Particle image velocimetry (PIV) processing method

Abstract

The invention discloses a particle image velocimetry (PIV) processing method comprising the following steps: A. constructing a level set function phi in an N*M pixel range omega, wherein the level set function phi is a continuous function defined in the N*M pixel range omega; dividing the range omega into a first subregion omega1 and a second subregion omega2 so that a motion field, a divergence field and a rotation field are smoothly connected in the two subregions of omega1 and omega2, but are not continuous on the boundary, wherein a region where the phi is larger than or equal to 0 corresponds to the first subregion omega1, and a region where the phi is smaller than 0 corresponds to the second subregion omega2; B. constructing an energy function E=E1+lambda *E2 of the range omega:, wherein the E is a global energy functional, the E1 is data constraint energy, and the E2 is smooth constraint energy; and C. minimizing the global energy functional to realize the optimal approximation of the motion field, the divergence field and the rotation field. The invention provides the PIV processing method with high precision and good reliability.

Description

technical field [0001] The invention relates to image processing, in particular to a particle image velocity measurement processing method. Background technique [0002] The motion forms of images in nature vary greatly. Fluid motion is a typical non-rigid body motion. Particle Image Velocimitry (PIV for short) is a new type of non-contact measurement technology. Usually, PIV is to put tracer particles into the fluid, and under the irradiation of the laser sheet light, use a camera to capture the particle image moving with the fluid in the direction perpendicular to the sheet light, and then process, analyze and calculate the image, and finally get the cross-section of the flow field A measurement method of two-dimensional velocity field. As a full flow field, non-contact, non-disturbance, high-precision flow visualization method, PIV is suitable for the measurement of time-varying complex flow fields such as turbulent flow and unsteady flow. Today, PIV is an interdiscipli...

AI Technical Summary

Problems solved by technology

[0006] 3. When the speed of particle movement is very slow (the displacement is at the sub-pixel level) or the particle density in some areas is too concentrated, when a larger window is used for correlation search, multiple peaks will appear

In addition to vector field calculations, there are still many unresolved problems on how to accurately calculate non-scattering fields, irrotational fields, viscous regions, and extract topological points (absorption points, divergence points), etc.

[0011] 2) Many variational optical flow calculations need to solve elliptic partial differential equations with boundary value problems, and are finally realized by solving linear or nonlinear equations, which require a large number of iterative calculations

Usually, these features are analyzed and processed separately and then combined and judged, but how to comprehensively describe and analyze the fluid motion field from the perspective of high-dimensional images is still a blank

Images