Application of smoothing algorithms to enhance quality of 3D reconstructed images in tissues and cultures

Abstract

Visualization of scientific data plays an ever increasingly vital role in the study of humans, plants, animals, and the world around us. An integral part of that visualization and what makes it possible is the use of modern technology, specifically computers. Computers give scientists and engineers freedom to study and manipulate a sample as often as necessary in order to understand its characteristices while only requiring them to sample a specimen and record data once. This has afforded a great opportunity for lending and testing different conditions and the effects of different stimuli with minimal intrusion and damage to the original human, plant, or animal from which the specimen was acquired. To that end, the focus of this research is to apply this very efficient, effective tool to smooth 3D reconstructed images of test spheres and plant cells. In order to reconstruct a series of segmented images in 3D it is necessary to use the Marching Cubes algorithm which takes a specified scalar value and uses it to construct a 3D surface called an isosurface from the information contained within the segmented images. There is a variety of software packages currently on the market that use this algorithm as part of their functionality. However, few are as efficient, effective, low cost, and easy to use as the Visualization Toolkit developed by Bill Lorensen, Will Schroeder and Ken Martin. One distinct advantage to using the Visualization Toolkit (VTK) as a tool is that it was created with the help of the original developer of the Marching Cubes algorithm, Bill Lorenesen. VTK offers a low cost, easy to use method of accomplishing very sophisticated postprocessing of images without the need for extensive programming or high end computers such as mainframes.