Skip to main content

Advertisement

Fixed point solutions for variational inequalities in image restoration over q-uniformly smooth Banach spaces

Article metrics

  • 736 Accesses

  • 2 Citations

Abstract

In this paper, we introduce new implicit and explicit iterative methods for finding a common fixed point set of an infinite family of strict pseudo-contractions by the sunny nonexpansive retractions in a real q-uniformly and uniformly convex Banach space which admits a weakly sequentially continuous generalized duality mapping. Then we prove the strong convergence under mild conditions of the purposed iterative scheme to a common fixed point of an infinite family of strict pseudo-contractions which is a solution of some variational inequalities. Furthermore, we apply our results to study some strong convergence theorems in L p and p spaces with 1<p<. Our results mainly improve and extend the results announced by Ceng et al. (Comput. Math. Appl. 61:2447-2455, 2011) and many authors from Hilbert spaces to Banach spaces. Finally, we give some numerical examples for support our main theorem in the end of the paper.

MSC:47H09, 47H10, 47H17, 47J25, 49J40.

1 Introduction

Let C 1 , C 2 ,, C n be nonempty, closed, and convex subsets of a real Hilbert space H such that i = 1 n C i . The problem of image recovery in a Hilbert space setting by using convex of metric projections P C i , may be stated as follows: the original unknown image z is known a priori to belong to the intersection of { C i } i = 1 n ; given only the metric projections P C i of H onto C i for i=1,2,,n recover z by an iterative scheme. Youla and Webb [1] first used iterative methods for applied in image restoration. The problems of image recovery have been studied in a Banach space setting by Kitahara and Takahashi [2] (see also [3, 4]) by using convex combinations of sunny nonexpansive retractions in uniformly convex Banach spaces. On the other hand, Alber [5] studied the problem of image recovery by the products of generalized projections in a uniformly convex and uniformly smooth Banach space whose duality mapping is weakly sequentially continuous (see also [6, 7]). Nakajo et al. [8] and Kimura et al. [9] considered this problem by the sunny nonexpansive retractions and proved convergence of the iterative sequence to a common point of countable nonempty, closed, and convex subsets in a uniformly convex and smooth Banach space, and in a strictly convex, smooth and reflexive Banach space having the Kadec-Klee property, respectively. Some iterative methods have been studied in problem of image recovery by numerous authors (see [25, 1012]).

The problems of image recovery are connected with the convex feasibility problem, convex minimization problems, multiple-set split feasibility problems, common fixed point problems, and variational inequalities. In particular, variational inequality theory has been studied widely in several branches of pure and applied sciences. This field is dynamics and is experiencing an explosive growth in both theory and applications. Indeed, applications of the variational inequalities span as diverse disciplines as differential equations, time-optimal control, optimization, mathematical programming, mechanics, finance, and so on. Note that most of the variational problems, including minimization or maximization of functions, variational inequality problems, quasivariational inequality problems, decision and management sciences, and engineering sciences problems. Recently, some iterative methods have been developed for solving the fixed point problems and variational inequality problems in q-uniformly smooth Banach spaces by numerous authors (see [1324]).

Let A be a strongly positive bounded linear operator on H, that is, there exists a constant γ ¯ >0 such that

Ax,x γ ¯ x 2 for all xH.
(1.1)

Remark 1.1 From the definition of operator A, we note that a strongly positive bounded linear operator A is a A-Lipschitzian and η-strongly monotone operator.

A typical problem is to minimize a quadratic function over the set of the fixed points of a nonexpansive mapping on a real Hilbert space H:

min x C 1 2 Ax,xx,u,
(1.2)

where C is the fixed point set of a nonexpansive mapping T on H and u is a given point in H.

In 2006, Marino and Xu [25] introduced and considered the following a general iterative method:

x n + 1 = α n γf( x n )+(I α n A)T x n ,n0,
(1.3)

where A is a strongly positive bounded linear operator on a real Hilbert space H. They proved that if the sequence { α n } satisfies appropriate conditions, then the sequence { x n } generated by (1.3) converges strongly to the unique solution of the variational inequality

( γ f A ) x , x x 0,xFix(T),
(1.4)

which is the optimality condition for the minimization problem

min x C 1 2 Ax,xh(x),
(1.5)

where C is the fixed point set of a nonexpansive mapping T and h is a potential function for γf (i.e., h (x)=γf(x) for all xH).

On the other hand, Yamada [26] introduced a hybrid steepest descent method for a nonexpansive mapping T as follows:

x n + 1 =T x n μ λ n F(T x n ),n0,
(1.6)

where F is a κ-Lipschitzian and η-strongly monotone operator on a real Hilbert space H with constants κ,η>0 and 0<μ< 2 η κ 2 . He proved that if { λ n } satisfy the appropriate conditions, then the sequence { x n } generated by (1.6) converges strongly to the unique solution of the variational inequality

F x , x x 0,xFix(T).
(1.7)

Tian [27] combined the iterative method (1.3) with the Yamada method (1.6) and considered a general iterative method for a nonexpansive mapping T on a real Hilbert space H as follows:

x n + 1 = α n γf( x n )+(I α n μF)T x n ,n0.
(1.8)

Then he proved that the sequence { x n } generated by (1.8) converges strongly to the unique solution of variational inequality

( γ f μ F ) x , x x 0,xFix(T).
(1.9)

In 2011, Ceng et al. [28] combined the iterative method (1.3) with Tian’s method (1.8) and consider the following a general composite iterative method:

x n + 1 =(I α n A)T x n + α n [ T x n β n ( μ F T x n γ f ( x n ) ) ] ,n0,
(1.10)

where A is a strongly positive bounded linear operator on H with coefficient γ ¯ (1,2), and { α n }(0,1) and { β n }(0,1] satisfy appropriate conditions. Then they proved that the sequence { x n } generated by (1.10) converges strongly to the unique solution x C of the variational inequality

( I A ) x , x x 0,xC,
(1.11)

where C=Fix(T).

In this paper, motivated by the above facts, we introduce new implicit and explicit iterative methods for finding a common fixed point set of an infinite family of strict pseudo-contractions by the sunny nonexpansive retractions in a real q-uniformly and uniformly convex Banach space X which admits a weakly sequentially continuous generalized duality mapping. Consequently, we prove the strong convergence under mild conditions of the purposed iterative scheme to a common fixed point of an infinite family of strict pseudo-contractions of nonempty, closed, and convex subsets of X which is a solution of some variational inequalities. Furthermore, we apply our results to the study of some strong convergence theorems in L p and p spaces with 1<p<. Our results extend the main result of Ceng et al. [28] in several aspects and the work of many authors from Hilbert spaces to Banach spaces. Finally, we give some numerical examples to support our main theorem in the end of the paper.

2 Preliminaries

Throughout this paper, we denote by X and X a real Banach space and the dual space of X, respectively. Let q>1 be a real number. The generalized duality mapping J q :X 2 X is defined by

J q (x)= { f X : x , f = x q , f = x q 1 } ,

where , denotes the duality pairing between X and X . In particular, J q = J 2 is called the normalized duality mapping and J q (x)= x q 2 J 2 (x) for x0. If X:=H is a real Hilbert space, then J=I, where I is the identity mapping. It is well known that if X is smooth, then J q is single-valued, which is denoted by j q (see [29]).

A Banach space X is said to be strictly convex if x + y 2 <1 for all x,yX with x=y=1 and xy. A Banach space X is said to be uniformly convex if, for each ϵ>0, there exists δ>0 such that for x,yX with x,y1 and xyϵ, x + y 2 1δ holds. Let S(X)={xX:x=1}. The norm of X is said to be Gâteaux differentiable (or X is said to be smooth) if the limit

lim t 0 x + t y x t

exists for each x,yS(X). The norm of X is said to be uniformly Gâteaux differentiable, if, for each yS(X), the limit is attained uniformly for xS(X).

Let ρ X :[0,)[0,) be the modulus of smoothness of X defined by

ρ X (τ)=sup { 1 2 ( x + y + x y ) 1 : x S ( X ) , y τ } .

