Registration of Partially Focused Images for 2D and 3D Reconstruction of Oversized Samples.

Methods of fracture surface 3D reconstruction from a series of partially focused images acquired in a small field of view (e.g., by confocal microscope or CCD camera) are well known. In this case, projection rays can be considered parallel and recorded images do not differ in any geometrical transformation from each other. In the case of larger samples (oversized for microscope or CCD camera), it is necessary to use a wider viewing field (e.g., standard cameras); taken images primarily differ in scaling but may also differ in shifting and rotation. These images cannot be used for reconstruction directly; they must be registered; that is, we must determine all transformations in which the images differ and eliminate their effects. There are several ways to do this. This paper deals with the registration based on phase correlation.

Three-dimensional reconstructions of solid surfaces using conventional microscopes.

The three-dimensional digital replicas of solid surfaces are subject of interest of different branches of science and technology. The present paper in its introductory parts brings an overview of the various microscopic reconstructive techniques based on optical sectioning. The main attention is devoted to conventional reconstruction methods and especially to that one employing the Fourier transform. The three-dimensional replicas of this special reconstructive frequency method are compared graphically and numerically with the three-dimensional replicas of the confocal method. Based on the comparative study it has been concluded that the quality of the conventional replicas of surfaces possessing textures of intermediate height irregularities is acceptable and almost comparable with the quality of confocal replicas. This study is relevant both for identifying a convenient technique that provides good qualities of three-dimensional replicas and for selecting the hardware whose price is affordable even for small research groups studying rougher surface textures.

Direct volume rendering methods for cell structures.

The study of the complicated architecture of cell space structures is an important problem in biology and medical research. Optical cuts of cells produced by confocal microscopes enable two-dimensional (2D) and three-dimensional (3D) reconstructions of observed cells. This paper discuses new possibilities for direct volume rendering of these data. We often encounter 16 or more bit images in confocal microscopy of cells. Most of the information contained in these images is unsubstantial for the human vision. Therefore, it is necessary to use mathematical algorithms for visualization of such images. Present software tools as OpenGL or DirectX run quickly in graphic station with special graphic cards, run very unsatisfactory on PC without these cards and outputs are usually poor for real data. These tools are black boxes for a common user and make it impossible to correct and improve them. With the method proposed, more parameters of the environment can be set, making it possible to apply 3D filters to set the output image sharpness in relation to the noise. The quality of the output is incomparable to the earlier described methods and is worth increasing the computing time. We would like to offer mathematical methods of 3D scalar data visualization describing new algorithms that run on standard PCs very well.

The two-dimensional and three-dimensional processing of images provided by conventional microscopes.

In this paper, I present an interesting processing method of microscopic images. High-pass type filters are generally used for image focusing. They enhance the high spatial frequencies. They are, however, efficient only in cases when the picture is not sharp because of the low contrast on high frequencies (for example in a TV picture). These filters, are not appropriate if the lack of sharpness has been caused by other factors. In this case, it is not possible to construct a three-dimensional model of the observed object. Better results and a three-dimensional model can be obtained by applying the following theory. As part of this paper, an original program based on this theory is described.