Abstract: In recent years, flow visualization has become a research hotspot in computer graphics, and one of its most important research goals is the extraction and visualization of vortex features. Since there is still no general definition of vortex, the evidences to determine whether vortex exists are different in the pertinent literature. In order to make a systematic review of the vortex feature extraction methods of fluid, the paper firstly explained the relevant research directions of vortex extraction, and then reviewed and summarized the development of the vortex feature extraction methods of fluids. The commonly used vortex extraction methods were classified into four categories, including point-based, line-based, geometry-based and machine learning-based methods. For the newly proposed invariance of reference frame, the vortex extraction methods were divided into three types, Galilean invariance, rotation invariance and Lagrange invariance. To compare the advantages and disadvantages of various methods, several classical methods of each type were given in the review, therefore providing a clear outline for future studies. Finally, the difficulties and problems of each method were concluded, and the possible future research focus were introduced.

Key words: vortex extraction method, feature visualization, reference frame invariance, flow feature; vortex core