Skip to main content
Springer logoLink to Springer
. 2018 Sep 19;2018(1):247. doi: 10.1186/s13660-018-1836-2

Some inequalities for (p,q)-mixed volume

Bin Chen 1, Weidong Wang 1,2,✉
PMCID: PMC6154050  PMID: 30839665

Abstract

Lutwak, Yang, and Zhang introduced the concept of (p,q)-mixed volume whose special cases contain the Lp-mixed volume and the Lp-dual mixed volume. In this article, associated with the (p,q)-mixed volumes, we establish related cyclic inequalities, monotonic inequalities, and product inequalities.

Keywords: (p,q)-mixed volume; Cyclic inequality; Monotonic inequality; Product inequality

Introduction and main results

At the end of the nineteenth century, Brunn and Minkowski pioneered the classical Brunn–Minkowski theory of convex bodies, which is the product of Minkowski linear combination of vectors and volumes in the Euclidean space. The core of this theory are mixed volume, mixed area measure, and the basic Brunn–Minkowski inequality. In recent years, Brunn–Minkowski theory attracted wide attention (see [1, 2]).

By the 1960s, Firey put forward the concept of Lp-Minkowski combination of convex bodies (see [2]). In 1993, Lutwak [3] introduced the Lp-Minkowski linear combination of convex bodies to the classical Brunn–Minkowski theory, proposed the notions of Lp-mixed volume, Lp-mixed quermassintegrals, and Lp-surface area measure, and obtained the corresponding integral expression, which extended the classical Brunn–Minkowski theory to Lp space (called the Lp Brunn–Minkowski theory). This new theory has attracted a large number of researchers’ interests in recent years (see [4–22]). Especially, the concept of Lp-mixed volume (p≥1) plays an important role in Lp Brunn–Minkowski theory (see [3, 23]).

The classical dual Brunn–Minkowski theory of star bodies was introduced by Lutwak [24] in 1975. In 1996, on the basis of Lp harmonic radial combination, Lutwak [23] put forward the concept of Lp-dual mixed volume (p≥1) and gave its integral expression. This means that the preliminary Lp dual Brunn–Minkowski theory has been established. Afterwards, Grinberg and Zhang defined the notion of Lp radial combination (p>0). In 2002, Gardner improved p>0 to p≠0 in Lp radial combination, and got a more extensive class of Lp-dual mixed volume (p≠0). For more information about the classical dual Brunn–Minkowski theory and Lp dual Brunn–Minkowski theory, please refer to [25–35].

Very recently, Huang et al. [10] constructed the dual curvature measure in dual Brunn–Minkowski theory. These measures are dual to Federer’s curvature measures which are fundamental in the classical Brunn–Minkowski theory. In 2018, Lutwak, Yang, and Zhang [36] made further work and introduced Lp dual curvature measures which include Lp surface area measure, dual curvature measures, and Lp integral curvatures. Using this new concept, they introduced the (p,q)-mixed volume, which unifies Lp-mixed volume and Lp-dual mixed volume. Thus, Lp Brunn–Minkowski theory and Lp dual Brunn–Minkowski theory are partially unified.

Let K be a convex body if K is a compact, convex subset in an n-dimensional Euclidean space Rn with nonempty interior. The set of all convex bodies in Rn is written as Kn. Let Kon denote the set of convex bodies containing the origin in their interiors. Let Son denote the set of star bodies (about the origin) in Rn. We write u for the unit vector and B for the unit ball centered at the origin, the surface of B denoted by Sn−1. We shall use V(K) for the n-dimensional volume of the body K in Rn.

Suppose that R is the set of real numbers. If E∈Kn, the support function of E, hE=h(E,⋅): Rn→R, is defined by (see [1, 2])

h(E,x)=max{x⋅y:y∈E},x∈Rn,

where x⋅y denotes the standard inner product of x and y in Rn.

