Topic 7: Digital extended depth of focus¶

1. Read and discuss the paper by Liu et al.
2. Describe the SIFT image descriptor.
3. Try to acquire your own focus stack with the aim of extending the depth of focus.
4. Use two of the objective evaluation metrics in section 4.1.3. of the paper to evaluate your results.
5. Additionally, you can generate you own dataset with groundtruth images if you have access to a camera/lens with deep/shallow depth of focus (cameras need you allow to change the aperture in order to this - it is called the f-number on cameras often). For this purpose, you first acquire test images with objects and background in-focus and out-of-focus, respectively. Then, you also acquire an image with deep depth of focus. This image serves as your ground-truth image. You apply your code on test images and then compare it with the groundtruth.

References

Yu Liu, Shuping Liu, Zengfu Wang, Multi-focus image fusion with dense SIFT, Information Fusion, Volume 23, 2015, Pages 139-155, ISSN 1566-2535.