Loading [MathJax]/jax/output/SVG/jax.js
Research article

STD-YOLOv8: A lightweight small target detection algorithm for UAV perspectives

  • When recognizing targets by unmanned aerial vehicles (UAVs), problems such as small size, dense dispersion, and complex background are likely to occur, resulting in low recognition rates. In order to solve the above problems, this work proposed a lightweight small target detection algorithm based on the YOLOv8n: STD-YOLOv8 algorithm. First, the regression problem of small targets in different training periods was optimized, the penalty term in the original loss was improved, and a new LIoU loss function was proposed, so that the size of the penalty term could be dynamically adjusted before and after training, thus improving the performance of the algorithm. Second, in order to better adapt to the small target scale and enhance the ability of small target feature acquisition, the SPD-Conv module was integrated in the backbone network, replacing the original stepwise convolutional layer and pooling layer, so as to solve the problems of loss of fine-grained information and low efficiency of feature representation existing in the current convolutional neural network (CNN) structure. In the neck part, nearest-neighbor upsampling was replaced by the feature reassembly assembly of features operator CARAFE (content-aware reassembly of features), which enabled the model to aggregate contextual information in a larger perceptual field and enhanced the feature representation in the neck. Finally, validation experiments were conducted by comparing different algorithms under the same VisDrone-2021 dataset. The results of the ablation experiments show that the algorithms proposed in this thesis have improved the recall (R), mAP50, and mAP95 by 4.7, 5.8 and 5.7%, respectively, compared with YOLOv8n. The results of the model generalization experiments on the TinyPerson dataset show that the algorithm in this paper has superior small target detection performance with only 1.2 M model parameters (1 M = 106).

    Citation: Dong Wu, Jiechang Li, Weijiang Yang. STD-YOLOv8: A lightweight small target detection algorithm for UAV perspectives[J]. Electronic Research Archive, 2024, 32(7): 4563-4580. doi: 10.3934/era.2024207

    Related Papers:

    [1] Sun Shin Ahn, Ravikumar Bandaru, Young Bae Jun . Imploring interior GE-filters in GE-algebras. AIMS Mathematics, 2022, 7(1): 855-868. doi: 10.3934/math.2022051
    [2] Young Bae Jun, Ravikumar Bandaru, Amal S. Alali . Endomorphic GE-derivations. AIMS Mathematics, 2025, 10(1): 1792-1813. doi: 10.3934/math.2025082
    [3] Yanjiao Liu, Jianhua Yin . The generalized Turán number of 2S. AIMS Mathematics, 2023, 8(10): 23707-23712. doi: 10.3934/math.20231205
    [4] Chang-Xu Zhang, Fu-Tao Hu, Shu-Cheng Yang . On the (total) Roman domination in Latin square graphs. AIMS Mathematics, 2024, 9(1): 594-606. doi: 10.3934/math.2024031
    [5] Chin-Lin Shiue, Tzu-Hsien Kwong . Distance 2-restricted optimal pebbling in cycles. AIMS Mathematics, 2025, 10(2): 4355-4373. doi: 10.3934/math.2025201
    [6] Feifei Qu, Xin Wei, Juan Chen . Uncertainty principle for vector-valued functions. AIMS Mathematics, 2024, 9(5): 12494-12510. doi: 10.3934/math.2024611
    [7] Fei Hou, Bin Chen . On triple correlation sums of Fourier coefficients of cusp forms. AIMS Mathematics, 2022, 7(10): 19359-19371. doi: 10.3934/math.20221063
    [8] Haocong Song, Wen Wu . Hankel determinants of a Sturmian sequence. AIMS Mathematics, 2022, 7(3): 4233-4265. doi: 10.3934/math.2022235
    [9] Fei Hou, Bin Chen . Triple correlation sums of coefficients of θ-series. AIMS Mathematics, 2023, 8(10): 25275-25287. doi: 10.3934/math.20231289
    [10] Tingting Li, Junyang Gao . The Hausdorff dimension of the Julia sets concerning generated renormalization transformation. AIMS Mathematics, 2022, 7(1): 939-956. doi: 10.3934/math.2022056
  • When recognizing targets by unmanned aerial vehicles (UAVs), problems such as small size, dense dispersion, and complex background are likely to occur, resulting in low recognition rates. In order to solve the above problems, this work proposed a lightweight small target detection algorithm based on the YOLOv8n: STD-YOLOv8 algorithm. First, the regression problem of small targets in different training periods was optimized, the penalty term in the original loss was improved, and a new LIoU loss function was proposed, so that the size of the penalty term could be dynamically adjusted before and after training, thus improving the performance of the algorithm. Second, in order to better adapt to the small target scale and enhance the ability of small target feature acquisition, the SPD-Conv module was integrated in the backbone network, replacing the original stepwise convolutional layer and pooling layer, so as to solve the problems of loss of fine-grained information and low efficiency of feature representation existing in the current convolutional neural network (CNN) structure. In the neck part, nearest-neighbor upsampling was replaced by the feature reassembly assembly of features operator CARAFE (content-aware reassembly of features), which enabled the model to aggregate contextual information in a larger perceptual field and enhanced the feature representation in the neck. Finally, validation experiments were conducted by comparing different algorithms under the same VisDrone-2021 dataset. The results of the ablation experiments show that the algorithms proposed in this thesis have improved the recall (R), mAP50, and mAP95 by 4.7, 5.8 and 5.7%, respectively, compared with YOLOv8n. The results of the model generalization experiments on the TinyPerson dataset show that the algorithm in this paper has superior small target detection performance with only 1.2 M model parameters (1 M = 106).