A Banach space X is said to be uniformly smooth if ρ X ( t ) t 0 as t0. Suppose that q>1, then X is said to be q-uniformly smooth if there exists c>0 such that ρ X (t)c t q for all t>0. It is shown in [30] (see also [31]) that there is no Banach space which is q-uniformly smooth with q>2. If X is q-uniformly smooth, then X is uniformly smooth. It is well known that each uniformly convex Banach space X is reflexive and strictly convex and every uniformly smooth Banach space X is a reflexive Banach space with uniformly Gâteaux differentiable norm (see [29]). Typical examples of both uniformly convex and uniformly smooth Banach spaces are L p , where p>1. More precisely, L p is min{p,2}-uniformly smooth for every p>1.

Let C be a nonempty, closed, and convex subset of X and T be a self-mapping on C. We denote the fixed points set of the mapping T by Fix(T)={xC:Tx=x}.

Definition 2.1 A mapping T:CC is said to be:

  1. (i)

    λ-strictly pseudo-contractive [32] if, for all x,yC, there exist λ>0 and j q (xy) J q (xy) such that

    T x T y , j q ( x y ) x y q λ ( I T ) x ( I T ) y q ,
    (2.1)

    or equivalently

    ( I T ) x ( I T ) y , j q ( x y ) λ ( I T ) x ( I T ) y q .
    (2.2)
  2. (ii)

    L-Lipschitzian if, for all x,yC, there exists a constant L>0 such that

    TxTyLxy.

If 0<L<1, then T is a contraction and if L=1, then T is a nonexpansive mapping. By the definition, we know that every λ-strictly pseudo-contractive mapping is ( 1 + λ λ )-Lipschitzian (see [33]).

Remark 2.2 Let C be a nonempty subset of a real Hilbert space H and T:CC be a mapping. Then T is said to be k-strictly pseudo-contractive [32] if, for all x,yC, there exists k[0,1) such that

T x T y 2 x y 2 +k ( I T ) x ( I T ) y 2 .
(2.3)

It is well known that (2.3) is equivalent to the following inequality:

TxTy,xy x y 2 1 k 2 ( I T ) x ( I T ) y 2 .

A mapping F:CX is said to be accretive if, for all x,yC, there exists j q (xy) J q (xy) such that

F x F y , j q ( x y ) 0.

For some η>0, F:CX is said to be strongly accretive if, for all x,yC, there exists j q (xy) J q (xy) such that

F x F y , j q ( x y ) η x y q .

Remark 2.3 If X:=H is a real Hilbert space, accretive and strongly accretive mappings coincide with monotone and strongly monotone mappings, respectively.

Let D be a nonempty subset of C. A mapping Q:CD is said to be sunny [34] if

Q ( Q x + t ( x Q x ) ) =Qx,

whenever Qx+t(xQx)C for xC and t0. A mapping Q:CD is said to be retraction if Qx=x for all xD. Furthermore, Q is a sunny nonexpansive retraction from C onto D if Q is a retraction from C onto D which is also sunny and nonexpansive. A subset D of C is called a sunny nonexpansive retraction of C if there exists a sunny nonexpansive retraction from C onto D. It is well known that if X:=H is a real Hilbert space, then a sunny nonexpansive retraction Q is coincident with the metric projection from X onto C.

Lemma 2.4 ([14])

Let C be a closed and convex subset of a smooth Banach space X. Let D be a nonempty subset of C. Let Q:CD be a retraction and let j, j q be the normalized duality mapping and generalized duality mapping on X, respectively. Then the following are equivalent:

  1. (a)

    Q is sunny and nonexpansive.

  2. (b)

    Q x Q y 2 xy,j(QxQy) for all x,yC.

  3. (c)

    xQx,j(yQx)0 for all xC and yD.

  4. (d)

    xQx, j q (yQx)0 for all xC and yD.

Lemma 2.5 ([35])

Suppose that q>1. Then the following inequality holds:

ab 1 q a q + ( q 1 q ) b q q 1

for arbitrary positive real numbers a, b.

In a real q-uniformly smooth Banach space, Xu [36] proved the following important inequality:

Lemma 2.6 ([36])

Let X be a real q-uniformly smooth Banach space. Then the following inequality holds:

x + y q x q +q y , J q ( x ) + C q y q
(2.4)

for all x,yX and for some C q >0.

Remark 2.7 The constant C q satisfying (2.4) is called the best q-uniform smoothness constant.

Lemma 2.8 ([21])

Let C be a nonempty and convex subset of a real q-uniformly smooth Banach space X and T:CC be a λ-strict pseudo-contraction. For γ(0,1), define Sx=(1γ)x+γTx. Then, as γ(0,ν), ν=min{1, ( q λ C q ) 1 q 1 }, S:CC is nonexpansive and Fix(S)=Fix(T), where C q is the best q-uniform smoothness constant.

Definition 2.9 ([37])

Let C be a nonempty, closed, and convex subset of a real q-uniformly smooth Banach space X. Let T n , k = θ n , k S k +(1 θ n , k )I, where S k :CC is λ k -strict pseudo-contraction and { t n } be a nonnegative real sequence with 0 t n 1, nN. For n1, define a mapping W n :CC as follows:

U n , n + 1 = I , U n , n = t n T n , n U n , n + 1 + ( 1 t n ) I , U n , k = t k T n , k U n , k + 1 + ( 1 t k ) I , U n , k 1 = t k 1 T n , k 1 U n , k + ( 1 t k 1 ) I , U n , 2 = t 2 T n , 2 U n , 3 + ( 1 t 2 ) I , W n = U n , 1 = t 1 T n , 1 U n , 2 + ( 1 t 1 ) I .
(2.5)

Such a mapping W n is called the W-mapping generated by T n , n , T n , n 1 ,, T n , 1 and t n , t n 1 ,, t 1 .

Throughout this paper, we will assume that { θ n , k } satisfies the following conditions:

(H1) θ n , k (0,ν], ν=min{1, ( q λ ¯ C q ) 1 q 1 } with λ ¯ =inf λ k >0, n,kN;

(H2) | θ n + 1 , k θ n , k | a n , nN and 1kn with n = 1 a n <;

The hypothesis (H2) secures the existence of lim n θ n , k , kN. Set θ 1 , k := lim n θ n , k , nN. Furthermore, we assume

(H3) θ 1 , k >0, kN.

It is obvious that θ 1 , k satisfies (H1). Using condition (H3), from T n , k = θ n , k S k +(1 θ n , k )I, we define mappings T 1 , k x:= lim n T n , k x= θ 1 , k S k x+(1 θ 1 , k )x, xC.

Lemma 2.10 ([37])

Let C be a nonempty, closed, and convex subset of a real q-uniformly smooth and strictly convex Banach space X. Let T n , i = θ n , i S i +(1 θ n , i )I, where S i :CC (i=1,2,) is λ i -strict pseudo-contraction with n = 1 Fix( S n ) and inf λ i >0. Let t 1 , t 2 , be nonnegative real numbers such that 0< t n b<1, n1. Assume the sequence { θ n , k } satisfies (H1)-(H3). Then

  1. (1)

    W n is nonexpansive and Fix( W n )= n = 1 Fix( S n ) for each n1;

  2. (2)

    for each xC and for each positive integer k, the limit lim n U n , k exists;

  3. (3)

    the mapping W:CC defined by

    Wx:= lim n W n x= lim n U n , 1 x,xC,

is a nonexpansive mapping satisfying Fix(W)= n = 1 Fix( S n ) and it is called the W-mapping generated by S 1 , S 2 , and t 1 , t 2 , and θ n , k , nN and 1kn.

Lemma 2.11 ([37])

Let C be a nonempty, closed, and convex subset of a real q-uniformly smooth and strictly convex Banach space X. Let T n , i = θ n , i S i +(1 θ n , i )I, where S i :CC (i=1,2,) is λ i -strict pseudo-contraction with n = 1 Fix( S n ) and inf λ i >0. Let t 1 , t 2 , be nonnegative real numbers such that 0< t n b<1, n1. Assume the sequence { θ n , k } satisfies (H1)-(H3). If { ω n } is a bounded sequence in C, then

lim n W ω n W n ω n =0.