For a compact star-shaped (about the origin) E in Rn, the radial function ρE of E, ρE=ρ(E,⋅): Rn∖{0}→[0,+∞), is defined by (see [1, 2])

ρ(E,x)=max{λ≥0:λx∈E},x∈Rn∖{0}.

If ρE is positive and continuous, then E is called a star body.

If E∈Rn is a nonempty subset, the polar set of E, E∗, is defined by (see [1, 2])

E∗={x∈Rn:x⋅y≤1,y∈E}.

From this, it is easy to get that (E∗)∗=E for all E∈Kon.

From the definition of polar, we know that if E∈Kon, the support and radial function of E∗, the polar body of E, have the following relationships (see [1, 2]):

h(E∗,⋅)=1ρ(E,⋅),ρ(E∗,⋅)=1h(E,⋅). 1.1

Very recently, Lutwak et al. defined a new concept (i.e., Lp dual curvature measures) as follows (see [36]): For p,q∈R, K∈Kon, and L∈Son, the Lp dual curvature measures C˜p,q(K,L,⋅) on Sn−1 is defined by

∫Sn−1g(v)dC˜p,q(K,L,v)=1n∫Sn−1g(αK(u))hK−p(αK(u))ρKq(u)ρLn−q(u)du 1.2

for each continuous g:Sn−1→R. Here αK is the radial Gauss map (see [36]).

By (1.2), Lutwak, Yang, and Zhang [36] defined the (p,q)-mixed volumes as follows: For K,L∈Kon, M∈Son, and p,q∈R, the (p,q)-mixed volume V˜p,q(K,L,M) of K,L,M is defined by

V˜p,q(K,L,M)=∫Sn−1hLp(v)dC˜p,q(K,M,v).

For the (p,q)-mixed volumes, the authors [36] gave the following integral formula:

V˜p,q(K,L,M)=1n∫Sn−1(hLhK)p(αK(u))ρKq(u)ρMn−q(u)du. 1.3

By (1.3), Lutwak et al. introduced the Lp mixed volume for p∈R. For K,L∈Kon and any real p, the Lp mixed volume Vp(K,L) is given by

Vp(K,L)=1n∫Sn−1hLp(u)dSp(K,u).

Here Sp(K,⋅) denotes the Lp surface area measure (see [3]). The case of p≥1 is Lutwak’s Lp mixed volume (see [3]).

At the same time, for K,L∈Son and q∈R, they also defined the qth dual mixed volume V˜q(K,L) by

V˜q(K,L)=1n∫Sn−1ρKq(u)ρLn−q(u)du.

In addition, they gave several special cases of (p,q)-mixed volume: For p,q∈R, K,L∈Kon, and M∈Son, then

V˜p,q(K,K,K)=V(K)=1n∫Sn−1ρKn(u)du, 1.4
V˜p,q(K,K,M)=V˜q(K,M), 1.5
V˜p,q(K,L,K)=Vp(K,L), 1.6
V˜0,q(K,L,M)=V˜q(K,M), 1.7
V˜p,n(K,L,M)=Vp(K,L). 1.8

In this paper, we further study the (p,q)-mixed volumes and establish some inequalities including cyclic inequalities, monotonic inequalities, and product inequalities. First, we give a class of cyclic inequalities as follows.

Theorem 1.1

Suppose p,q,r,s∈R satisfy 1≤p<q<r≤n. If K,L∈Kon and M∈Son, then

V˜q,s(K,L,M)r−p≤V˜p,s(K,L,M)r−qV˜r,s(K,L,M)q−p 1.9

with equality if and only if K,L, and M are dilates.

Theorem 1.2

Suppose p,q,r,s∈R satisfy 1≤p<q<r≤n. If K,L∈Kon and M∈Son, then

V˜s,q(K,L,M)r−p≤V˜s,p(K,L,M)r−qV˜s,r(K,L,M)q−p 1.10

with equality if and only if K,L, and M are dilates.

Then we obtain a type of monotonic inequalities as follows.

