I have a question about an estimate of the surface area of a set.

Let $B(r)$ denotes the open ball of $\mathbb{R}^{d}$ centered at origin with radius $r>0$. Let $F:\mathbb{R}^{d} \to \mathbb{R}^{d}$ be a Lipschitz continuous function and define \begin{equation} \text{Lip}(F)=\inf\{L>0:|F(x)-F(y)| \le L|x-y|,x,y \in \mathbb{R}^{d} \}. \end{equation}


Define $B^{\ast}(r)=F(B(r))$. I want to obtain an upper estimate of $\sigma(B^{\ast}(r))$, where $\sigma(B^{\ast}(r))$ is the surface area of $B^{\ast}(r)$. Strictly speaking, $\sigma(B^{\ast}(r))$ is the $d-1$ dimensional Hausdorff measure of $\partial B^{\ast}(r)$. That is, $\sigma(B^{\ast}(r))=\mathcal{H}^{d-1}(\partial B^{\ast}(r))$.

My attempt

It is known that the following inequality: \begin{equation*} \mathcal{H}^{d-1}(F(E)) \le \text{Lip}(F)^{d-1}\mathcal{H}^{d-1}(E), \end{equation*} for all $E \subset \mathbb{R}^{d}$.

But can we show $\partial B^{\ast}(r) \subset F \bigl( \partial B(r) \bigr)$ (if necessary, we may assume $F$ is smooth)? If this is true, we have \begin{align*} \sigma(B^{\ast}(r))&=\mathcal{H}^{d-1}(\partial B^{\ast}(r)) \\ & \le \text{Lip}(F)^{d-1}\mathcal{H}^{d-1}(\partial B(r))\\ &= \text{Lip}(F)^{d-1}\sigma( B(r)). \end{align*}

Can we have an upper estimate like as $ \sigma(B^{\ast}(r)) \le Cr^{d-1} ? $

Thank you in advance.

  • 1
    $\begingroup$ In addition to what Mikhail Katz said, see for example this duplicate : math.stackexchange.com/questions/917494/…. Furthermore, while $\partial B^\ast(r)$ is $(d-1)$-rectifiable and has finite $\mathcal{H}^{d-1}$ measure, I don't know why you should have the same decay with $r^{d-1}$. $\endgroup$ – Paul-Benjamin Oct 26 '16 at 16:04
  • $\begingroup$ Could you please show me a reference for the inequality \begin{equation*} \mathcal{H}^{d-1}(F(E)) \le \text{Lip}(F)^{d-1}\mathcal{H}^{d-1}(E), \end{equation*}? $\endgroup$ – Humed Feb 1 at 20:59
  • $\begingroup$ Please see Section 2.4.1 in this book: crcpress.com/Measure-Theory-and-Fine-Properties-of-Functions/…. $\endgroup$ – sharpe Feb 3 at 10:50

There is no reason to assume that the boundary of the image is contained in the image of the boundary, even in the case $n=1$; consider for example the function $y=x^2$ on $[-1,1]$.


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.