Sparse Noise Texture-Base Visualization of Image Spaces Advection
Ziyan He, Yunpeng Li, Sheng Zhang, Biaosong Chen · 2024
Texture-based visualization methods are frequently utilized in the field of flow field data visualization. This study proposes an image space advection algorithm based on sparse noise textures to address the limitations of line integral convolution (LIC) algorithm in expressing directionality. The algorithm involves generating grayscale texture images with short integration step lengths using LIC algorithm and filtering them to produce sparse noise images. The graphics obtained from image space advection are blended with sparse noise images to enable the observation of flow field directions through dynamic images. The gray images are then mapped with vector field data to the HSV color space and converted to RGB, resulting in color images. The proposed method for image space advection, which is based on sparse noise textures, effectively conveys the flow direction of the flow field. The experimental results support the effectiveness of this method. This method addresses the limitations of traditional texture images in expressing directionality. Furthermore, by utilizing the algorithm on the GPU, it achieves real-time rendering computational efficiency.