Theorem 1.3

Suppose p,q∈R satisfy 1≤p<q<n. If K,L∈Kon and M∈Son, then

[V˜n−p,p(K,L,M)V(K)]1n−p≥[V˜n−q,q(K,L,M)V(K)]1n−q 1.11

with equality if and only if K and M are dilates.

Theorem 1.4

Suppose p,q∈R satisfy 1≤p<q<n. If K,L∈Kon and M∈Son, then

[V˜p,p(K,L,M)V(M)]1p≤[V˜q,q(K,L,M)V(M)]1q 1.12

with equality if and only if K and M are dilates.

Finally, we set up a type of product inequalities as follows.

Theorem 1.5

Suppose p>0,q∈R. If K,L∈Kon and M∈Son, then

V˜p,q(K,L,M)V˜p,q(K,L∗,M)≥V˜p,q(K,B,M)2 1.13

with equality if and only if L is a ball centered at the origin.

Theorem 1.6

Suppose p,q∈R and q>n. If K,L,M∈Kon, then

V˜p,q(K,L,M)V˜p,q(K,L,M∗)≥V˜p,q(K,L,B)2 1.14

with equality if and only if M is a ball centered at the origin.

The proofs of Theorems 1.1–1.6 will be completed in the next section.

Proofs of theorems

In this part, we give the proofs of Theorems 1.1–1.6.

Proof of Theorem 1.1

For p,q,r,s∈R, K,L∈Kon, and M∈Son. Since 1≤p<q<r≤n, then r−pr−q>1. From (1.3) and Hölder’s integral inequality, we get that for u∈Sn−1

V˜p,s(K,L,M)r−qr−pV˜r,s(K,L,M)q−pr−p=[1n∫Sn−1(hLhK)p(αK(u))ρKs(u)ρMn−s(u)du]r−qr−p⋅[1n∫Sn−1(hLhK)r(αK(u))ρKs(u)ρMn−s(u)du]q−pr−p=[1n∫Sn−1((hLhK)p(r−q)r−p(αK(u))ρKs(r−q)r−p(u)ρM(n−s)(r−q)r−p(u))r−pr−qdu]r−qr−p⋅[1n∫Sn−1((hLhK)r(q−p)r−p(αK(u))ρKs(q−p)r−p(u)ρM(n−s)(q−p)r−p(u))r−pq−pdu]q−pr−p≥1n∫Sn−1(hLhK)q(αK(u))ρKs(u)ρMn−s(u)du=V˜q,s(K,L,M),

i.e.,

V˜q,s(K,L,M)r−p≤V˜p,s(K,L,M)r−qV˜r,s(K,L,M)q−p. 2.1

This yields (1.9). According to the equality condition of Hölder’s integral inequality, we see that equality holds in (2.1) if and only if K,L, and M are dilates. □

In (2.1), if M=K or s=n, by (1.6) or (1.8), we can get the following result (see [23]).

Corollary 2.1

Suppose that p,q,r∈R satisfy 1≤p<q<r≤n. If K,L∈Kon, then

Vq(K,L)r−p≤Vp(K,L)r−qVr(K,L)q−p

with equality if and only if K and L are dilates.

Proof of Theorem 1.2

For p,q,r,s∈R, K,L∈Kon, and M∈Son. Since 1≤p<q<r≤n, then r−pr−q>1. From (1.3) and Hölder’s integral inequality, we get that for u∈Sn−1

V˜s,p(K,L,M)r−qr−pV˜s,r(K,L,M)q−pr−p=[1n∫Sn−1(hLhK)s(αK(u))ρKp(u)ρMn−p(u)du]r−qr−p⋅[1n∫Sn−1(hLhK)s(αK(u))ρKr(u)ρMn−r(u)du]q−pr−p=[1n∫Sn−1((hLhK)s(r−q)r−p(αK(u))ρKp(r−q)r−p(u)ρM(n−p)(r−q)r−p(u))r−pr−qdu]r−qr−p⋅[1n∫Sn−1((hLhK)s(q−p)r−p(αK(u))ρKr(q−p)r−p(u)ρM(n−r)(q−p)r−p(u))r−pq−pdu]q−pr−p≥1n∫Sn−1(hLhK)s(αK(u))ρKq(u)ρMn−q(u)du=V˜s,q(K,L,M).