In the following, the notation and → denote the weak and strong convergence, respectively. The duality mapping J q from a smooth Banach space X into X is said to be weakly sequentially continuous generalized duality mapping if, for all { x n }X, x n x implies J q ( x n ) J q (x).

A Banach space X is said to be satisfy Opial’s condition [38], that is, for any sequence { x n } in X, x n x implies that

lim inf n x n x< lim inf n x n y,yX with xy.

By Theorem 3.2.8 in [39], it is well known that if X admits a weakly sequentially continuous generalized duality mapping, then X satisfies Opial’s condition.

Lemma 2.12 ([13])

Let C be a nonempty, closed, and convex subset of a real q-uniformly smooth Banach space X which admits weakly sequentially continuous generalized duality mapping j q from X into X . Let T:CC be a nonexpansive mapping. Then, for all { x n }C, if x n x and x n T x n 0, then x=Tx.

Lemma 2.13 ([40])

Let { a n }, { μ n }, and { δ n } be real sequences of nonnegative numbers such that

a n + 1 (1 σ n ) a n + μ n + δ n ,n1,

where σ n (0,1), n = 1 σ n =, μ n =( σ n ) and n = 1 δ n <. Then lim n a n =0.

3 Main results

In order to prove our main result, the following lemma is needed.

Lemma 3.1 Let C be a nonempty, closed, and convex subset of a real q-uniformly smooth Banach space X with the best q-uniform smoothness constant C q >0. Let F:CX be a κ-Lipschitzian and η-strongly accretive operator with constants κ,η>0. Let 0<μ< ( q η C q κ q ) 1 q 1 and τ=μ(η C q μ q 1 κ q q ). Then for t(0,min{1, 1 q τ }), the mapping S:CX defined by S:=ItμF is a contraction with constant 1tτ.

Proof Since 0<μ< ( q η C q κ q ) 1 q 1 and t(0,min{1, 1 q τ }). This implies that 1tτ(0,1). From Lemma 2.6, for all x,yC, we have

S x S y q = ( I t μ F ) x ( I t μ F ) y q = ( x y ) t μ ( F x F y ) q x y q q t μ F x F y , j q ( x y ) + C q t q μ q F x F y q x y q q t μ η x y q + C q t q μ q κ q x y q [ 1 t μ ( q η C q μ q 1 κ q ) ] x y q = [ 1 t μ q ( η C q μ q 1 κ q q ) ] x y q [ 1 t μ ( η C q μ q 1 κ q q ) ] q x y q = ( 1 t τ ) q x y q .

It follows that

SxSy(1tτ)xy.

Hence, we have S:=ItμF is a contraction with constant 1tτ. □

Lemma 3.2 Let C be a nonempty, closed, and convex subset of a real q-uniformly smooth Banach space X and G:CX be a mapping.

  1. (i)

    If G is a δ-strongly accretive and λ-strictly pseudo-contractive mapping wit δ+λ>1, then IG is a contraction with constant L δ , λ := ( 1 δ λ ) 1 q .

  2. (ii)

    If G is a δ-strongly accretive and λ-strictly pseudo-contractive mapping with δ+λ>1. For a fixed number t(0,1), then ItG is a contraction with constant 1(1 L δ , λ )t.

Proof (i) For all x,yC, from (2.2), we have

λ ( I G ) x ( I G ) y q x y q G x G y , j q ( x y ) ( 1 δ ) x y q .

Observe that

δ+λ>1 ( 1 δ λ ) 1 q (0,1).

It follows that

( I G ) x ( I G ) y ( 1 δ λ ) 1 q xy:= L δ , λ xy.

Hence, IG is a contraction with constant L δ , λ .

(ii) Since IG is a contraction with constant L δ , λ . For all t(0,1), we have

( I t G ) x ( I t G ) y = ( x y ) t ( G x G y ) = ( 1 t ) ( x y ) + t [ ( I G ) x ( I G ) y ] ( 1 t ) x y + t ( I G ) x ( I G ) y ( 1 ( 1 L δ , λ ) t ) x y .

Hence, ItG is a contraction with constant 1(1 L δ , λ )t. This completes the proof. □

3.1 Implicit iteration scheme

Let C be a nonempty, closed, and convex subset of a real reflexive and q-uniformly smooth Banach space X which admits a weakly sequentially continuous generalized duality mapping j q . Let Q C be a sunny nonexpansive retraction from X onto C. Let F:CX be a κ-Lipschitzian and η-strongly accretive operator with constants κ,η>0, G:CX be a δ-strongly accretive and λ-strictly pseudo-contractive mapping with δ+λ>1, V:CX be an L-Lipschitzian mapping with constant L0 and T:CC be a nonexpansive mapping such that Fix(T). Let 0<μ< ( q η C q κ q ) 1 q 1 and 0γL<τ, where τ=μ(η C q μ q 1 κ q q ). For each σ( L δ , λ τ γ L ,min{1, 1 q τ , 1 + L δ , λ τ γ L }) and t(0,1), we define a mapping S t :CC defined by

Sx:= Q C [ ( I t G ) T x + t ( T x σ ( μ F T x γ V x ) ) ] ,xC.

It is easy to see immediately that S t is a contraction. Indeed, for all x,yC, from Lemmas 3.1 and 3.2(ii), we have

S t x S t y = Q C [ ( I t G ) T x + t ( T x σ ( μ F T x γ V x ) ) ] Q C [ ( I t G ) T y + t ( T y σ ( μ F T y γ V y ) ) ] ( I t G ) ( T x T y ) + t [ ( I σ μ F ) ( T x T y ) + σ γ ( V x V y ) ] ( 1 t ( 1 L δ , λ ) ) x y + t [ σ γ V x V y + ( I σ μ F ) ( T x T y ) ] ( 1 t ( 1 L δ , λ ) ) x y + t ( 1 σ ( τ γ L ) ) x y = [ 1 t ( σ ( τ γ L ) L δ , λ ) ] x y = ( 1 t θ ) x y ,
(3.1)

where θ:=σ(τγL) L δ , λ . Since τγL>0 and L δ , λ (0,1), observe that

L δ , λ τ γ L <σ<min { 1 , 1 q τ , 1 + L δ , λ τ γ L } 1 + L δ , λ τ γ L .

It follows that

σ< 1 + L δ , λ τ γ L θ=σ(τγL) L δ , λ <1

and

L δ , λ τ γ L <σθ=σ(τγL) L δ , λ >0.

This implies that θ=σ(τγL) L δ , λ (0,1), which together with t(0,1) gives

1t ( σ ( τ γ L ) L δ , λ ) (0,1).

Hence S t is a contraction. By the Banach contraction principle, S t has a unique fixed point, denote by x t , which uniquely solves the fixed point equation

x t = Q C [ ( I t G ) T x t + t ( T x t σ ( μ F T x t γ V x t ) ) ] .
(3.2)

The following proposition summarizes the properties of the net { x t }.

Proposition 3.3 Let { x t } be defined by (3.2). Then the following hold:

  1. (i)

    { x t } is bounded for each t(0,1);

  2. (ii)

    lim t 0 x t T x t =0;

  3. (iii)

    { x t } defines a continuous curve from (0,1) into C.

Proof (i) Take pFix(T), and denote a mapping S t :CC by

S t x:= Q C [ ( I t G ) T x + t ( T x σ ( μ F T x γ V x ) ) ] ,xC.

From (3.1), we have

x t p S t x t S t p + S t p p ( 1 t θ ) x t p + Q C [ ( I t G ) T p + t ( T p σ ( μ F T p γ V p ) ) ] Q C p ( 1 t θ ) x t p + t G p + p σ ( μ F p γ V p ) ( 1 t θ ) x t p + t [ I G p + σ μ F p + σ γ V p ] ,

where θ:=σ(τγL) L δ , λ . It follows that

x t p I G p + σ μ F p + σ γ V p θ .

Hence { x t } is bounded, so are {V x t }, {FT x t }, and {GT x t }.