    Henkin and Skolem introduced Hilbert algebras in the fifties for investigations in intuitionistic and other non-classical logics. Diego [4] proved that Hilbert algebras form a variety which is locally finite. Bandaru et al. introduced the notion of GE-algebras which is a generalization of Hilbert algebras, and investigated several properties (see [1,2,7,8,9]). The notion of interior operator is introduced by Vorster [12] in an arbitrary category, and it is used in [3] to study the notions of connectedness and disconnectedness in topology. Interior algebras are a certain type of algebraic structure that encodes the idea of the topological interior of a set, and are a generalization of topological spaces defined by means of topological interior operators. Rachůnek and Svoboda [6] studied interior operators on bounded residuated lattices, and Svrcek [11] studied multiplicative interior operators on GMV-algebras. Lee et al. [5] applied the interior operator theory to GE-algebras, and they introduced the concepts of (commutative, transitive, left exchangeable, belligerent, antisymmetric) interior GE-algebras and bordered interior GE-algebras, and investigated their relations and properties. Later, Song et al. [10] introduced the notions of an interior GE-filter, a weak interior GE-filter and a belligerent interior GE-filter, and investigate their relations and properties. They provided relations between a belligerent interior GE-filter and an interior GE-filter and conditions for an interior GE-filter to be a belligerent interior GE-filter is considered. Given a subset and an element, they established an interior GE-filter, and they considered conditions for a subset to be a belligerent interior GE-filter. They studied the extensibility of the belligerent interior GE-filter and established relationships between weak interior GE-filter and belligerent interior GE-filter of type 1, type 2 and type 3. Rezaei et al. [7] studied prominent GE-filters in GE-algebras. The purpose of this paper is to study by applying interior operator theory to prominent GE-filters in GE-algebras. We introduce the concept of a prominent interior GE-filter, and investigate their properties. We discuss the relationship between a prominent GE-filter and a prominent interior GE-filter and the relationship between an interior GE-filter and a prominent interior GE-filter. We find and provide examples where any interior GE-filter is not a prominent interior GE-filter and any prominent GE-filter is not a prominent interior GE-filter. We provide conditions for an interior GE-filter to be a prominent interior GE-filter. We provide conditions under which an internal GE-filter larger than a given internal GE filter can become a prominent internal GE-filter, and give an example describing it. We also introduce the concept of a prominent interior GE-filter of type 1 and type 2, and investigate their properties. We discuss the relationship between a prominent interior GE-filter and a prominent interior GE-filter of type 1. We give examples to show that A and B are independent of each other, where A and B are:

    (1) { A: prominent interior GE-filter of type 1. B: prominent interior GE-filter of type 2.

    (2) { A: prominent interior GE-filter. B: prominent interior GE-filter of type 2.

    (3) { A: interior GE-filter. B: prominent interior GE-filter of type 1.

    (4) { A: interior GE-filter. B: prominent interior GE-filter of type 2.

    Definition 2.1. [1] By a GE-algebra we mean a non-empty set X with a constant 1 and a binary operation satisfying the following axioms:

    (GE1) uu=1,

    (GE2) 1u=u,

    (GE3) u(vw)=u(v(uw))

    for all u,v,wX.

