Resultant 3-D Images

CytoSpec software was employed to generate a 2-D ANN image for every tissue section calculated from the fully trained NeuroDeveloper ANN model. In the final step, the 2-D images are provided as inputs for the SCI Run software to render the completed 3-D ANN image. 

---

Comparing the 3-D images simulated and the experimental photographs in Fig. 10, it can be seen that the droplet shapes are well reproduced by the present model. During the first 3.5 ms of the impact (frames 1-3), a liquid film with flattened disc shape is formed immediately after the impact. The inertial force drives the liquid to continue spreading on the solid surface, while the surface tension and the viscous forces resist the spreading of the liquid film. As a result, the droplet spreading speed decreases and the fluid mass starts to accumulate at... 

Fig. 10-3 shows a 3-D view of the true model. Fig. 10-6 presents a 3-D image of the LQL inversion result. Only values of resistivity below a certain threshold have been plotted out of the inverted cube-shaped area. Both the location and the depth of the dike are determined very well. Its dip is also remarkably well defined. We can even distinguish some of the details of its shape. Its resistivity, however, is overestimated. These results demonstrate the stability of the method in the presence of noise (results of inversion of noise-free data, not shown here, are very similar). 

Figure 10-4 3-D image of the LQL inversion result for a model of a dipping dike.

If we were to watch the movie without the special glasses provided by the movie house, we would see the blurry combination of the two movies. However, the 3-D glasses consist of polarized lenses, with planes of polarization that ate mutually perpendicular. The left lens, polarized in one direction, blocks the image that is polarized perpendicular to it. The right lens, polarized in the other direction, blocks the other image. Oiu eyes are tricked into seeing two different movies, which our brain combines to form one 3-D image. The results of this process can be quite impressive ... 

Submicrofibril and triple-stranded left-hand helical microfibrils are found in tobacco primary cell wall and bacterial A. xylinum cellulose. We suspect from our results and the literature survey outlined in reference (1) that the triple stranded structures are prominent in the primary plant cell wall. The highly crystalline cellulose of plant and algae secondary cell wall appears by X-ray fiber diffraction (18,19) and TEM lattice imaging (20-23) to be largely crystalline arrays of planar straight chains of (l-4)-/3-D-glucan chains. 

D being were looking at a ball, the resultant image is produced by the excitation of a ball-shaped pattern of nerve endings in the spherical retina. Each point on the 3-D ball sends a light ray upsilon to a ballshaped region in the spherical retina (Fig. 3.4). 

Fig. 25. 2-D MR image of the deposition of 80-pm particles (fines) within a bed packed with 5-mm diameter spherical glass beads. The water flow rate was 300mLmin . All images were acquired in 3-D with isotropic spatial resolution of 188 pm x 188 pm x 188 pm. Two local regions associated with a buildup of fines are highlighted and are identified by the low apparent H spin density from these regions resulting from low voidage and relaxation time effects. Flow was in the + z direction. Reprinted from reference (54), with permission of Springer Science and Business Media.

In addition to the immediate practical application of imaging flow fields within reactors, data sets such as these are of use in developing, implementing, and validating numerical codes for the prediction of flow and chemical conversion in process units. In particular, if the 3-D MR image of the internal structure of the reactor is used directly as the simulation lattice in the flow-field simulation code, the subsequent comparison of MR velocity data with the predictions of the code enables a direct evaluation of the ability of the code to predict the flow field. This point is important because often when numerical predictions are compared with experimental results, assumptions have to be made in constructing the simulation lattice itself, which means that the final comparison is influenced not only by the ability of the code to predict the flow field but also the ability to construct a... 

---