  1. (ii)

    By definition of { x t }, we have

    x t T x t = Q C [ ( I t G ) T x t + t ( T x t σ ( μ F T x t γ V x t ) ) ] Q C T x t t ( I G ) T x t σ ( μ F T x t γ V x t ) 0 as  t 0 .
  2. (iii)

    Take t, t 0 (0,1) and calculate

    x t x t 0 = Q C [ ( I t G ) T x t + t ( T x t σ ( μ F T x t γ V x t ) ) ] Q C [ ( I t 0 G ) T x t 0 + t ( T x t 0 σ ( μ F T x t 0 γ V x t 0 ) ) ] ( t 0 t ) G T x t + ( I t 0 G ) ( T x t T x t 0 ) + ( t t 0 ) [ T x t σ ( μ F T x t γ V x t ) ] + t 0 [ T x t σ ( μ F T x t 0 γ V x t 0 ) [ T x t 0 σ ( μ F T x t 0 γ V x t 0 ) ] ] = ( t 0 t ) G T x t + ( I t 0 G ) ( T x t T x t 0 ) + ( t t 0 ) [ T x t σ ( μ F T x t γ V x t ) ] + t 0 [ σ γ ( V x t V x t 0 ) + ( I σ μ F ) ( T x t T x t 0 ) ] | t t 0 | G T x t + ( 1 ( 1 L δ , λ ) ) x t x t 0 + | t t 0 | T x t σ ( μ F T x t γ V x t ) + t 0 ( 1 σ ( τ γ L ) ) x t x t 0 .

It follows that

x t x t 0 G T x t + T x t σ ( μ F T x t γ V x t ) t 0 ( σ ( τ γ L ) L δ , λ ) |t t 0 |.

Since {V x t }, {FT x t }, and {GT x t } are bounded. Hence { x t } defines a continuous curve from (0,1) into C. □

Theorem 3.4 Assume that { x t } is defined by (3.2), then { x t } converges strongly to x Fix(T) as t0, where x is the unique solution of the variational inequality

( G I + σ ( μ F γ V ) ) x , j q ( x v ) 0,vFix(T).
(3.3)

Proof We observe that

C q μ q 1 κ q q > 0 η C q μ q 1 κ q q < η μ ( η C q μ q 1 κ q q ) < μ η τ < μ η .
(3.4)

It follows that

0γL<τ<μη.
(3.5)

First, we show the uniqueness of solution of the variational inequality. Suppose that x ˜ , x Fix(T) are solutions of (3.3), then

( G I + σ ( μ F γ V ) ) x , j q ( x x ˜ ) 0
(3.6)

and

( G I + σ ( μ F γ V ) ) x ˜ , j q ( x ˜ x ) 0.
(3.7)

Adding up (3.6) and (3.7), and from Lemma 3.2(i), we obtain

0 ( G I + σ ( μ F γ V ) ) x ( G I + σ ( μ F γ V ) ) x ˜ , j q ( x x ˜ ) = ( G I ) x ( G I ) x ˜ , j q ( x x ˜ ) + σ ( μ F γ V ) x ( μ F γ V ) x ˜ , j q ( x x ˜ ) = ( I G ) x ( I G ) x ˜ , j q ( x x ˜ ) + σ μ F x F x ˜ , j q ( x x ˜ ) σ γ V x V x ˜ , j q ( x x ˜ ) L δ , λ x x ˜ q + σ μ η x x ˜ q σ γ V x V x ˜ x x ˜ q 1 ( σ ( μ η γ L ) L δ , λ ) x x ˜ q .

On the other hand, we observe from (3.5) that

L δ , λ τ γ L < σ L δ , λ < σ ( τ γ L ) L δ , λ < σ ( μ η γ L ) 0 < σ ( μ η γ L ) L δ , λ .
(3.8)

Note that (3.8) implies that x = x ˜ and the uniqueness is proved. Below, we use x ˜ to denote the unique solution of the variational inequality (3.3).

Next, we show that x t x as t0. Set x t = Q C y t , where y t =(ItG)T x t +t(T x t σ(μFT x t γV x t )). Assume that { t n }(0,1) is a sequence such that t n 0 as n. Put x n := x t n and y n := y t n . For zFix(T), we note that

x n z = Q C y n y n + y n z = Q C y n y n + ( I t n G ) ( T x n z ) + t n ( T x n σ ( μ F T x n γ V x n ) G z ) = Q C y n y n + ( I t n G ) ( T x n z ) + t n [ ( I σ μ F ) T x n + σ γ V x n G z ] = Q C y n y n + ( I t n G ) ( T x n z ) + t n [ ( I σ μ F ) ( T x n z ) + σ γ ( V x n V z ) ] + t n [ ( I σ μ F ) z + σ γ V z G z ] .
(3.9)

By Lemma 2.4, we have

Q C y n y n , j q ( Q C y n z ) 0.
(3.10)

It follows from (3.9) and (3.10) that

x n z q = Q C y n y n , j q ( Q C y n z ) + y n z , j q ( x n z ) ( I t n G ) ( T x n z ) , j q ( x n z ) + t n ( I σ μ F ) ( T x n z ) , j q ( x n z ) + t n σ γ V x n V z , j q ( x n z ) + t n ( I σ μ F ) z + σ γ V z G z , j q ( x n z ) [ 1 t n ( σ ( τ γ L ) L δ , λ ) ] x n z q + t n ( I σ μ F ) z + σ γ V z G z , j q ( x n z ) ,

which implies that

x n z q 1 σ ( τ γ L ) L δ , λ ( I σ μ F ) z + σ γ V z G z , j q ( x n z ) .

In particular, we have

x n i z q 1 σ ( τ γ L ) L δ , λ ( I σ μ F ) z + σ γ V z G z , j q ( x n i z ) .
(3.11)

By reflexivity of a Banach space X and boundedness of { x n }, there exists a subsequence { x n i } of { x n } such that x n i x ˜ as i. Since a Banach space X has a weakly sequentially continuous generalized duality mapping and by (3.11), we obtain x n i x ˜ . By Proposition 3.3(ii), we have x n i T x n i 0 as i. Hence, it follows from Lemma 2.12 that x ˜ Fix(T).

Next, we show that x ˜ solves the variational inequality (3.3). We note that

x t = Q C y t = Q C y t y t +(ItG)Tx+t ( T x t σ ( μ F T x t γ V x t ) ) ,

we derive

( G I + σ ( μ F γ V ) ) x t = 1 t ( Q C y t y t ) 1 t ( ( I t G ) ( I T ) x t + t ( I σ μ F ) ( I T ) x t ) .
(3.12)

Since IT is accretive (i.e., (IT)x(IT)y, j q (xy)0 for x,yC). For all vFix(T), it follows from (3.10) and (3.12) that

( G I + σ ( μ F γ V ) ) x t , j q ( x t v ) = 1 t Q C y t y t , j q ( Q C y t v ) 1 t ( I t G ) ( I T ) x t , j q ( x t v ) ( I σ μ F ) ( I T ) x t , j q ( x t v ) 1 t ( I T ) x t ( I T ) v , j q ( x t v ) + G ( I T ) x t , j q ( x t v ) ( I T ) x t ( I T ) v , j q ( x t v ) + σ μ F ( I T ) x t , j q ( x t v ) G ( I T ) x t , j q ( x t v ) + σ μ F ( I T ) x t , j q ( x t v ) G x t T x t x t v q 1 + σ μ F x t T x t x t v q 1 x t T x t M 1 ,
(3.13)

where M 1 >0 is an appropriate constant such that M 1 = sup t ( 0 , 1 ) {G x t v q 1 ,σμF x t v q 1 }. Now, replacing t in (3.13) with t n and taking the limit as n, we notice that x t n T x t n x ˜ T x ˜ =0, we obtain

( G I + σ ( μ F γ V ) ) x ˜ , j q ( x ˜ v ) 0.