    In a GE-algebra X, a binary relation "" is defined by

    (x,yX)(xyxy=1). (2.1)

    Definition 2.2. [1,2,8] A GE-algebra X is said to be transitive if it satisfies:

    (x,y,zX)(xy(zx)(zy)). (2.2)

    Proposition 2.3. [1] Every GE-algebra X satisfies the following items:

    (uX)(u1=1). (2.3)
    (u,vX)(u(uv)=uv). (2.4)
    (u,vX)(uvu). (2.5)
    (u,v,wX)(u(vw)v(uw)). (2.6)
    (uX)(1uu=1). (2.7)
    (u,vX)(u(vu)u). (2.8)
    (u,vX)(u(uv)v). (2.9)
    (u,v,wX)(uvwvuw). (2.10)

    If X is transitive, then

    (u,v,wX)(uvwuwv,vwuw). (2.11)
    (u,v,wX)(uv(vw)(uw)). (2.12)

    Lemma 2.4. [1] In a GE-algebra X, the following facts are equivalent each other.

    (x,y,zX)(xy(zx)(zy)). (2.13)
    (x,y,zX)(xy(yz)(xz)). (2.14)

    Definition 2.5. [1] A subset F of a GE-algebra X is called a GE-filter of X if it satisfies:

    1F, (2.15)
    (x,yX)(xyF,xFyF). (2.16)

    Lemma 2.6. [1] In a GE-algebra X, every filter F of X satisfies:

    (x,yX)(xy,xFyF). (2.17)

    Definition 2.7. [7] A subset F of a GE-algebra X is called a prominent GE-filter of X if it satisfies (2.15) and

    (x,y,zX)(x(yz)F,xF((zy)y)zF). (2.18)

    Note that every prominent GE-filter is a GE-filter in a GE-algebra (see [7]).

    Definition 2.8. [5] By an interior GE-algebra we mean a pair (X,f) in which X is a GE-algebra and f:XX is a mapping such that

    (xX)(xf(x)), (2.19)
    (xX)((ff)(x)=f(x)), (2.20)
    (x,yX)(xyf(x)f(y)). (2.21)

    Definition 2.9. [10] Let (X,f) be an interior GE-algebra. A GE-filter F of X is said to be interior if it satisfies:

    (xX)(f(x)FxF). (2.22)

    Definition 3.1. Let (X,f) be an interior GE-algebra. Then a subset F of X is called a prominent interior GE-filter in (X,f) if F is a prominent GE-filter of X which satisfies the condition (2.22).

    Example 3.2. Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 1.

    Table 1.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 4 4
    3 1 1 1 5 5
    4 1 2 3 1 1
    5 1 2 2 1 1

     | Show Table
    DownLoad: CSV

    Then X is a GE-algebra. If we define a mapping f on X as follows:

    f:XX,x{1if x{1,4,5},2if x{2,3},

    then (X,f) is an interior GE-algebra and F={1,4,5} is a prominent interior GE-filter in (X,f).

    It is clear that every prominent interior GE-filter is a prominent GE-filter. But any prominent GE-filter may not be a prominent interior GE-filter in an interior GE-algebra as seen in the following example.

    Example 3.3. Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 2,

    Table 2.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 3 4 1
    3 1 2 1 4 5
    4 1 2 3 1 5
    5 1 1 3 4 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x{1,2,3,5},4if x=4.

    Then (X,f) is an interior GE-algebra and F:={1} is a prominent GE-filter of X. But it is not a prominent interior GE-filter in (X,f) since f(2)=1F but 2F.

    We discuss relationship between interior GE-filter and prominent interior GE-filter.

    Theorem 3.4. In an interior GE-algebra, every prominent interior GE-filter is an interior GE-filter.

    Proof. It is straightforward because every prominent GE-filter is a GE-filter in a GE-algebra.

    In the next example, we can see that any interior GE-filter is not a prominent interior GE-filter in general.

    Example 3.5. Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 3.

    Table 3.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 4 4
    3 1 2 1 4 4
    4 1 1 3 1 1
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    Then X is a GE-algebra. If we define a mapping f on X as follows:

    f:XX,x{1if x=1,2if x{2,4,5},3if x=3,

    then (X,f) is an interior GE-algebra and F={1} is an interior GE-filter in (X,f). But it is not a prominent interior GE-filter in (X,f) since 1(23)=1F but ((32)2)3=3F.

    Proposition 3.6. Every prominent interior GE-filter F in an interior GE-algebra (X,f) satisfies:

    (x,yX)(f(xy)F((yx)x)yF). (3.1)

    Proof. Let F be a prominent interior GE-filter in (X,f). Let x,yX be such that f(xy)F. Then xyF by (2.22), and so 1(xy)=xyF by (GE2). Since 1F, it follows from (2.18) that ((yx)x)yF.

    Corollary 3.7. Every prominent interior GE-filter F in an interior GE-algebra (X,f) satisfies:

    (x,yX)(xyF((yx)x)yF). (3.2)

    Proof. Let F be a prominent interior GE-filter in (X,f). Then F is an interior GE-filter in (X,f) by Theorem 3.4. Let x,yX be such that xyF. Since xyf(xy) by (2.19), it follows from Lemma 2.6 that f(xy)F. Hence ((yx)x)yF by Proposition 3.6.