Thus, we get

V˜s,q(K,L,M)r−p≤V˜s,p(K,L,M)r−qV˜s,r(K,L,M)q−p. 2.2

This yields (1.10). According to the equality condition of Hölder’s integral inequality, we see that equality holds in (2.2) if and only if K,L, and M are dilates. □

Combined with (1.5) and (1.7), taking L=K or s=0 in (2.2), we obtain the following corollary.

Corollary 2.2

Suppose p,q,r∈R satisfy 1≤p<q<r≤n. If K∈Kon and M∈Son, then

V˜q(K,M)r−p≤V˜p(K,M)r−qV˜r(K,M)q−p

with equality if and only if K and M are dilates.

Proof of Theorem 1.3

For p,q∈R, K,L∈Kon, and M∈Son. Since 1≤p<q, then n−qn−p<1. From (1.3), (1.4), and Hölder’s integral inequality, we obtain that for u∈Sn−1

V˜n−q,q(K,L,M)n−pn−qV(K)p−qn−q=[1n∫Sn−1(hLhK)n−q(αK(u))ρKq(u)ρMn−q(u)du]n−pn−q⋅[1n∫Sn−1ρKn(u)du]p−qn−q=[1n∫Sn−1((hLhK)n−p(αK(u))ρKq(n−p)n−q(u)ρMn−p(u))n−qn−pdu]n−pn−q⋅[1n∫Sn−1(ρKn(p−q)n−q(u))n−qp−qdu]p−qn−q≤1n∫Sn−1(hLhK)n−p(αK(u))ρKp(u)ρMn−p(u)du=V˜n−p,p(K,L,M),

i.e.,

[V˜n−p,p(K,L,M)V(K)]1n−p≥[V˜n−q,q(K,L,M)V(K)]1n−q. 2.3

This gives (1.11). According to the equality condition of Hölder’s integral inequality, we know that equality holds in (2.3) if and only if K and M are dilates. □

Proof of Theorem 1.4

For p,q∈R, K,L∈Kon, and M∈Son. Since 1≤p<q, then qp>1. From (1.3), (1.4), and Hölder’s integral inequality, we obtain that for u∈Sn−1

V˜q,q(K,L,M)pqV(M)q−pq=[1n∫Sn−1(hLhK)q(αK(u))ρKq(u)ρMn−q(u)du]pq⋅[1n∫Sn−1ρMn(u)du]q−pq=[1n∫Sn−1((hLhK)p(αK(u))ρKp(u)ρMp(n−q)q(u))qpdu]pq⋅[1n∫Sn−1(ρMn(q−p)q(u))qq−pdu]q−pq≥1n∫Sn−1(hLhK)p(αK(u))ρKp(u)ρMn−p(u)du=V˜p,p(K,L,M),

i.e.,

[V˜p,p(K,L,M)V(M)]1p≤[V˜q,q(K,L,M)V(M)]1q. 2.4

This gives (1.12). According to the equality condition of Hölder’s integral inequality, we know that equality holds in (2.4) if and only if K and M are dilates. □

We can get the following corollary by (1.5) and (1.7) in (2.4).

Corollary 2.3

Suppose p,q∈R satisfy 1≤p<q. If K∈Kon and M∈Son, then

[V˜p(K,M)V(M)]1p≤[V˜q(K,M)V(M)]1q

with equality if and only if K and M are dilates.

Proof of Theorem 1.5

For p>0,q∈R, K,L∈Kon, and M∈Son. From the definitions of support function and radial function, we know

ρL∗(u)≤hL∗(u) 2.5

with equality if and only if L is a ball centered at the origin.

