Centre of Computer Graphics and Data Visualisation

home
people
grants
research
publications
education
seminars
vacancies
informal

Digital Holography

Result of the Month | Introduction | People | Developed Methods | Publications

Result of the Month

Speckle based holography

Figure 1: Five strips in the image show how a tiny diffuse circle emits light; in a distance z well known speckle pattern appears. Although speckle is the enemy number one in holography, we can let it do the work. Speckle pattern of a single patch can be estimated quite easily - and if we can split a scene into individual patches, we can estimate light emitted from the scene as well. The image in the corner shows the numerical reconstruction of a computer generated hologram of a 3D scene split into such patches; the principle works, although a lot of work has to be yet done. (October 2010)

Older results of the month can be found here.

Introduction

A holography is an alternative way of taking pictures of real world objects. It uses the in fact same films like photography but the special light setup, image formation and developing process make the objects in the taken pictures look solid and three dimensional. The three dimensional perception is very realistic and natural. As a result, there are efforts to exploit the holographic technology to create 3D display. Sadly, the holographic technology application constitutes difficulties that need to be resolved first.

The holography is a discipline that relays on the wave nature of light. A hologram is a recording of light structure leaving the recorded scene. The hologram also allows to reconstruct the light according to the recording, see Figure 2. The hologram contains a range of view directions and a range of focal depths.

Hologram Recording/Replay

Figure 2: A comparison of obeserving an object directly and through a hologram. (a) An object reflects light (the rings represents individual wavefronts) and it is registered by a camera/viewer. (b) Reflected light structure is recorded on a hologram (notice that there is no objective in front of the hologram plate). (c) The light structure is replayed from the hologram and the camera/viewer sees the very same image as the one seen without the hologram.

For explaining how the holographic displaying works let us revise the current 2D imaging process illustrated in the Figure 3 first. There are two possible ways of acquiring an image of some object (a lovely flower in our case). The first way is to take a photo of a real world object. If we use an old fashion analog camera we capture the image onto a photographic film. We need to digitise (using a scanner) the image before we are able to display it on an LCD. The digitisation is not necessary if a modern digital camera is used. The digitised image is then stored in a computer in a form of bitmap image which then sends it to an LCD for displaying. An alternative and more flexible way of acquiring an image is to create a virtual model of the object (flower) and render its image using the computer graphics approach.

2D imaging process

Figure 3: Image acquiring and displaying.

The holography based imaging process follows exactly the same schema. The only difference is the lighting conditions used when hologram of a real world object is taken (LASER light is required) and the holographic camera construction which requires the lens to be removed (it means a ruined picture in a case of photography but in a case of holography it is essential). Also acquiring a hologram of a non-existing is possible. The geometric model is created first and a hologram generation algorithm is used for computing the hologram (digital hologram). Finally, the hologram, either taken from a real object or calculated, is then sent to the holographic display for viewing.

So where are the difficulities? Since hologram contains more information than a photo does the digitisation must be done using much higher sampling rate. While it is sufficient to digitise a photo with the rate of 92 samples per one inch (92 DPI) hologram needs to be digitised with the rate of 25 000 samples per one inch (25 000 DPI) or better! Inherently, the digitisation devices, i.e., the digital camera, scanner, and display, also need to provide the same resolution. This is a technological problem and the contemporary devices are still quite far from such resolution. Also the hologram of a virtual object needs to be computed with such resolution. Not to mention that a proper hologram generation algorithm is still work in progress.

In our research, we focus on the problem of computing digital holograms of virtual objects. As we noted above, this problem has two objectives. The first objective is to find a proper hologram generation algorithm. The second objective is to create such algorithm implementation that will be able to handle the massive workload required for computing even the smallest holograms. As an illustration of the problem we give the example of 17 inch display. The standard 2D content is sufficiently displayed using 1280×1024 pixels. Similar sized holographic content needs resolution at least 340 000×255 000 pixels.

Our goal is to develop a hologram generation algorithm capable of handling virtual objects described in a format usual in a contemporary computer graphics. For achieving sufficient performance we aim on reusing the acceleration techniques known from the computer graphics discipline. Note, that our research has nothing to do with the efforts to exploit holography as a tool to enhance the computer graphics features such as a simulation of various effects due to focus of a camera. This branch of research deals with holograms that have small resolution thus the actual physical size is measured in milimetres.

This research is supported by a national project MŠMT LC-06008. This research was supported by the EU project 3DTV: The True Vision.

People

Supervisor

Researchers

Students

Developed Methods and Facilities

Publications

This is a list of reviewed publications. For a complete list including technical reports refer to individual methods.

  • (2011) Lobaz, P. Reference calculation of light propagation between parallel planes of different sizes and sampling rates. Optics Express, Vol. 19, Issue 1, pp. 32-39. eISSN: 1094-4087
  • (2010) Hanák, I. Accelerating Digital Hologram Generation. PhD. thesis. University of West Bohemia, 2010.
  • (2010) Hanák, I., Herout, A., Zemčík, P. Acceleration of detail driven method for hologram generation. Optical Engineering, 2010, vol. 49, no. 8. ISSN 0091-3286.
  • (2010) Hanák, I., Janda, M., Skala, V. Detail-driven digital hologram generation. The Visual Computer, 2010, vol. 26, no. 2, pp. 83-96. ISSN 0178-2789.
  • (2009) Hanák, I., Zemčík, P. Zadník, M., Herout, A. Hologram synthesis accelerated in field programmable gate array by partial quadratic interpolation. Optical Engineering, 2009, vol. 48, no. 8, pp. 1-7. ISSN 0091-3286.
  • (2008) Janda, M., Hanák, I., Onural, L. Hologram synthesis for photorealistic reconstruction. Journal of the Optical Society of America, 2008, vol. 25, no. 12, pp. 3083-3096. ISSN 1084-7529.
  • (2007) Hanák I., Janda M., Skala V. Full-Parallax Hologram Synthesis Of Triangular Meshes Using A Graphical Processing Unit. 3DTV Conference 2007.
  • (2007) Janda M., Hanák I., Skala V. HPO Hologram Synthesis For Full-Parallax Reconstruction Setup. 3DTV Conference 2007.
  • (2006) Janda, M., Hanák, I., Skala, V. Digital HPO Hologram Rendering Pipeline. EUROGRAPHICS 2006.
  • (2006) Janda, M., Hanák, I., Skala, V. Scanline Rendering of Digital HPO Holograms and Hologram Numerical Reconstruction. SCCG 2006.

Technical reports

This is a list of technical reports concerning (digital) holography published by Department of Computer Science and Engineering.

Last Update: 06.01.2011