    Corollary 3.8. Every prominent interior GE-filter F in an interior GE-algebra (X,f) satisfies:

    (x,yX)(xyFf(((yx)x)y)F).

    Proof. Straightforward.

    Corollary 3.9. Every prominent interior GE-filter F in an interior GE-algebra (X,f) satisfies:

    (x,yX)(f(xy)Ff(((yx)x)y)F).

    Proof. Straightforward.

    In the following example, we can see that any interior GE-filter F in an interior GE-algebra (X,f) does not satisfy the conditions (3.1) and (3.2).

    Example 3.10. Consider the interior GE-algebra (X,f) in Example 3.4. The interior GE-filter F:={1} does not satisfy conditions (3.1) and (3.2) since f(23)=f(1)=1F and 23=1F but ((32)2)3=3F.

    We provide conditions for an interior GE-filter to be a prominent interior GE-filter.

    Theorem 3.11. If an interior GE-filter F in an interior GE-algebra (X,f) satisfies the condition (3.1), then F is a prominent interior GE-filter in (X,f).

    Proof. Let F be an interior GE-filter in (X,f) that satisfies the condition (3.1). Let x,y,zX be such that x(yz)F and xF. Then yzF. Since yzf(yz) by (2.19), it follows from Lemma 2.6 that f(yz)F. Hence ((zy)y)zF by (3.1), and therefore F is a prominent interior GE-filter in (X,f).

    Lemma 3.12. [10] In an interior GE-algebra, the intersection of interior GE-filters is also an interior GE-filter.

    Theorem 3.13. In an interior GE-algebra, the intersection of prominent interior GE-filters is also a prominent interior GE-filter.

    Proof. Let {FiiΛ} be a set of prominent interior GE-filters in an interior GE-algebra (X,f) where Λ is an index set. Then {FiiΛ} is a set of interior GE-filters in (X,f), and so {FiiΛ} is an interior GE-filter in (X,f) by Lemma 3.12. Let x,yX be such that f(xy){FiiΛ}. Then f(xy)Fi for all iΛ. It follows from Proposition 3.6 that ((yx)x)yFi for all iΛ. Hence ((yx)x)y{FiiΛ} and therefore {FiiΛ} is a prominent interior GE-filter in (X,f) by Theorem 3.11.

    Theorem 3.14. If an interior GE-filter F in an interior GE-algebra (X,f) satisfies the condition (3.2), then F is a prominent interior GE-filter in (X,f).

    Proof. Let F be an interior GE-filter in (X,f) that satisfies the condition (3.2). Let x,y,zX be such that x(yz)F and xF. Then yzF and thus ((zy)y)zF. Therefore F is a prominent interior GE-filter in (X,f).

    Given an interior GE-filter F in an interior GE-algebra (X,f), we consider an interior GE-filter G which is greater than F in (X,f), that is, we take two interior GE-filters F and G such that FG in an interior GE-algebra (X,f). We are now trying to find the condition that G can be a prominent interior GE-filter in (X,f).

    Theorem 3.15. Let (X,f) be an interior GE-algebra in which X is transitive. Let F and G be interior GE-filters in (X,f). If FG and F is a prominent interior GE-filter in (X,f), then G is also a prominent interior GE-filter in (X,f).

    Proof. Assume that F is a prominent interior GE-filter in (X,f). Then it is an interior GE-filter in (X,f) by Theorem 3.4. Let x,yX be such that f(xy)G. Then xyG by (2.22), and so 1=(xy)(xy)x((xy)y) by (GE1) and (2.6). Since 1F, it follows from Lemma 2.6 that x((xy)y)F. Hence ((((xy)y)x)x)((xy)y)FG by Corollary 3.7. Since

    ((((xy)y)x)x)((xy)y)(xy)(((((xy)y)x)x)y)

    by (2.6), we have (xy)(((((xy)y)x)x)y)G by Lemma 2.6. Hence

    ((((xy)y)x)x)yG.

    Since y(xy)y, it follows from (2.11) that

    ((((xy)y)x)x)y((yx)x)y.

    Thus ((yx)x)yG by Lemma 2.6. Therefore G is a prominent interior GE-filter in (X,f). by Theorem 3.11.

    The following example describes Theorem 3.15.

    Example 3.16. Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 4,

    Table 4.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 5 5
    3 1 1 1 5 5
    4 1 3 3 1 1
    5 1 3 3 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,3if x{2,3},5if x{4,5}.