From (1.3), (1.1), (2.5), and Cauchy’s integral inequality, and noticing that h(B,⋅)=1, we have

V˜p,q(K,L,M)12V˜p,q(K,L∗,M)12=[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)ρMn−q(u)du]12⋅[1n∫Sn−1(hL∗hK)p(αK(u))ρKq(u)ρMn−q(u)du]12=[1n∫Sn−1ρL∗−p(αK(u))hK−p(αK(u))ρKq(u)ρMn−q(u)du]12⋅[1n∫Sn−1hL∗p(αK(u))hK−p(αK(u))ρKq(u)ρMn−q(u)du]12≥1n∫Sn−1(hL∗ρL∗)p2(αK(u))hK−p(αK(u))ρKq(u)ρMn−q(u)du≥1n∫Sn−1hK−p(αK(u))ρKq(u)ρMn−q(u)du=1n∫Sn−1(hBhK)p(αK(u))ρKq(u)ρMn−q(u)du=V˜p,q(K,B,M),

i.e.,

V˜p,q(K,L,M)V˜p,q(K,L∗,M)≥V˜p,q(K,B,M)2. 2.6

Obviously, equality holds in (2.6) if and only if L is a ball centered at the origin. □

If we take M=K or q=n in (2.6) and associate with (1.6) and (1.8), the following corollary can be obtained (see [37]).

Corollary 2.4

Suppose p>0. If K,L∈Kon, then

Vp(K,L)Vp(K,L∗)≥Vp(K,B)2

with equality if and only if L is a ball centered at the origin.

Proof of Theorem 1.6

For p,q∈R and q>n, K,L,M∈Kon. From the definitions of support function and radial function, we know

ρM(u)≤hM(u) 2.7

with equality if and only if M is a ball centered at the origin.

From (1.3), (1.1), (2.7), and Cauchy’s integral inequality, and together with ρ(B,⋅)=1, we obtain

V˜p,q(K,L,M)V˜p,q(K,L,M∗)=[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)ρMn−q(u)du]⋅[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)ρM∗n−q(u)du]=[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)ρMn−q(u)du]⋅[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)hMq−n(u)du]≥[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)(hMρM)q−n2(u)du]2≥[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)du]2=[1n∫Sn−1(hLhK)p(αK(u))ρKq(u)ρBn−q(u)du]2=V˜p,q(K,L,B)2,

This gives (1.14). Obviously, according to the equality of (2.7), we know that equality holds in (1.14) if and only if M is a ball centered at the origin. □

By (1.5) and (1.7), taking L=K or p=0 in Theorem 1.6, we also obtain the following corollary.

Corollary 2.5

Suppose q∈R and q>n. If K,M∈Kon, then

V˜q(K,M)V˜q(K,M∗)≥V˜q(K,B)2

with equality if and only if M is a ball centered at the origin.

Acknowledgements

The authors want to express earnest thankfulness for the referees who provided extremely precious and helpful comments and suggestions. This made the article more accurate and readable.

Authors’ contributions

The authors were devoted equally to the writing of this article. The authors read and approved the final manuscript.

Funding

Research is supported in part by the Natural Science Foundation of China (Grant No. 11371224) and Innovation Foundation of Graduate Student of China Three Gorges University (Grant No. 2018SSPY136).

Competing interests

The authors state that they have no competitive interests.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Bin Chen, Email: chenb121223@163.com.

Weidong Wang, Email: wangwd722@163.com.

