method of any of claims 1-7, wherein the determining, in the linear domain, an anti-ghosting weight further comprising:

accessing a look-up table (LUT) using a luminance value from the useful image and the differential luminance value to locate the anti-ghosting weight.

9. The method of any of claims 1-8, the method further comprising:

populating the LUT with anti-ghosting weights during image exposure.

10. The method of any of claims 1-9, wherein a number of entries in the LUT are delimited by a defined degree of interpolation between entries and real-time execution.

11. A system, comprising:

image sensors configured to capture images; and

an image processor configured to receive the captured images and comprised of: an anti-ghosting unit configured to determine an anti-ghosting weight based on a useful image and a differential luminance value for a pixel, wherein the useful image is a captured image that has a largest dynamic range for the pixel and the differential luminance value is calculated from the luminance values of the captured images; and

a blending unit configured to use the anti -ghosting weight to fuse the captured images into an output high dynamic range image; and

the image processor further configured to store, display, or transmit the output high dynamic range image based on at least the anti-ghosting weight.

12. The system of claim 11, the image processor further comprising:

at least one alignment unit configured to normalize grey level of same objects in the captured images using gain factors;

a look-up table (LUT), the look-up table including anti-ghosting weights, wherein the anti-ghosting unit is further configured to access the LUT using a luminance value associated with the useful image and the differential luminance value to locate the anti-ghosting weight; a luminance computation unit configured to compute the luminance values of the captured images; and

a luminance blending unit configured to generate a blended image based on the luminance values.

13. A method comprising:

receiving input images from image sensors in a linear domain, each input image having different exposures or gains;

blending, in the linear domain, luminance values at each pixel from each of the input images to generate a blended image;

selecting, in the linear domain, a useful image based on degree of useful information for a pixel;

calculating, in the linear domain, a differential luminance value from the input images for the pixel;

locating, in the linear domain, from a look-up table (LUT), an anti-ghosting weight using the useful image and the differential luminance value for the pixel;

proportionally applying, in the linear domain, the located anti-ghosting weight to the pixel for each of the input images to generate an output image; and

storing, displaying, or transmitting the output image based on at least the anti ghosting weight.

14. A method comprising:

receiving input images from image sensors in a linear domain, each input image having a different exposure time;

aligning, in the linear domain, in luminance by applying a gain factor to appropriate input images so same objects with different exposure times have a same grey level;

blending, in the linear domain, luminance values at each pixel from each of the aligned images to generate a blended image;

determining, in the linear domain, a useful image by selecting which of the aligned images has useful information for a pixel;

calculating, in the linear domain, a distance value from the aligned images for the pixel;

determining, in the linear domain, an anti -ghosting weight based on the useful image and the distance value for the pixel;

applying, in the linear domain, proportionally the determined anti-ghosting weight to the pixel for each of the aligned images to generate an output image; and

storing, displaying, or transmitting the output image based on at least the anti ghosting weight.

15. A system, comprising:

image sensors configured to capture images; and

an image processor configured to receive the captured images and comprised of: an anti-ghosting unit configured to determine an anti-ghosting weight based on a useful image and a distance value for a pixel, wherein the useful image is a captured image that has a largest dynamic range for the pixel and the distance value is calculated from defined image properties of the captured images; and

a blending unit configured to use the anti -ghosting weight to fuse the captured images into an output high dynamic range image; and

the image processor further configured to store, display, or transmit the output high dynamic range image based on at least the anti-ghosting weight.