That is, x ˜ Fix(T) is the solution of the variational inequality (3.3). Consequently, x ˜ = x by uniqueness. In a summary, we have shown that each cluster point of { x t } is equal to x . Therefore x t x as t0. This completes the proof. □

3.2 Explicit iteration scheme

Theorem 3.5 Let C be a nonempty, closed, and convex subset of a real q-uniformly smooth and uniformly convex Banach space X which admits a weakly sequentially continuous generalized duality mapping j q . Let Q C be a sunny nonexpansive retraction such that X onto C. Let F:CX be a κ-Lipschitzian and η-strongly accretive operator with constants κ,η>0, G:CX be a δ-strongly accretive and λ-strictly pseudo-contractive mapping with δ+λ>1, V:CX be an L-Lipschitzian mapping with constant L0. Let { S i } i = 1 be an infinite family of λ i -strictly pseudo-contractive mapping from C into itself such that F:= i = 1 Fix( S i ). For given x 1 C, define the sequence { x n } by

x n + 1 = Q C [ ( I α n G ) W n x n + α n ( W n x n σ ( μ F W n x n γ V x n ) ) ] ,n1,
(3.14)

where { α n } is a sequence in (0,1) which satisfies the following conditions:

(C1) lim n α n =0 and n = 1 α n =;

(C2) | α n + 1 α n |( α n )+ σ n with n = 1 σ n <.

Suppose in addition that { θ n , k } satisfies (H1)-(H3). Then the sequence { x n } defined by (3.14) converges strongly to x F as n, where x is the unique solution of the variational inequality

( G I + σ ( μ F γ V ) ) x , j q ( x v ) 0,vF.
(3.15)

Proof From the condition (C1), we may assume, without loss of generality, that α n min{1, 1 q τ } for all nN. First, we show that { x n } is bounded. Take pF, and denote a mapping S n α n :CC by

S n α n x:= Q C [ ( I α n G ) W n x + α n ( W n x σ ( μ F W n x γ V x ) ) ] ,xC.

Then we have

S n α n p= Q C [ ( I α n G ) W n p + α n ( W n p σ ( μ F W n p γ V p ) ) ] .

From (3.1), we have

x n + 1 p S n α n x n S n α n p + S n α n p p ( 1 α n θ ) x n p + Q C [ ( I α n G ) p + α n ( p σ ( μ F p γ V p ) ) ] Q C p ( 1 α n θ ) x n p + α n G p + p σ ( μ F p γ V p ) ( 1 α n θ ) x n p + α n ( I G p + σ μ F p γ V p ) ( 1 α n θ ) x n p + α n θ I G p + σ μ F p + γ V p θ max { x n p , I G p + σ μ F p + γ V p θ } ,

where θ:=σ(τγL) L δ , λ . By induction, we obtain

x n pmax { x 1 p , I G p + σ μ F p + γ V p θ } ,n1.

Hence, { x n } is bounded, so are {V x n }, {F W n x n }, and {G W n x n }.

Next, we show that x n + 1 x n 0 as n. Set S n α n x n = Q C y n , where y n =(I α n G) W n x n + α n ( W n x n σ(μF W n x n γV x n )). From (2.5), we have

W n + 1 x n W n x n = t 1 T n + 1 , 1 U n + 1 , 2 x n + ( 1 t 1 ) x n t 1 T n , 1 U n , 2 ( 1 t 1 ) x n = t 1 T n + 1 , 1 U n + 1 , 2 x n T n , 1 U n , 2 x n = t 1 ( θ n + 1 , 1 S 1 + ( 1 θ n + 1 , 1 ) ) U n + 1 , 2 x n T n , 1 U n , 2 x n = t 1 ( θ n , 1 S 1 + ( 1 θ n , 1 ) ) U n + 1 , 2 x n T n , 1 U n , 2 x n + ( θ n + 1 , 1 θ n , 1 ) ( S 1 U n + 1 , 2 x n U n + 1 , 2 x n ) t 1 T n , 1 U n + 1 , 2 x n T n , 1 U n , 2 x n + t 1 | θ n + 1 , 2 θ n , 1 | S 1 U n + 1 , 2 x n U n + 1 , 2 x n t 1 T n , 1 U n + 1 , 2 x n T n , 1 U n , 2 x n + t 1 | θ n + 1 , 2 θ n , 1 | M t 1 T n , 1 U n + 1 , 2 x n T n , 1 U n , 2 x n + t 1 a n M 1 i = 1 n t i U n + 1 , n + 1 x n U n , n + 1 x n + ( a n j = 1 n i = 1 j t i ) M 1 i = 1 n t i t n + 1 T n + 1 , n + 1 x n + ( 1 t n + 1 ) x n x n + b 1 b a n M 1 i = 1 n + 1 t i T n + 1 , n + 1 x n x n + b 1 b a n M 1 ( b n + 1 + b 1 b a n ) M 1 ,
(3.16)

where M 1 = inf i = 1 , 2 , ( 1 + 2 λ i q 1 λ i q 1 ) sup n 1 { x n p} with pF.

On the other hand, we note that

y n + 1 y n = ( I α n + 1 G ) W n + 1 x n + α n + 1 [ W n + 1 x n σ ( μ F W n + 1 x n γ V x n ) ] ( I α n G ) W n x n α n [ W n x n σ ( μ F W n x n γ V x n ) ] = ( I α n + 1 G ) ( W n + 1 x n W n x n ) + ( α n α n + 1 ) G W n x n + ( α n + 1 α n ) W n + 1 x n + α n ( W n + 1 x n W n x n ) + σ ( α n α n + 1 ) ( μ F W n + 1 x n γ V x n ) σ α n [ μ F W n + 1 x n γ V x n ( μ F W n x n γ V x n ) ] = [ ( 1 + α n ) I α n + 1 G ] ( W n + 1 x n W n x n ) + ( α n + 1 α n ) [ W n + 1 x n G W n x n ] + σ ( α n α n + 1 ) [ μ F W n + 1 x n γ V x n ] σ α n μ F [ W n + 1 x n W n x n ] = [ ( 1 + α n ) I α n + 1 G σ α n μ F ] ( W n + 1 x n W n x n ) + ( α n + 1 α n ) [ W n + 1 x n G W n x n ] + σ ( α n α n + 1 ) [ μ F W n + 1 x n γ V x n ] .

Hence, we have

S n + 1 α n + 1 x n S n α n x n = Q C y n + 1 Q C y n y n + 1 y n ( 1 + α n ) I α n + 1 G σ α n μ F W n + 1 x n W n x n + | α n + 1 α n | W n + 1 x n G W n x n + σ | α n + 1 α n | μ F W n + 1 x n γ V x n ( W n + 1 x n W n x n + | α n + 1 α n | ) M 2 ,

where M 2 = sup n 1 {(1+ α n )I α n + 1 Gσ α n μF, W n + 1 x n G W n x n ,σμF W n + 1 x n γV x n }. It follows from (3.1) and (3.16) that

x n + 2 x n + 1 S n + 1 α n + 1 x n + 1 S n + 1 α n + 1 x n + S n + 1 α n + 1 x n S n α n x n ( 1 α n + 1 θ ) x n + 1 x n + ( | α n + 1 α n | + W n + 1 x n W n x n ) M 2 ( 1 α n + 1 θ ) x n + 1 x n + ( ( α n ) + σ n ) M 2 + W n + 1 x n W n x n M 2 ( 1 α n + 1 θ ) x n + 1 x n + ( α n ) M 2 + ( σ n + b n + 1 + b 1 b a n ) M ,
(3.17)

where M=max{ M 1 , M 2 }. Then, by Lemma 2.13, we have

lim n x n + 1 x n =0.
(3.18)

Next, we show that lim n x n W x n =0. Since

x n W n x n x n x n + 1 + x n + 1 W n x n = x n x n + 1 + Q C [ ( I α n G ) W n x n + α n ( W n x n σ ( μ F W n x n γ V x n ) ) ] Q C W n x n x n x n + 1 + α n ( I G ) W n x n σ ( μ F W n x n γ V x n ) .

From (3.18) and the condition (C1), we obtain

lim n x n W n x n =0.
(3.19)

At the same time, observe that

x n W x n x n W n x n + W n x n W x n .