    Then (X,f) is an interior GE-algebra in which X is transitive, and F:={1} and G:={1,4,5} are interior GE-filters in (X,f) with FG. Also we can observe that F and G are prominent interior GE-filters in (X,f).

    In Theorem 3.15, if F is an interior GE-filter which is not prominent, then G is also not a prominent interior GE-filter in (X,f) as shown in the next example.

    Example 3.17. Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 5,

    Table 5.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 4 1
    3 1 5 1 4 5
    4 1 1 1 1 1
    5 1 1 1 4 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,3if x=3,4if x=4,2if x{2,5}.

    Then (X,f) is an interior GE-algebra in which X is transitive, and F:={1} and G:={1,3} are interior GE-filters in (X,f) with FG. We can observe that F and G are not prominent interior GE-filters in (X,f) since 23=1F but ((32)2)3=(52)3=13=3F, and 42=1G but ((24)4)2=(44)2=12=2G.

    In Theorem 3.15, if X is not transitive, then Theorem 3.15 is false as seen in the following example.

    Example 3.18. Let X={1,2,3,4,5,6} be a set with the Cayley table which is given in Table 6.

    Table 6.  Cayley table for the binary operation "".
    1 2 3 4 5 6
    1 1 2 3 4 5 6
    2 1 1 1 6 6 6
    3 1 1 1 5 5 5
    4 1 1 3 1 1 1
    5 1 2 3 2 1 1
    6 1 2 3 2 1 1

     | Show Table
    DownLoad: CSV

    If we define a mapping f on X as follows:

    f:XX,x{1if x=1,4if x=4,5if x=5,6if x=6,2if x{2,3},

    then (X,f) is an interior GE-algebra in which X is not transitive. Let F:={1} and G:={1,5,6}. Then F is a prominent interior GE-filter in (X,f) and G is an interior GE-filter in (X,f) with FG. But G is not prominent interior GE-filter since 5(34)=55=1G and 5G but ((43)3)4=(33)4=14=4G.

    Definition 3.19. Let (X,f) be an interior GE-algebra and let F be a subset of X which satisfies (2.15). Then F is called:

    A prominent interior GE-filter of type 1 in (X,f) if it satisfies:

    (x,y,zX)(x(yf(z))F,f(x)F((f(z)y)y)f(z)F). (3.3)

    A prominent interior GE-filter of type 2 in (X,f) if it satisfies:

    (x,y,zX)(x(yf(z))F,f(x)F((zf(y))f(y))zF). (3.4)

    Example 3.20. (1). Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 7,

    Table 7.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 1 1
    3 1 2 1 2 2
    4 1 1 1 1 1
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x{1,3}2if x=2,4if x=4,5if x=5.

    Then (X,f) is an interior GE-algebra and F:={1,3} is a prominent interior GE-filter of type 1 in (X,f).

    (2). Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 8,

    Table 8.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 1 1
    3 1 1 1 4 1
    4 1 1 1 1 5
    5 1 1 3 4 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,2if x{2,3,4,5}.

    Then (X,f) is an interior GE-algebra and F:={1,3} is a prominent interior GE-filter of type 2 in (X,f).

    Theorem 3.21. In an interior GE-algebra, every prominent interior GE-filter is of type 1.

    Proof. Let F be a prominent interior GE-filter in an interior GE-algebra (X,f). Let x,y,zX be such that x(yf(z))F and f(x)F. Then xF by (2.22). It follows from (2.18) that ((f(z)y)y)f(z)F. Hence F is a prominent interior GE-filter of type 1 in (X,f).

    The following example shows that the converse of Theorem 3.21 may not be true.

    Example 3.22. Let X={1,2,3,4,5} be a set with the Cayley table which is given in Table 9,

    Table 9.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 1 1
    3 1 1 1 1 5
    4 1 1 3 1 1
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,2if x{2,3},5if x{4,5}.

    Then (X,f) is an interior GE-algebra and F:={1} is a prominent interior GE-filter of type 1 in (X,f). But it is not a prominent interior GE-filter in (X,f) since 1(34)=1F but (43)3)4=4F.

    The following example shows that prominent interior GE-filter and prominent interior GE-filter of type 2 are independent of each other, i.e., prominent interior GE-filter is not prominent interior GE-filter of type 2 and neither is the inverse.

    Example 3.23. (1). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 10,

    Table 10.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 1 1
    3 1 5 1 1 5
    4 1 1 1 1 1
    5 1 3 3 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,4if x{3,4}5if x{2,5}.

