Method and device for fusion of images
A method of fusing images includes obtaining an optical image of a first scene with a first camera. A thermal image of the first scene is obtained with a second camera. The optical image is fused with the thermal image to generate a fused image.
1. A method of fusing images comprising:
obtaining, by a central processing unit (CPU) of a controller, an optical image of a first scene with a first camera;
obtaining, by the CPU, a thermal image of the first scene with a second camera;
fusing, by the CPU, the optical image with the thermal image to generate a fused image, wherein fusing the optical image with the thermal image includes cropping a portion of at least one of the optical image or the thermal image so as to generate a cropped version of the at least one of the optical image or the thermal image which does not include the cropped portion, the cropped portion corresponding to a portion of the at least one of the optical image or the thermal image which is not included in the other of the at least one of the optical image of the thermal image, and using the cropped version of the at least one of the optical image or the thermal image to generate the fused image,
wherein fusing the optical image with the thermal image includes matching pixels from the optical image with corresponding pixels from the thermal image on a pixel-by-pixel basis.
2. The method of claim 1 , wherein the fused image includes a field of view less than or equal to a field of view of the optical image and a field of view of the thermal image.
3. The method of claim 1 , wherein the optical image includes a different perspective of the first scene from the thermal image.
4. The method of claim 1 , wherein the optical image includes different distortions of objects in the first scene from the thermal image.
5. The method of claim 1 , wherein at least one object appears in one of the thermal image or the optical image but not the other of the thermal image or the optical image.
6. The method of claim 1 , wherein fusing the optical image with the thermal image is performed by an algorithm on a central processing unit of a controller.
7. The method of claim 6 , wherein the algorithm relies on machine learning for fusing the optical image with the thermal image.
8. The method of claim 7 , wherein the machine learning occurs on the CPU.
9. The method of claim 1 , including performing object detection on the fused image to identify at least one relevant portion of the fused image.
10. The method of claim 1 , wherein the first scene includes low light conditions.
11. An assembly for capturing and fusing images comprising:
an optical camera;
a thermal camera;
a controller comprising a central processing unit (CPU) in communication with the optical camera and the thermal camera, wherein the controller is configured for:
obtaining, by the CPU, an optical image of a first scene with a first camera;
obtaining, by the CPU, a thermal image of the first scene with a second camera; and
fusing, by the CPU, the optical image with the thermal image to generate a fused image, wherein fusing the optical image with the thermal image includes cropping a portion of at least one of the optical image or the thermal image to generate a cropped version of the at least one of the optical image or the thermal image, the cropped portion corresponding to a portion of the at least one of the optical image or the thermal image which is not included in the other of the at least one of the optical image or the thermal image, and using the cropped version of the at least one of the optical image or the thermal image to generate the fused image,
wherein fusing the optical image with the thermal image includes matching pixels from the optical image with corresponding pixels from the thermal image on a pixel-by-pixel basis.
12. The assembly of claim 11 , wherein the fused image includes a field of view less than or equal to a field of view of the optical image and a field of view of the thermal image.
13. The assembly of claim 11 , wherein fusing the optical image with the thermal image is performed by an algorithm executed by the CPU of the controller.
14. The assembly of claim 13 , wherein the algorithm relies on machine learning for fusing the optical image with the thermal image.
15. The assembly of claim 11 , including performing object detection on the fused image to identify at least one relevant portion of the fused image.
16. The assembly of claim 15 , wherein performing the object detecting occurs on a graphics processing unit.