Method and device for combining at least two images to form a panoramic image
AssigneeDiehl Bgt Defence GmbH & Co. KG
InventorFlorian Gagel
Life of this patent
Estimating the term…
Abstract
In order to make it possible to reduce stereographic disturbances without image processing methods, it is proposed that pixels in the images each be associated with a distance from a distance model. The distance is an assumed distance from an object, which is imaged at that pixel, to at least one predetermined point, the distance model associates the distance in dependence on the position of the pixels in the recorded image. The image fusion is carried out using the associated distances.
Claims20
A method for image fusion of at least two images for forming a panoramic image, which comprises the steps of:
- providing a processing device, the processing device performing the further steps of:
- associating pixels in the two images each with a distance from a distance model, the distance is an assumed distance from an object, being imaged at a pixel, to at least one predetermined point, the distance model associating the distance in dependence on a position of the pixels in the images resulting in associated distances, wherein the distance model assumes an at least generally planar topography, such that, because of a known height of a camera above the planar topography, the distance to the object being imaged at the pixel can be deduced from an elevation angle of the pixel in the image, wherein the pixels in the two images which are combined to form one pixel in the panoramic image are those which are associated with a same distance;
- and
- carrying out the image fusion using the associated distances.
An apparatus for image fusion, comprising:
- at least two cameras spaced at a distance from one another, for recording a surrounding area from different perspectives;
- an image display for displaying a panoramic image;
- and
- a processing device for fusing an image from a first of said cameras with an image from a second of said cameras to form the panoramic image, with pixels in the two images each being associated with a distance from a distance model, with the distance being an assumed distance from an object, being imaged at the pixel, to at least one predetermined point, with the distance model associating the distance in dependence on a position of the pixels in the image recorded, wherein the distance model assumes an at least generally planar topography, such that, because of a known height of said cameras above the planar topography, the distance to the object being imaged at the pixel can be deduced from an elevation angle of the pixel in the image, wherein the pixels in the two images which are combined to form one pixel in the panoramic image are those which are associated with a same distance, and with the image fusion being carried out using the associated distances.
Dependencies are inferred from the claim text.