    Then (X,f) is an interior GE-algebra and F:={1} F is a prominent interior GE-filter in (X,f). But it is not a prominent interior GE-filter of type 2 since 1(5f(2))=55=1F and f(1)=1F but ((2f(5))f(5))2=((25)5)2=(15)2=52=3F.

    (2). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 11,

    Table 11.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 1 1
    3 1 2 1 1 1
    4 1 1 1 1 1
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,5if x{2,3,4,5}.

    Then (X,f) is an interior GE-algebra and F:={1} is a prominent interior GE-filter of type2 in (X,f). But it is not a prominent interior GE-filter in (X,f) since 1(23)=11=1F and 1F but ((32)2)3=(22)3=13=3F.

    The following example shows that prominent interior GE-filter of type 1 and prominent interior GE-filter of type 2 are independent of each other.

    Example 3.24. (1). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 12,

    Table 12.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 5 5
    3 1 1 1 1 1
    4 1 1 1 1 1
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,3if x{2,3},5if x{4,5}.

    Then (X,f) is an interior GE-algebra and F:={1,2,4} is a prominent interior GE-filter of type 1 in (X,f). But it is not a prominent interior GE-filter of type 2 since 1(5f(2))=1(53)=11=1F and f(1)=1F but ((2f(5))f(5))2=((25)5)2=(55)2=12=2F.

    (2). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 13,

    Table 13.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 4 4 5
    3 1 1 1 1 1
    4 1 2 2 1 5
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,2if x=2,4if x=4,3if x{3,5}.

    Then (X,f) is an interior GE-algebra and F:={1} is a prominent interior GE-filter of type 2 in (X,f). But it is not a prominent interior GE-filter of type 1 in (X,f) since 1(5f(2))=1(52)=11=1F and f(1)F but ((f(2)5)5)f(2)=((25)5)2=(55)2=12=2F.

    The following example shows that interior GE-filter and prominent interior GE-filter of type 1 are independent of each other.

    Example 3.25. (1). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 14,

    Table 14.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 5 5 5
    3 1 1 1 1 1
    4 1 1 1 1 1
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,2if x=2,5if x{3,4,5}.

    Then (X,f) is an interior GE-algebra and F:={1} is an interior GE-filter in (X,f). But F is not prominent interior GE-filter of type 1 since 1(5f(2))=1(52)=11=1F and f(1)=1F but ((f(2)5)5)2=((25)5)2=(55)2=12=2F.

    (2). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 15,

    Table 15.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 5 1 5
    3 1 2 1 1 1
    4 1 1 3 1 5
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x{1,2,4},5if x{3,5}.

    Then (X,f) is an interior GE-algebra and F:={1,2} is a prominent interior GE-filter of type 1 in (X,f). But it is not an interior GE-filter in (X,f) since 24=1 and 2F but 4F.

    The following example shows that interior GE-filter and prominent interior GE-filter of type 2 are independent of each other.

    Example 3.26. (1). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 16,

    Table 16.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 1 1
    3 1 2 1 1 2
    4 1 2 3 1 5
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x{1,4}2if x=2,3if x=3,5if x=5.

    Then (X,f) is an interior GE-algebra and F:={1,4} is an interior GE-filter in (X,f). But F is not prominent interior GE-filter of type 2 since 4(2f(3))=4(23)=41=1F and f(4)=1F but ((3f(2))f(2))3=((32)2)3=(22)3=13=3F.

    (2). Let X={1,2,3,4,5} be a set with the Cayley table which is given in the following Table 17,

    Table 17.  Cayley table for the binary operation "".
    1 2 3 4 5
    1 1 2 3 4 5
    2 1 1 1 1 5
    3 1 1 1 1 1
    4 1 1 1 1 5
    5 1 1 1 1 1

     | Show Table
    DownLoad: CSV

    and define a mapping f on X as follows:

    f:XX,x{1if x=1,3if x{2,3,4,5}.

    Then (X,f) is an interior GE-algebra and F:={1,2,5} is a prominent interior GE-filter of type 2 in (X,f). But it is not an interior GE-filter in (X,f) since 54=1 and 5F but 4F.

    Before we conclude this paper, we raise the following question.

    Question. Let (X,f) be an interior GE-algebra. Let F and G be interior GE-filters in (X,f). If FG and F is a prominent interior GE-filter of type 1 (resp., type 2) in (X,f), then is G also a prominent interior GE-filter of type 1 (resp., type 2) in (X,f)?

