Image fusion is the process of combining relevant information in two or more images of a scene into a single
highly informative image. Depending on what is considered relevant information, the images can be combined
differently. Using image contrast as the relevant information, multi-focus images can be combined into a sharp
image.
highly informative image. Depending on what is considered relevant information, the images can be combined
differently. Using image contrast as the relevant information, multi-focus images can be combined into a sharp
image.
Sharp images contain more information than blurred images. Often due to great variations in a scene's depth, it
is not possible to capture an image where all areas of the scene appear sharp. Only scene areas that are at the
focus plane appear sharp, and areas in front of or behind the focus plane appear blurred. The ability to create
a single image where all scene areas appear sharp is desired not only in digital photography but it is also
needed in various machine vision applications.
A scene area at and near the focus plane appears sharp in an image and has a higher contrast. The algorithm
used to identify the best-exposed images within an area of interest may be used to identify the best-focused
image within an area of interest. The image domain is subdivided into equal regions and the image appearing
the sharpest within each region is selected and blended together to create an image that is sharp everywhere.
Figs. 1a and 1b show two images of a newspaper scene obtained at different focus levels. By fusing the
images, the image shown in Fig 2c is obtained. Figs. 2a and 2b show two images of books taken by a camera at
different distances, and Fig. 2c shows the fused image. Note that if an area appears blurred in all images, the
area will appear blurred in the fused process. The fusion process does not sharpen an image. For image
sharpening, see the image enhancement page.
is not possible to capture an image where all areas of the scene appear sharp. Only scene areas that are at the
focus plane appear sharp, and areas in front of or behind the focus plane appear blurred. The ability to create
a single image where all scene areas appear sharp is desired not only in digital photography but it is also
needed in various machine vision applications.
A scene area at and near the focus plane appears sharp in an image and has a higher contrast. The algorithm
used to identify the best-exposed images within an area of interest may be used to identify the best-focused
image within an area of interest. The image domain is subdivided into equal regions and the image appearing
the sharpest within each region is selected and blended together to create an image that is sharp everywhere.
Figs. 1a and 1b show two images of a newspaper scene obtained at different focus levels. By fusing the
images, the image shown in Fig 2c is obtained. Figs. 2a and 2b show two images of books taken by a camera at
different distances, and Fig. 2c shows the fused image. Note that if an area appears blurred in all images, the
area will appear blurred in the fused process. The fusion process does not sharpen an image. For image
sharpening, see the image enhancement page.
| (a) (b) |
(c)
Fig. 1. (a), (b) Two images of a newspaper scene obtained at different focus levels. (c) The result of fusing
images (a) and (b).
images (a) and (b).
(a)
(b)
(c)
Fig. 2. (a), (b) Two images of books placed at different distances to the camera and obtained at different focus
levels. Each image was obtained by focusing the camera on one of the books. (c) The result of fusing images
(a) and (b). In the fused image both books appear sharp.
levels. Each image was obtained by focusing the camera on one of the books. (c) The result of fusing images
(a) and (b). In the fused image both books appear sharp.
This image fusion software reads in a set of images and combines them into a single high-contrast image. The
relevant information used to combine the images is image contrast.
relevant information used to combine the images is image contrast.
To obtain a license for this multi-focus image fusion software, follow this link =>
 | |  |
| Multi-focus image fusion |
| Image Registration and Fusion Systems |
