Design Of Transfer Function For Direct Volume Rendering

Volume visualization is a method of extracting meaningful information from volumetric data using graphics and imaging. However, direct volume rendering is a key technology for volume visualization, which is mostly used to visualize volumetric data from scientific or medical applications.Transfer functions are particularly important to the quality of direct volume rendered results, which are used to assign optical properties, such as color and opacity, to original values of the data set being visualized. Unfortunately, finding optimal transfer functions are extremely difficult in general. Therefore, it is a challenging problem to generate an accurate and rapid transfer function. This paper presents 1D and 2D transfer functions based on trail-and-error approach firstly, which benefit the position information of voxel from color histogram additionally. Furthermore, owing to the frustrating process of finding a good transfer function, an automatic generation of transfer function using an underlying data model is proposed. Finally, to harness the GPU's power for computing, numerous computation of automatic generation of transfer function is deployed on GPUs and the performance is improved greatly. Experimental results of rendering CT and MRI datasets illustrate the effectiveness and rapidness of the proposed algorithm.