It follows from (3.19) and Lemma 2.11, we have

lim n x n W x n =0.
(3.20)

Next, we show that

lim sup n [ I G + σ ( γ V μ F ) ] x , j q ( x n x ) 0,

where x is the same as in Theorem 3.4. Since { x n } is bounded, there exists a subsequence { x n i } of { x n } such that

lim sup n [ I G + σ ( γ V μ F ) ] x , j q ( x n x ) = lim i [ I G + σ ( γ V μ F ) ] x , j q ( x n i x ) .

By reflexivity of a Banach space X and boundedness of { x n }, without loss of generality, we may assume that x n i v as i. It follows from (3.20) and Lemma 2.12 that vF. Since a Banach space X has a weakly sequentially continuous generalized duality mapping, we obtain

lim sup n [ I G + σ ( γ V μ F ) ] x , j q ( x n x ) = lim i [ I G + σ ( γ V μ F ) ] x , j q ( x n i x ) = [ I G + σ ( γ V μ F ) ] x j q ( v x ) 0 .
(3.21)

Finally, we show that x n x as n. Set x n + 1 = Q C y n , where y n =(I α n G) W n x n + α n ( W n x n σ(μ W n x n γV x n )). From Lemmas 2.4 and 2.5, we have

x n + 1 x q = y n x , j q ( x n + 1 x ) + Q C y n y n , j q ( x n + 1 x ) y n x , j q ( x n + 1 x ) = ( I α n G ) ( W n x n x ) , j q ( x n + 1 x ) + α n ( I σ μ F ) ( W n x n x ) , j q ( x n + 1 x ) + α n σ γ V x n V x , j q ( x n + 1 x ) + α n ( I σ μ F ) x + σ γ V x G x , j q ( x n + 1 x ) ( 1 α n ( 1 L δ , λ ) ) x n x x n + 1 x q 1 + α n ( 1 σ τ ) x n x x n + 1 x q 1 + α n σ γ L x n x x n + 1 x q 1 + α n x G x + σ ( γ V x μ F x ) , j q ( x n + 1 x ) = ( 1 α n ( σ ( τ γ L ) L δ , λ ) ) x n x x n + 1 x q 1 + α n x G x + σ ( γ V x μ F x ) , j q ( x n + 1 x ) ( 1 α n ( σ ( τ γ L ) L δ , λ ) ) [ 1 q x n x q + ( q 1 q ) x n + 1 x q ] + α n x G x + σ ( γ V x μ F x ) , j q ( x n + 1 x ) ,

which implies that

x n + 1 x q ( 1 α n ( σ ( τ γ L ) L δ , λ ) ) x n x q + q α n 1 + ( q 1 ) ( σ ( τ γ L ) L δ , λ ) × x G x + σ ( γ V x μ F x ) , j q ( x n + 1 x ) .
(3.22)

We can write (3.22) to the formula

x n + 1 x q (1 τ n ) x n x q + ξ n ,
(3.23)

where τ n :=(σ(τγL) L δ , λ ) α n and ξ n := q α n 1 + ( q 1 ) ( σ ( τ γ L ) L δ , λ ) x G x +σ(γV x μF x ), j q ( x n + 1 x ). Put c n =max{0, ξ n }, from (3.21), we have c n 0 as n. Then we can rewrite (3.23) as

x n + 1 x q ( 1 τ n ) x n x q + c n ( 1 τ n ) x n x q + ( α n ) .

Therefore, by Lemma 2.13, we conclude that x n x as n. This completes the proof. □

4 Some applications

In this section, we will utilize Theorems 3.4 and 3.5 to study some strong convergence theorems in L p (or p ) spaces with 1<p<. It well known that Hilbert spaces, L p (or p ) spaces with 1<p< and the Sobolev spaces W m p with 1<p< are q-uniformly smooth, i.e.,

L p (or  p ) or  W m p  is { 2 -uniformly smooth , if  2 p < , p -uniformly smooth , if  1 < p 2 .

Furthermore, we have the following properties of L p (or p ) spaces with 1<p< (see [36, 39]):

  1. (1)

    For 2p<, the spaces L p (or p ) are 2-uniformly smooth with C q = C 2 =p1.

  2. (2)

    For 1<p2, the spaces L p (or p ) are p-uniformly smooth with C q = C p =(1+ t p p 1 ) ( 1 + t p ) 1 p , where t p is the unique solution of the equation

    (p2) t p 1 +(p1) t p 2 1=0,0<t<1.
  3. (3)

    Every Hilbert spaces are 2-uniformly smooth with C q = C 2 =1.

  4. (4)

    Every L p (or p ) spaces with 1<p< are q-uniformly smooth and uniformly convex.

  5. (5)

    Every p spaces with 1<p< have weakly sequentially continuous generalized duality mappings, but L p spaces (1<p<, p2) do not have weakly sequentially continuous generalized duality mappings.

Lemma 4.1 Let X:= L p (or p ) with 1<p2. Let C be a nonempty, closed, and convex subset of X. Let F:CX be a κ-Lipschitzian and η-strongly accretive operator with constants κ,η>0. Let 0<μ< ( p η D p κ p ) 1 p 1 and τ=μ(η D p μ p 1 κ p p ). Then for t(0,min{1, 1 p τ }), the mapping S:CX defined by S:=ItμF is a contraction with constant 1tτ.

Lemma 4.2 Let X:= L p (or p ) with 2p<. Let C be a nonempty, closed, and convex subset of X. Let F:CX be a κ-Lipschitzian and η-strongly accretive operator with constants κ,η>0. Let 0<μ< 2 η ( p 1 ) κ 2 and τ=μ(η ( p 1 ) μ κ 2 2 ). Then for t(0,min{1, 1 2 τ }), the mapping S:CX defined by S:=ItμF is a contraction with constant 1tτ.

Lemma 4.3 Let X:=H be a real Hilbert space. Let C be a nonempty, closed, and convex subset of X. Let F:CX be a κ-Lipschitzian and η-strongly accretive operator with constants κ,η>0. Let 0<μ< 2 η κ 2 and τ=μ(η μ κ 2 2 ). Then for t(0,min{1, 1 2 τ }), the mapping S:CX defined by S:=ItμF is a contraction with constant 1tτ.

4.1 Implicit iteration schemes

Theorem 4.4 Let C be a nonempty, closed, and convex subset of an p space for 1<p2. Let Q C , F, G, V, and T be the same as in Theorem  3.4. Assume that 0<μ< ( p η D p κ p ) 1 p 1 and 0γL<τ, where τ=μ(η D p μ p 1 κ p p ). For σ( L δ , λ τ γ L ,min{1, 1 p τ , 1 + L δ , λ τ γ L }) and t(0,1), the sequence { x t } defined by (3.2) converges strongly to x Fix(T) as t0, where x is the unique solution of the variational inequality (3.3).

Theorem 4.5 Let C be a nonempty, closed, and convex subset of an p space for 2p<. Let Q C , F, G, V, and T be the same as in Theorem  3.4. Assume that 0<μ< 2 η ( p 1 ) κ 2 and 0γL<τ, where τ=μ(η ( p 1 ) μ κ 2 2 ). For σ( L δ , λ τ γ L ,min{1, 1 2 τ , 1 + L δ , λ τ γ L }) and t(0,1), the sequence { x t } defined by (3.2) converges strongly to x Fix(T) as t0, where x is the unique solution of the variational inequality (3.3).

Remark 4.6 If the spaces L p has a weakly sequentially continuous generalized duality mappings, then we obtain Theorems 4.4 and 4.5 hold for L p spaces with 1<p<, p2.

4.2 Explicit iteration schemes

Theorem 4.7 Let C be a nonempty, closed, and convex subset of an p space for 1<p2. Let Q C , F, G, V, and W n be the same as in Theorem  3.5. Let { α n } and { β n } are sequences in (0,1) which satisfy the conditions (C1) and (C2) in Theorem  3.5 and { θ n , k } satisfies (H1)-(H3). Then the sequence { x n } defined by (3.14) converges strongly to x F as n, where x is the unique solution of the variational inequality (3.15).