    We have introduced the concept of a prominent interior GE-filter (of type 1 and type 2), and have investigated their properties. We have discussed the relationship between a prominent GE-filter and a prominent interior GE-filter and the relationship between an interior GE-filter and a prominent interior GE-filter. We have found and provide examples where any interior GE-filter is not a prominent interior GE-filter and any prominent GE-filter is not a prominent interior GE-filter. We have provided conditions for an interior GE-filter to be a prominent interior GE-filter. We have given conditions under which an internal GE-filter larger than a given internal GE filter can become a prominent internal GE-filter, and have provided an example describing it. We have considered the relationship between a prominent interior GE-filter and a prominent interior GE-filter of type 1. We have found and provide examples to verify that a prominent interior GE-filter of type 1 and a prominent interior GE-filter of type 2, a prominent interior GE-filter and a prominent interior GE-filter of type 2, an interior GE-filter and a prominent interior GE-filter of type 1, and an interior GE-filter and a prominent interior GE-filter of type 2 are independent each other. In future, we will study the prime and maximal prominent interior GE-filters and their topological properties. Moreover, based on the ideas and results obtained in this paper, we will study the interior operator theory in related algebraic systems such as MV-algebra, BL-algebra, EQ-algebra, etc. It will also be used for pseudo algebra systems and further to conduct research related to the very true operator theory.

    This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2016R1D1A1B02006812).

    The authors wish to thank the anonymous reviewers for their valuable suggestions.

    All authors declare no conflicts of interest in this paper.



