In this paper, we obtain some Bonnesen-style Minkowski inequalities of mixed volumes of convex bodies K and L in the Euclidean space . Let L be the unit ball; we get some better Bonnesen-style isoperimetric inequalities than Dinghas’s result for .
MSC:52A20, 52A40.
Hinweise
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
The authors read and approved the final manuscript.
1 Introduction
It is well known that the ball has the maximum volume among bodies of fixed surface area in the Euclidean space . That is, of all domains K with surface area and volume (cf. [1, 2]),
(1)
with equality if and only if K is a ball. Here denotes the volume of the unit ball,
Anzeige
where is the Gamma function.
The isoperimetric deficit
(2)
measures the deficit between the domain K and a ball of radius . A Bonnesen-style isoperimetric inequality is of the form (cf. [2‐4])
(3)
where the quantity is a non-negative invariant of geometric significance of K and vanishes only when K is a ball.
Anzeige
Bonnesen himself proved several inequalities of the form (3) in the Euclidean plane (cf. [5, 6]), but he was not able to obtain direct generalizations of his two-dimensional results. This was done much later, first by Hadwiger [7] for , and then by Dinghas [8] for arbitrary dimension. From then on, some Bonnesen-style inequalities in the higher dimensions and generalizations have been obtained by Osserman (cf. [1, 2]), Santaló (cf. [9]), Groemer and Schneider (cf. [10]), Zhang (cf. [11]), Zhou (cf. [4, 12]) and others. See references [13‐36] for more details. The following well-known Bonnesen-style inequality for a convex body K in the Euclidean space is due to Dinghas (cf. [8]):
(4)
where r is the in-radius of K, and equality holds if and only if K is a ball.
In [11], some different forms of Bonnesen-style isoperimetric inequalities have been established associated with the mean width of K. Zhang obtained (cf. [11])
where and R are the mean width and out-radius of K, respectively.
The Minkowski inequality of mixed volume is a natural generalization of the isoperimetric inequality (1) in the Euclidean space (cf. [27, 37‐39]). Let K, L be convex bodies in , then
(5)
where is the mixed volume of K and L and the equality holds if and only if K and L are homothetic.
Motivated by (2), we define the Minkowski homothetic deficit as
(6)
The Minkowski homothetic deficit measures the homothety between K and L. Then a Bonnesen-style Minkowski inequality would be of the form
(7)
where the quantity is an invariant of geometric significance about K and L with the following basic properties:
1.
is non-negative;
2.
vanishes only when K and L are homothetic.
Note that let L be the unit ball B and by , the surface area of K, then the Minkowski homothetic deficit is just the isoperimetric deficit. Therefore, the Bonnesen-style Minkowski inequality (7) is more general than the Bonnesen-style isoperimetric inequality (3).
In this paper, we focus on Bonnesen-style Minkowski inequalities of type (7). Some are obtained. Let L be the unit ball; then we obtain stronger Bonnesen-style isoperimetric inequalities K than (4).
2 Preliminaries
A set of points K in the Euclidean space is convex if for all and , . A domain is a set with nonempty interiors. A convex body is a compact convex domain. The set of convex bodies in is denoted by . Let be the class of members of containing the origin in their interiors. Write V for an n-dimensional Lebesgue measure and for an -dimensional Hausdorff measure. denotes the surface of the unit ball in .
A convex body is uniquely determined by its support function , where , for . For the support function of the dilate of a convex body K we have
(8)
Note that support functions are positively homogeneous of degree one and subadditive. It follows immediately from the definition of support functions that for convex bodies K and L
(9)
For a convex body K and each Borel set , the reverse spherical image , of K at ω is the set of all boundary points of K which have an outer unit normal belonging to the set ω. Associated with each convex body there is a Borel measure on called the Aleksandrov-Fenchel surface area measure of K, defined by
for each Borel set . Observe that for the surface area measure of the dilate cK of K we have
The Minkowski sum of convex sets in is defined by
The mixed volume of compact convex sets in is defined by
The Aleksandrov-Fenchel inequality about the i th mixed volume is
(10)
where
with appears times and appears i times and (10) holds as an equality if and only if K and L are homothetic.
