Accelerating techniques in volume rendering of irregular data

Abstract

Volume rendering of irregular data has become a challenging research topic in the area of scientific visualization in recent years. In comparison with regular data rendering, irregular data volume rendering faces many new problems. Because cells in an irregular volume have various shape and size, volume rendering by the raycasting method has more difficulty in calculating the intersection between rays and cells, and the projection method has even greater trouble in queuing up cells, in particular concave cells. To accelerate volume rendering of irregular data, this paper presents a hybrid method which possesses the advantages of both raycasting and projection methods. The main strategy in the method is using a polygon scan conversion algorithm in 3-D space to set up a ray path with high efficiency, and by establishing intersection links to allow the operation of color composition to be locally conducted on-the-fly. As a result, it is possible to produce high quality images without the requirement of the heavy task for cell queuing up and complex ray intersection calculation.