DENMARK, S.C. (courtesy voorhees.edu) — Dr. Xiqiang Zheng, associate professor in mathematics and computer science, was recently awarded a Henry C. McBay faculty research fellowship (the year 2019 solicitation) from the United Negro College Fund (UNCF) to support his research on computer simulation for CT image segmentation using morphological operations on hexagonal lattices.
The fellowship will help Zheng implement the research topics on CT machines. A CT machine performs X-ray scans around an object from different directions. The data from the scans are used to compute the internal image of the object. Image segmentation is one of the image processing tools that may be applied to the CT image to determine the sizes of the internal organs, bones, or cancers of the scanned object.
During the fellowship, one student majoring in computer science will have the opportunity to participate in the research and receive a stipend. The student will expand his or her knowledge of CT image processing and gain research experiences
Zheng said he is taking novel approaches for CT image computation and segmentation. “The usual two-dimensional CT computations are done on a square grid and over a square domain, where a square grid is tessellated with congruent squares,” Zheng said. “Because of rotations of CT machines, we can assume that the CT images are defined on circular regions.”
He added, “Since a circular region can be embedded into a regular hexagonal region more compactly into the square region, if the CT image is computed on a hexagonal grid (tessellated with congruent regular hexagons) and the regular hexagonal region, better computational results may be achieved or the image segmentation time may be reduced,” Zheng said.
Mathematical morphology is a method based on set theory and topology for quantitative analysis of geometric objects. In a hexagonal grid, each pixel is well connected to six adjacent pixels. Hence morphological operations on hexagonal grids may achieve better image processing effects than on square grids.
In this project, computer codes for morphological operations on hexagonal grids will be developed and applied to CT image segmentation. In the future, other image processing tools on hexagonal grids will be considered.
Zheng has been teaching and performing research at Voorhees since 2008. He earned a master’s degree in mathematics from Jiangxi University in China and earned a doctorate in applied mathematics from the University of Florida.
For more information, contact the Office of Communications, at 803-780-1191 or at communications@voorhees.edu.