Note that
(11)
The following inequality for mixed volumes is the general Aleksandrov-Fenchel inequality: Let and . Then
Hence
(12)
Let , then , the i th quermassintegral of the convex body .
The mixed volume has monotonicity: If , then
The mixed volume of the convex bodies has the integral form
(13)
Since
we have
If B is the unit ball, then
the surface area of K. The mean width of K is
that is,
The in-radius , out-radius of K with respect to L are, respectively, defined by
Notice that always
When L is the unit ball, and are the radius of maximal inscribed and minimal circumscribed balls of K, respectively.
3 Bonnesen-style Minkowski inequalities associated with
In this section, we derive some Bonnesen-style Minkowski inequalities associated with in-radius of K with respect to L. In [26], Diskant improved the Minkowski inequality of mixed volumes as follows.
Lemma 1LetK, Lbe convex bodies in the Euclidean space , then
(14)
with equality if and only ifKis homothetic toL.
Note that the right-hand side of (14) is non-negative for (). By (13) we have
From Lemma 1 and using the inequality (for ), a lower bound of the Minkowski deficit follows (cf. [26, 27]).
Proposition 1LetK, Lbe convex bodies in the Euclidean space , then
(15)
where the inequality holds as an equality if and only ifKandLare homothetic.
The following Bonnesen-style Minkowski inequality is stronger than (15) for .
Theorem 1LetK, Lbe convex bodies in the Euclidean space , then
(16)
with equality if and only ifKis homothetic toL.
Proof Since and by (for ), we have
Note that (14) can be rewritten as
Multiplying by on both sides, we have
By these inequalities, we complete the proof of the theorem. □
Let L be the unit ball and notice in (16), we obtain the following Bonnesen-style isoperimetric inequality that strengthens Dinghas’s inequality (4) for .
Corollary 1LetKbe a convex body inandrbe the in-radius ofK, then
(17)
with equality if and only ifKis a ball.
For , we obtain a stronger Bonnesen-style Minkowski inequality as follows.
Theorem 2LetK, Lbe convex bodies in the Euclidean space (), then
with equality if and only ifKis homothetic toL.
Proof Let and , then .
That is,
Multiplying by both sides of (14) and via the formula
we obtain
We complete the proof of Theorem 2. □
Let L be the unit ball and by in Theorem 2; we obtain the following stronger Bonnesen-style isoperimetric inequality than Dinghas’s inequality (4) for .
Corollary 2LetKbe a convex body in () andrbe the in-radius ofK, then
with equality if and only ifKis a ball.
4 Bonnesen-style Minkowski inequalities associated with the mean width
In this section, we derive some Bonnesen-style Minkowski inequalities associated with the mean width.
Lemma 2LetK, Lbe convex bodies in , then
(18)
with equality if and only ifKandLare homothetic.
Proof By inequality (12), we have
By the Aleksandrov-Fenchel inequality (10) we have
Therefore
□
Theorem 3LetK, Lbe convex bodies in , then
(19)
with equality if and only ifKandLare homothetic.
Proof Via (18), we have
That is
□
The following Bonnesen-style inequality is a direct consequence of Theorem 3.
Theorem 4LetKbe a convex body in , then
with equality if and only ifKis a ball.
Lemma 3LetK, Lbe convex bodies in , then
(20)
Proof The Minkowski inequality (5) gives
The above inequality together with (18) leads to Lemma 3. □
We are now in a position to prove the following Bonnesen-style Minkowski inequality.
Theorem 5LetK, Lbe convex bodies in , then
(21)
with equality if and only ifKandLare homothetic.
Proof From (20) we have
□
The following Bonnesen-style inequality is a direct consequence of Theorem 5 when L is the unit ball.
Theorem 6LetKbe a convex body in , then
with equality if and only ifKis a ball.
Acknowledgements
The authors would like to thank two anonymous referees for many helpful comments and suggestions that directly lead to the improvement of the original manuscript. The authors are supported in part by NSFC (No. 11271302 and No. 11326073), Fundamental Research Funds for the Central Universities (No. XDJK2014C164), the Ph.D. Program of Higher Education Research Funds (No. 2012182110020).
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.