Multidimensional representation of ultrasonic data processed by Reconfigurable Ultrasonic System-on-Chip using OpenCL high-level synthesis

Ultrasonic non-destructive testing has been widely used to determine the properties of materials, and more importantly their integrity. However, these techniques often require capturing massive amounts of data and intensive signal processing for processes such as image formation, analysis, characterization, classification and diagnosis. Our study is focused around utilizing a Reconfigurable Ultrasonic System-on-Chip Hardware (RUSH) platform to process ultrasonic signals in real-time. To this end, OpenCL has been utilized in the RUSH platform to provide a means to accelerate computation using the FPGA fabric while providing consistent memory and execution models to enable portability. To visualize the ultrasound data, a Graphical User Interface (GUI) for the Analysis of Multidimensional Ultrasonic data on RUSH (GAMUR) has been incorporated. GAMUR utilizes C++/QT and OpenGL to enable enhanced visualization and control features within the RUSH platform. The interface not only features the ability to view the ultrasound signals in one dimension, two dimensions, or three dimensions; but also to command and configure the hardware accelerators built using OpenCL. Therefore, the system provides a means for analyzing, visualizing and accelerating the extraction of information from multi-dimensional ultrasound data.