Theorem 4.8 Let C be a nonempty, closed, and convex subset of an p space for 2p<. Let Q C , F, G, V, and W n be the same as in Theorem  3.5. Let { α n } and { β n } are sequences in (0,1) which satisfy the conditions (C1) and (C2) in Theorem  3.5 and { θ n , k } satisfies (H1)-(H3). Then the sequence { x n } defined by (3.14) converges strongly to x F as n, where x is the unique solution of the variational inequality (3.15).

Remark 4.9 If the spaces L p has a weakly sequentially continuous generalized duality mappings, then we obtain Theorems 4.7 and 4.8 hold for L p spaces with 1<p<, p2.

5 Numerical examples

In this section, we give a simple example and some numerical experiment result to explain the convergence of the sequence (3.14) as follows:

Example 5.1 Let X=R and C=[0, 1 2 ]. Let q=2 and j q =I. We define a mapping Q C as follows:

Q C x={ x | x | , x ( , 0 ) ( 1 2 , ) , x , x [ 0 , 1 2 ] .

In terms of Theorem 3.5, set σ=μ=γ=1 and α n = 1 n . Then we see that α n = 1 n satisfies (C1) and (C2) with σ n = 1 n 2 . Moreover, we define the mappings F, G, and V as follows:

Fx= 1 3 ( x 2 + 2 x ) ,Gx=xandVx= x 2 .

It is easy to observe that F is 1-Lipschitzian and 2 3 -strongly accretive, G is 1-strongly accretive and λ-strictly pseudo-contraction for λ>0 and V is 1-Lipschitzian. For each nN, set S n =I. We show that W n =I. Since T n , k = θ n , k S k +(1 θ n , k )I, where S k is a λ k -strictly pseudo-contractive mapping and { θ n , k } satisfies (H1)-(H3). It is observe that T n , k is a nonexpansive mapping. From (2.5), we have

W 1 = U 1 , 1 = t 1 T 1 , 1 U 1 , 2 + ( 1 t 1 ) I , W 2 = U 2 , 1 = t 1 T 2 , 1 U 2 , 2 + ( 1 t 1 ) I W 2 = t 1 T 2 , 1 ( t 2 T 2 , 2 U 2 , 3 + ( 1 t 2 ) I ) + ( 1 t 1 ) I W 2 = t 1 t 2 T 2 , 1 T 2 , 2 U 2 , 3 + t 1 ( 1 t 2 ) T 2 , 1 + ( 1 t 1 ) I , W 3 = U 3 , 1 = t 1 T 3 , 1 U 3 , 2 + ( 1 t 1 ) I W 3 = t 1 T 3 , 1 ( t 2 T 3 , 2 U 3 , 3 + ( 1 t 2 ) I ) + ( 1 t 1 ) I W 3 = t 1 t 2 T 3 , 1 T 3 , 2 U 3 , 3 + t 1 ( 1 t 2 ) T 3 , 1 + ( 1 t 1 ) I W 3 = t 1 t 2 T 3 , 1 T 3 , 2 ( t 3 T 3 , 3 U 3 , 4 + ( 1 t 3 ) I ) + t 1 ( 1 t 2 ) T 3 , 1 + ( 1 t 1 ) I W 3 = t 1 t 2 t 3 T 3 , 1 T 3 , 2 T 3 , 3 + t 1 t 2 ( 1 t 3 ) T 3 , 1 T 3 , 2 + t 1 ( 1 t 2 ) T 3 , 1 + ( 1 t 1 ) I

and we compute (2.5) in a similar way to above, we obtain

W n = U n , 1 = t 1 t 2 t n T n , 1 T n , 2 T n , n + t 1 t 2 t n 1 ( 1 t n ) T n , 1 T n , 2 T n , n 1 + t 1 t 2 t n 2 ( 1 t n 1 ) T n , 1 T n , 2 T n , n 2 + + t 1 ( 1 t 2 ) T n , 1 + ( 1 t 1 ) I .

Since S n =I and t n =α, for all nN, we have

W n = [ α n + α n 1 ( 1 α ) + + α ( 1 α ) + ( 1 α ) ] =I.

Under the above assumptions, (3.14) is simplified as follows:

{ x 1 C : = [ 0 , 1 2 ] , x n + 1 = ( 1 2 3 n ) x n + 2 3 n x n 2 .
(5.1)

Since the assumptions of Theorem 3.5 are satisfied in Example 5.1, the sequence (5.1) converges to x =0, which is the unique fixed point of S n .

Next, we show the numerical results by using MATLAB 7.11.0. We presented numerical comparisons for two cases of iteration process with different initial values, which show the convergence of the sequence (5.1).

When we choose x 1 =0.05 and x 1 =0.1, we see that the iteration process of sequence { x n } converges to x =0 at n=8,615 and n=28,946, respectively, as shown in Table 1 and Figures 1 and 2.

Figure 1
figure1

The iteration process with initial value x 1 =0.05 .

Figure 2
figure2

The iteration process with initial value x 1 =0.1 .

Table 1 The value of sequence { x n } with iteration values x 1 =0.05 and x 1 =0.1

From the figures, we can see that { x n } is a monotone decreasing sequence and converges to 0, but an iterative process with initial value x 1 =0.05 is converges faster than an iterative process with initial value x 1 =0.1.

Remark 5.2 Note that Lemma 3.1 and Lemma 3.2 play an important role in the proof of Theorems 3.4 and 3.5. These are proved in the framework of the more general q-uniformly smooth Banach space.

Remark 5.3 Our main result extends the main result of Ceng et al. [28] in the following respects:

  1. (1)

    An iterative process (1.10) is to extend to a general iterative process defined over the set of fixed points of an infinite family of strict pseudo-contractions in a more general q-uniformly smooth Banach space.

  2. (2)

    The self contraction mapping f:HH in [[28], Theorem 3.2] is extended to the case of a nonself Lipschitzian mapping V:CX on a nonempty, closed, and convex subset C of a real q-uniformly smooth Banach space X.

  3. (3)

    The control condition (C3) in [[28], Theorem 3.2] is removed by weaker than control condition | α n + 1 α n |( α n )+ σ n with n = 1 σ n <.

Furthermore, our method is extended to develop a new iterative method and method of proof is very different from that in Ceng et al. [28] because our method involves the sunny nonexpansive retraction and the infinite family of strict pseudo-contractions.

References

  1. 1.

    Youla DC, Webb H: Image restoration by the method of convex projections: part 1 - theory. IEEE Trans. Med. Imaging 1982, 1: 81–94.

  2. 2.

    Kitahara S, Takahashi W: Image recovery by convex combinations of sunny nonexpansive retractions. Topol. Methods Nonlinear Anal. 1993,2(2):333–342.

  3. 3.

    Takahashi W, Tamura T: Limit theorems of operators by convex combinations of nonexpansive retractions in Banach spaces. J. Approx. Theory 1997,91(3):386–397. 10.1006/jath.1996.3093

  4. 4.

    Kimura Y, Nakajo K: The problem of image recovery by the metric projections in Banach spaces. Abstr. Appl. Anal. 2013., 2013: Article ID 817392

  5. 5.

    Alber YI: Metric and generalized projections operators in Banach spaces: properties and applications. Lecture Notes in Pure and Applied Mathematics 178. In Theory and Applications of Nonlinear Operators of Accretive and Monotone Type. Edited by: Kartsatos AG. Dekker, New York; 1996:15–50.

  6. 6.

    Alber YI, Reich S: An iterative method for solving a class of nonlinear operator equations in Banach spaces. Panam. Math. J. 1994,4(2):39–54.

  7. 7.

    Kamimura S, Takahashi W: Strong convergence of a proximal-type algorithm in a Banach space. SIAM J. Optim. 2002,13(3):938–945. 10.1137/S105262340139611X

  8. 8.

    Nakajo K, Shimoji K, Takahashi W: Strong convergence theorems by the hybrid method for families of mappings in Banach spaces. Nonlinear Anal., Theory Methods Appl. 2009,71(3–4):812–818. 10.1016/j.na.2008.10.108

  9. 9.