References

  • 1.Gardner R.J. Geometric Tomography. 2. Cambridge: Cambridge University Press; 2006. [Google Scholar]
  • 2.Schneider R. Convex Bodies: The Brunn–Minkowski Theory. 2. Cambridge: Cambridge University Press; 2014. [Google Scholar]
  • 3.Lutwak E. The Brunn–Minkowski–Firey theory I: mixed volumes and the Minkowski problem. J. Differ. Geom. 1993;38:131–150. doi: 10.4310/jdg/1214454097. [DOI] [Google Scholar]
  • 4.Böröczky K., Lutwak E., Yang D., Zhang G.Y. The log-Brunn–Minkowski inequality. Adv. Math. 2012;231:1974–1997. doi: 10.1016/j.aim.2012.07.015. [DOI] [Google Scholar]
  • 5.Böröczky K., Lutwak E., Yang D., Zhang G.Y. The logarithmic Minkowski problem. J. Am. Math. Soc. 2013;26:831–852. doi: 10.1090/S0894-0347-2012-00741-3. [DOI] [Google Scholar]
  • 6.Chen W. Lp Minkowski problem with not necessarily positive data. Adv. Math. 2006;201:77–89. doi: 10.1016/j.aim.2004.11.007. [DOI] [Google Scholar]
  • 7.Chou K., Wang X. The Lp Minkowski problem and the Minkowski problem in centroaffine geometry. Adv. Math. 2006;205:33–82. doi: 10.1016/j.aim.2005.07.004. [DOI] [Google Scholar]
  • 8.Fleury B., Guédon O., Paouris G. A stability result for mean width of Lp centroid bodies. Adv. Math. 2007;214:865–877. doi: 10.1016/j.aim.2007.03.008. [DOI] [Google Scholar]
  • 9.He B.W., Leng G.S., Li K.H. Projection problem for symmetric polytopes. Adv. Math. 2006;207:73–90. doi: 10.1016/j.aim.2005.11.006. [DOI] [Google Scholar]
  • 10.Huang Y., Lutwak E., Yang D., Zhang G.Y. Geometric measures in the dual Brunn–Minkowski theory and their associated Minkowski problems. Acta Math. 2016;216:325–388. doi: 10.1007/s11511-016-0140-6. [DOI] [Google Scholar]
  • 11.Lutwak E., Yang D., Zhang G.Y. Lp affine isoperimetric inequalities. J. Differ. Geom. 2000;56:1–13. doi: 10.4310/jdg/1090347527. [DOI] [Google Scholar]
  • 12.Lutwak E., Yang D., Zhang G.Y. On the Lp Minkowski problem. Trans. Am. Math. Soc. 2004;356:4359–4370. doi: 10.1090/S0002-9947-03-03403-2. [DOI] [Google Scholar]
  • 13.Lutwak E., Yang D., Zhang G.Y. Lp John ellipsoids. Proc. Lond. Math. Soc. 2005;90:497–520. doi: 10.1112/S0024611504014996. [DOI] [Google Scholar]
  • 14.Ludwig M. General affine surface area. Adv. Math. 2009;224:2346–2360. doi: 10.1016/j.aim.2010.02.004. [DOI] [Google Scholar]
  • 15.Stancu A. The discrete planar L0 Minkowski problem. Adv. Math. 2002;167:160–174. doi: 10.1006/aima.2001.2040. [DOI] [Google Scholar]
  • 16.Stancu A. On the number of solutions to the discrete two-dimensional L0 Minkowski problem. Adv. Math. 2003;180:290–323. doi: 10.1016/S0001-8708(03)00005-7. [DOI] [Google Scholar]
  • 17.Werner E., Ye D. New Lp affine isoperimetric inequalities. Adv. Math. 2008;218:762–780. doi: 10.1016/j.aim.2008.02.002. [DOI] [Google Scholar]
  • 18.Werner E. On Lp-affine surface area. Indiana Univ. Math. J. 2007;56:2305–2324. doi: 10.1512/iumj.2007.56.3099. [DOI] [Google Scholar]
  • 19.Werner E. Rényi divergence and Lp-affine surface area for convex bodies. Adv. Math. 2012;230:1040–1059. doi: 10.1016/j.aim.2012.03.015. [DOI] [Google Scholar]
