Samson Shen, Ward Melville High School

Abstract:

Imaging datasets are often complex and typically deal with huge number of data points. Typical analyses of imaging datasets focus either on average characteristics such as particle size distributions, or on specific small regions of interest such as morphology changes of a few specific particles. Such approaches work well in many applications including X-ray imaging at synchrotrons. In this poster, we show that significant new information can be obtained from statistically significant residuals beyond the simple average properties for the entire dataset.  

We illustrate this method by a careful analysis of the residuals or outliers in nonlinear regression analysis of histogram distributions of the intervein angles in the venation patterns of dragonfly wings. We found that these significant residuals indicate a set of preferred angles that are the results of the golden-rule partitions of the regular polygon angle intervals. These golden-rule optimized patterns provide the best possible biomechanical support and aerodynamic performance for the dragonfly wings. 

Our new analysis represents a new imaging analysis method, and can be easily adapted and applied to imaging analysis in general, including synchrotron X-ray imaging.

This research used resources of the National Synchrotron Light Source II, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Brookhaven National Laboratory under Contract No. DE-SC0012704. 

Poster Session Link: https://gather.town/invite?token=0pEoq7VP

If you have any questions for the presenter, please contact them via email: junwang@bnl.gov