The anisotropic diffusion is defined as

$$\frac{\partial I}{\partial t} = \mathrm{div} \left( c(x,y,t) \nabla I \right)= \nabla c \cdot \nabla I + c(x,y,t) \Delta I$$

where $\operatorname{div}$ is the divergence operator, $c(x,y,t)$ is the diffusion coefficient (which depends on the $x$ and $y$ coordinates and on the time $t$ of the diffusion), $\nabla I$ is the gradient of the image and $\Delta I$ is the Laplacian of the image and $I(\cdot ,t):\Omega \rightarrow {\mathbb {R}} $ is the actual image with domain $\Omega$ which is a subset of $\mathbb{R}^2$.

Why exactly is the anisotropic diffusion equivalent to the heat equation when using a constant diffusion coefficient?


The heat equation for a function $I(x, y, t)$ of three variables $x$, $y$ (the coordinates) and $t$ (the time) is defined as follows

\begin{align} \frac{\partial I}{\partial t} &= \alpha\left(\frac{\partial^2I}{\partial x^2}+\frac{\partial^2I}{\partial y^2}\right) \\ &= \alpha \operatorname{div}(\nabla I) \end{align}

where $\alpha$ is a real coefficient called the diffusivity and it thus corresponds to a constant $c(x, y, t)$, which, in the case of the anisotropic diffusion equation, can change as a function of $x$, $y$ and $t$.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.