  • 20.Wang W.D., Leng G.S. Lp-Mixed affine surface area. J. Math. Anal. Appl. 2007;1:341–354. doi: 10.1016/j.jmaa.2007.01.046. [DOI] [Google Scholar]
  • 21.Wang W.D., Leng G.S. Some affine isoperimetric inequalities associated with Lp-affine surface area. Houst. J. Math. 2008;34:443–454. [Google Scholar]
  • 22.Zhu B.C., Zhou J.Z., Xu W.X. L0 mixed geominimal surface area. J. Math. Anal. Appl. 2015;422:1247–1263. doi: 10.1016/j.jmaa.2014.09.035. [DOI] [Google Scholar]
  • 23.Lutwak E. The Brunn–Minkowski–Firey theory II: affine and geominimal surface areas. Adv. Math. 1996;118:244–294. doi: 10.1006/aima.1996.0022. [DOI] [Google Scholar]
  • 24.Lutwak E. Dual mixed volumes. Pac. J. Math. 1975;58:531–538. doi: 10.2140/pjm.1975.58.531. [DOI] [Google Scholar]
  • 25.Bernig A. The isoperimetrix in the dual Brunn–Minkowski theory. Adv. Math. 2014;254:1–14. doi: 10.1016/j.aim.2013.12.020. [DOI] [Google Scholar]
  • 26.Dulio P., Gardner R.J., Peri C. Characterizing the dual mixed volume via additive functionals. Indiana Univ. Math. J. 2016;65:69–91. doi: 10.1512/iumj.2016.65.5765. [DOI] [Google Scholar]
  • 27.Feng Y.B., Wang W.D. Lp dual mixed geominimal surface area. Glasg. Math. J. 2014;56:224–239. doi: 10.1017/S0017089513000244. [DOI] [Google Scholar]
  • 28.Gardner R.J. The dual Brunn–Minkowski theory for bounded Borel set: dual affine quermassintegrals and inequalities. Adv. Math. 2007;216:358–386. doi: 10.1016/j.aim.2007.05.018. [DOI] [Google Scholar]
  • 29.Jiménez C.H., Villanueva I. Characterization of dual mixed volumes via polymeasures. J. Math. Anal. Appl. 2015;426:688–699. doi: 10.1016/j.jmaa.2015.01.068. [DOI] [Google Scholar]
  • 30.Milman E. Dual mixed volumes and the slicing problem. Adv. Math. 2005;207:565–598. [Google Scholar]
  • 31.Wang W., He B.W. Lp-dual affine surface area. J. Math. Anal. Appl. 2008;348:746–751. doi: 10.1016/j.jmaa.2008.08.006. [DOI] [Google Scholar]
  • 32.Wan X.Y., Wang W.D. Lp-dual mixed affine surface area. Ukr. Math. J. 2016;68:601–609. doi: 10.1007/s11253-016-1250-1. [DOI] [Google Scholar]
  • 33.Wang J.Y., Wang W.D. Lp-dual affine surface area forms of Busemann–Petty type problems. Proc. Indian Acad. Sci. 2015;125:71–77. [Google Scholar]
  • 34.Wang W.D., Leng G.S. Lp-dual mixed quermassintegrals. Indian J. Pure Appl. Math. 2005;36:177–188. [Google Scholar]
  • 35.Zhang T., Wang W.D., Si L. The mixed Lp-dual affine surface area for multiple star bodies. J. Nonlinear Sci. Appl. 2016;9:2813–2822. doi: 10.22436/jnsa.009.05.76. [DOI] [Google Scholar]
  • 36.Lutwak E., Yang D., Zhang G.Y. Lp dual curvature measures. Adv. Math. 2018;329:85–132. doi: 10.1016/j.aim.2018.02.011. [DOI] [Google Scholar]
  • 37.Wei B., Wang W.D. Some inequalities for the Lp-mixed quermassintegrals. Wuhan Univ. J. Nat. Sci. 2013;18:233–236. doi: 10.1007/s11859-013-0920-1. [DOI] [Google Scholar]

Articles from Journal of Inequalities and Applications are provided here courtesy of Springer

RESOURCES