Multifocus Image Fusion

 

Multifocus Image fusion is process of combining information of two or more images of a scene and as a result has "all-in-focus" image.

When one scene contains objects in different distance, the camera can be focused on each object one after the other, creating set of pictures. Then, using image fusion technique, an image with better focus accross all area can be generated. There are many multifocus image fusion methods, today. One of them is Empirical Mode Decomposition based multifocus image fusion.

Fig.1 Example of two defocused images from in-house dataset.

Dataset

One of the issues in this research was creating multifocus image dataset, that will be available for testing fusion methods. For that purpose, we used Nikon D5000 camera, and created in-house multifocus image dataset. All images are in .jpg format.

For our testing process, we used 27 pairs of images, where 21 of them are from in-house dataset, while 6 pairs are standard for testing multifocus image fusion methods. Standard images are in .bmp format.

 

All 27 pairs of defocused images are creating TestingImageDataset.

The images provided here are for research purposes only.

 

Evaluation

Example of All-in-focus image using proposed image fusion algorithm.

Fig.2 All-in-focus image.

Papers

Slavica Savic, "Multifocus Image Fusion Based on Empirical Mode Decomposition", Twentieth International Electrotechnical and Computer Science Conference, ERK 2011 (.pdf)