Patterns of Flow –like Images Revealed by Digital Image Processing

Automatic Systems to fluid flow visualization can help in solution of several Fluid Dynamic Phenomena. Usually, these systems can be concerned of three fundamental steps: data acquisition, processing and result display. There are several ways to achieve the first step, which can be grouped as follows: simulation of mathematics elements, and detection/description of fluid flow patterns characteristics. The detection/description is done in real images captured in laboratory experiments and can form the basis of moderns fluid flow visualization systems. These fluid flow patterns can be of several classes which, in Dynamical Systems Theory, are called Critical Points or Singular Points. Even in this theory, these patterns are modeled as different equation systems too. These models can revel intrinsic aspects of fluid flow phenomena and are difficult to achieved in real time computation. So, in this paper, we present a new efficient methodology concerning of four steps, which is also part of a larger effort toward fluid flow patterns description. The time required to execute the proposed method is less than the time required for the other methods here discussed. Some examples, where the proposed methodology can be applied, are presented and the results are discussed. Moreover, it is shown that the proposed method can find some kind of occluded patterns too. Key-Words: Pattern Analysis, Fluid Flow Images, Pattern Recognition.