    Kimura Y, Nakajo K, Takahashi W: Strongly convergent iterative schemes for a sequence of nonlinear mappings. J. Nonlinear Convex Anal. 2008,9(3):407–416.

  10. 10.

    Kohsaka F, Takahashi W: Iterative scheme for finding a common point of infinitely many convex sets in a Banach space. J. Nonlinear Convex Anal. 2004,5(3):407–414.

  11. 11.

    Matsushita S, Takahashi W: A strong convergence theorem for relatively nonexpansive mappings in a Banach space. J. Approx. Theory 2005,134(2):257–266. 10.1016/j.jat.2005.02.007

  12. 12.

    Matsushita S, Nakajo K, Takahashi W: Strong convergence theorems obtained by a generalized projections hybrid method for families of mappings in Banach spaces. Nonlinear Anal., Theory Methods Appl. 2010,73(6):1466–1480. 10.1016/j.na.2010.04.007

  13. 13.

    Sunthrayuth P, Kumam P: Iterative methods for variational inequality problems and fixed point problems of a countable family of strict pseudo-contractions in a q -uniformly smooth Banach space. Fixed Point Theory Appl. 2012., 2012: Article ID 65

  14. 14.

    Song Y, Ceng L-C: A general iteration scheme for variational inequality problem and common fixed point problems of nonexpansive mappings in q -uniformly smooth Banach spaces. J. Glob. Optim. 2013. 10.1007/s10898-012-9990-4

  15. 15.

    Cholamjiak P, Suantai S: Strong convergence for a countable family of strict pseudocontractions in q -uniformly smooth Banach spaces. Comput. Math. Appl. 2011, 62: 787–796. 10.1016/j.camwa.2011.06.008

  16. 16.

    Zhang H, Su Y: Strong convergence theorems for strict pseudo-contractions in q -uniformly smooth Banach spaces. Nonlinear Anal. 2009, 70: 3236–3242. 10.1016/j.na.2008.04.030

  17. 17.

    Cai G, Hu CS: Strong convergence theorems of a general iterative process for a finite family of λ i -strict pseudo-contractions in q -uniformly smooth Banach spaces. Comput. Math. Appl. 2010, 59: 149–160. 10.1016/j.camwa.2009.07.068

  18. 18.

    Song Y, Hu H, Ceng LC: On variational inclusion and common fixed point problems in q -uniformly smooth Banach spaces. J. Appl. Math. 2012., 2012: Article ID 865810

  19. 19.

    Cai G, Bu S: An iterative algorithm for a general system of variational inequalities and fixed point problems in q -uniformly smooth Banach spaces. Optim. Lett. 2013,7(2):267–287. 10.1007/s11590-011-0415-y

  20. 20.

    Zhou HY: Convergence theorems for λ -strict pseudo-contractions in q -uniformly smooth Banach spaces. Acta Math. Sin. Engl. Ser. 2010,26(4):743–758. 10.1007/s10114-010-7341-2

  21. 21.

    Zhang H, Su Y: Convergence theorems for strict pseudo-contractions in q -uniformly smooth Banach spaces. Nonlinear Anal. 2009, 71: 4572–4580. 10.1016/j.na.2009.03.033

  22. 22.

    Saewan S, Kumam P: A strong convergence theorem concerning a hybrid projection method for finding common fixed points of a countable family of relatively quasi-nonexpansive mappings. J. Nonlinear Convex Anal. 2012,13(2):313–330.

  23. 23.

    Saewan S, Kumam P: Computational of generalized projection method for maximal monotone operators and a countable family of relatively quasi-nonexpansive mappings. Optimization 2013. 10.1080/02331934.2013.824444

  24. 24.

    Saewan S, Kumam P, Kanjanasamranwong P: The hybrid projection algorithm for finding the common fixed points of nonexpansive mappings and the zeroes of maximal monotone operators in Banach spaces. Optimization 2013. 10.1080/02331934.2012.724686

  25. 25.

    Marino G, Xu HK: A general iterative method for nonexpansive mappings in Hilbert spaces. J. Math. Anal. Appl. 2006, 318: 43–52. 10.1016/j.jmaa.2005.05.028

  26. 26.

    Yamada I: The hybrid steepest descent method for the variational inequality problems over the intersection of fixed point sets of nonexpansive mappings. Stud. Comput. Math. 8. In Inherently Parallel Algorithms in Feasibility and Optimization and Their Applications. Edited by: Butnariu D, Censor Y, Reich S. North-Holland, Amsterdam; 2001:473–504.

  27. 27.

    Tian M: A general iterative algorithm for nonexpansive mappings in Hilbert spaces. Nonlinear Anal., Theory Methods Appl. 2010, 73: 689–694. 10.1016/j.na.2010.03.058

  28. 28.

    Ceng L-C, Guu S-M, Yao J-C: A general composite iterative algorithm for nonexpansive mappings in Hilbert spaces. Comput. Math. Appl. 2011, 61: 2447–2455. 10.1016/j.camwa.2011.02.025

  29. 29.

    Takahashi W: Nonlinear Functional Analysis. Yokohama Publishers, Yokohama; 2000.

  30. 30.

    Xu ZB, Roach GF: Characteristic inequalities of uniformly smooth Banach spaces. J. Math. Anal. Appl. 1991, 157: 189–210. 10.1016/0022-247X(91)90144-O

  31. 31.

    Takahashi Y, Hashimoto K, Kato M: On sharp uniform convexity, smoothness, and strong type, cotype inequalities. J. Nonlinear Convex Anal. 2002, 3: 267–281.

  32. 32.

    Browder FE, Petryshyn WV: Construction of fixed points of nonlinear mappings in Hilbert space. J. Math. Anal. Appl. 1967, 20: 197–228. 10.1016/0022-247X(67)90085-6

  33. 33.

    Chen R, Song Y, Zhou H: Convergence theorems for implicit iteration process for a finite family of continuous pseudocontractive mappings. J. Math. Anal. Appl. 2006, 314: 701–709. 10.1016/j.jmaa.2005.04.018

  34. 34.

    Reich S: Asymptotic behavior of contractions in Banach spaces. J. Math. Anal. Appl. 1973, 44: 57–70. 10.1016/0022-247X(73)90024-3

  35. 35.

    Mitrinović DS: Analytic Inequalities. Springer, New York; 1970.

  36. 36.

    Xu HK: Inequalities in Banach spaces with applications. Nonlinear Anal., Theory Methods Appl. 1991, 16: 1127–1138. 10.1016/0362-546X(91)90200-K

  37. 37.

    Dong Q-L, He S, Su F: Strong convergence of an iterative algorithm for an infinite family of strict pseudo-contractions in Banach spaces. Appl. Math. Comput. 2010, 216: 959–969. 10.1016/j.amc.2010.01.111

  38. 38.

    Opial Z: Weak convergence of the sequence of successive approximations for nonexpansive mappings. Bull. Am. Math. Soc. 1967, 73: 591–597. 10.1090/S0002-9904-1967-11761-0

  39. 39.

    Agarwal RP, O’Regan D, Sahu DR: Fixed Point Theory for Lipschitzian-Type Mappings with Applications. Springer, Berlin; 2009.

  40. 40.

    Liu LS: Iterative processes with errors for nonlinear strongly accretive mappings in Banach spaces. J. Math. Anal. Appl. 1980, 75: 114–292.

Download references

Acknowledgements

The second author was supported by the Thailand Research Fund and the King Mongkut’s University of Technology Thonburi (Grant No. RSA5780059).

Author information

Correspondence to Poom Kumam.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

All authors contributed equally and significantly in writing this paper. All authors read and approved the final manuscript.

Authors’ original submitted files for images

Below are the links to the authors’ original submitted files for images.

Authors’ original file for figure 1

Authors’ original file for figure 2

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0), which permits use, duplication, 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 license, and indicate if changes were made.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Keywords

  • variational inequality
  • Banach space
  • strong convergence
  • iterative method
  • common fixed point
  • strongly accretive operator
  • inverse strongly accretive operator