    [1] Z. Bi, L. Jing, C. Sun, M. Shan, YOLOX++ for transmission line abnormal target detection, IEEE Access, 11 (2023), 38157–38167. https://doi.org/10.1109/ACCESS.2023.3268106 doi: 10.1109/ACCESS.2023.3268106
    [2] R. Li, Y. Chen, C. Sun, W. Fu, Improved algorithm for small target detection of traffic signs on YOLOv5s, in 2023 4th International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI), (2023), 339–344. https://doi.org/10.1109/ICHCI58871.2023.10278065
    [3] F. A. Kurniadi, C. Setianingsih, R. E. Syaputra, Innovation in livestock surveillance: Applying the YOLO algorithm to UAV imagery and videography, in 2023 IEEE 9th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA), (2023), 246–251. https://doi.org/10.1109/ICSIMA59853.2023.10373473
    [4] H. Y. Jiang, F. Hu, X. Q. Fu, C. R. Chen, C. Wang, L. X. Tian, et al., YOLOv8-Peas: a lightweight drought tolerance method for peas based on seed germination vigor, Front. Plant Sci., 14 (2023). https://doi.org/10.3389/fpls.2023.1257947 doi: 10.3389/fpls.2023.1257947
    [5] R. Girshick, Fast R-CNN, in 2015 IEEE International Conference on Computer Vision (ICCV), (2015), 1440–1448. https://doi.org/10.1109/ICCV.2015.169
    [6] R. Girshick, J. Donahue, T. Darrell, J. Malik, Rich feature hierarchies for accurate object detection and semantic segmentation, in 2014 IEEE Conference on Computer Vision and Pattern Recognition, (2014), 580–587. https://doi.org/10.1109/CVPR.2014.81
    [7] K. He, G. Gkioxari, P. Dollár, R. Girshick, Mask R-CNN, in 2017 IEEE International Conference on Computer Vision (ICCV), (2017), 2980–2988. https://doi.org/10.1109/ICCV.2017.322
    [8] S. Ren, K. He, R. Girshick, J. Sun, Faster R-CNN: Towards real-time object detection with region proposal networks, in IEEE Transactions on Pattern Analysis and Machine Intelligence, 39 (2017), 1137–1149. https://doi.org/10.1109/TPAMI.2016.2577031 doi: 10.1109/TPAMI.2016.2577031
    [9] J. Bai, H. Zhang, Z. Li, The generalized detection method for the dim small targets by faster R-CNN integrated with GAN, in 2018 IEEE 3rd International Conference on Communication and Information Systems (ICCIS), (2018), 1–5. https://doi.org/10.1109/ICOMIS.2018.8644960
    [10] Z. Wang, Y. Cao, J. Li, A detection algorithm based on improved faster R-CNN for spacecraft components, in 2023 IEEE International Conference on Image Processing and Computer Applications (ICIPCA), (2023), 1–5. https://doi.org/10.1109/ICIPCA59209.2023.10257992
    [11] Z. H. He, X. Ye, Y. Li, Compact sparse R-CNN: Speeding up sparse R-CNN by reducing iterative detection heads and simplifying feature pyramid network, AIP Adv., 13 (2023). https://doi.org/10.1063/5.0146453 doi: 10.1063/5.0146453
    [12] J. Redmon, S. Divvala, R. Girshick, A. Farhadi, You Only look once: Unified, real-time object detection, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2016), 779–788. https://doi.org/10.1109/CVPR.2016.91
    [13] J. Redmon, A. Farhadi, YOLO9000: Better, faster, stronger, in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2017), 6517–6525. https://doi.org/10.1109/CVPR.2017.690
    [14] J. Redmon, A. Farhadi, YOLOv3: An incremental improvement, preprint, arXiv: 1804.02767. https://doi.org/10.48550/arXiv.1804.02767
    [15] A. Bochkovskiy, C. Y. Wang, H. Y. M. Liao, YOLOv4: Optimal speed and accuracy of object detection, preprint, arXiv: 2004.10934. https://doi.org/10.48550/arXiv.2004.10934
    [16] X. Zhu, S. Lyu, X. Wang, Q. Zhao, TPH-YOLOv5: Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios, in 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), (2021), 2778–2788. https://doi.org/10.1109/ICCVW54120.2021.00312
    [17] W. Liu, D. Anguelov, D. Erhan, C. Szegedy, S. E. Reed, C. Y. Fu, et al., SSD: Single shot multibox detector, in Computer Vision–ECCV 2016. ECCV 2016. Lecture Notes in Computer Science(), Springer, 9905 (2016), 21–37. https://doi.org/10.1007/978-3-319-46448-0_2
    [18] T. Y. Lin, P. Goyal, R. Girshick, K. He, P. Dollár, Focal loss for dense object detection, IEEE Trans. Pattern Anal. Mach. Intell., 42 (2020), 318–327. https://doi.org/10.1109/TPAMI.2018.2858826 doi: 10.1109/TPAMI.2018.2858826
    [19] Z. Yao, W. Douglas, S. O'Keeffe, R. Villing, Faster YOLO-LITE: Faster object detection on robot and edge devices, in RoboCup 2021: Robot World Cup XXIV, Springer, 13132 (2021), 226–237. https://doi.org/10.1007/978-3-030-98682-7_19
    [20] C. Sun, Y. B. Ai, S. Wang, W. D. Zhang, Mask-guided SSD for small-object detection, Appl. Intell., 51 (2021), 3311–3322. https://doi.org/10.1007/s10489-020-01949-0 doi: 10.1007/s10489-020-01949-0
    [21] H. Wang, H. Qian, S. Feng, GAN-STD: small target detection based on generative adversarial network, J. Real-Time Image Process., 21 (2024), 65. https://doi.org/10.1007/s11554-024-01446-4 doi: 10.1007/s11554-024-01446-4
    [22] C. Chen, M. Y. Liu, O. Tuzel, J. Xiao, R-CNN for small object detection, in Computer Vision—ACCV 2016, Springer, 10115 (2017), 214–230. https://doi.org/10.1007/978-3-319-54193-8_14
    [23] X. Chen, H. Fang, T. Y. Lin, R. Vedantam, S. Gupta, P. Dollár, et al., Microsoft COCO captions: Data collection and evaluation server, preprint, arXiv: 1504.00325. https://doi.org/10.48550/arXiv.1504.00325
    [24] G. S. Xia, X. Bai, J. Ding, Z. Zhu, S. Belongie, J. Luo, et al., DOTA: A large-scale dataset for object detection in aerial images, in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, (2018), 3974–3983. https://doi.org/10.1109/CVPR.2018.00418
    [25] S. Zhang, R. Benenson, B. Schiele, CityPersons: A diverse dataset for pedestrian detection, in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), (2017), 4457–4465. https://doi.org/10.1109/CVPR.2017.474
    [26] J. Wang, K. Chen, R. Xu, Z. Liu, C. C. Loy, D. Lin, CARAFE: Content-aware reassembly of features, in 2019 IEEE/CVF International Conference on Computer Vision (ICCV), (2019), 3007–3016. https://doi.org/10.1109/ICCV.2019.00310
    [27] R. Sunkara, T. Luo, No more strided convolutions or pooling: A new CNN building block for low-resolution images and small objects, in Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2022, Cham: Springer Nature Switzerland, 13715 (2022), 443–459. https://doi.org/10.1007/978-3-031-26409-2_27
    [28] H. Rezatofighi, N. Tsoi, J. Gwak, A. Sadeghian, I. Reid, S. Savarese, Generalized intersection over union: A metric and a loss for bounding box regression, in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), (2019), 658–666. https://doi.org/10.1109/CVPR.2019.00075
    [29] Z. Zheng, P. Wang, W. Liu, J. Li, R. Ye, D. Ren, Distance-IoU loss: Faster and better learning for bounding box regression, in AAAI Conference on Artificial Intelligence, (2019). https://doi.org/10.48550/arXiv.1911.08287
    [30] Y. Cao, Z. He, L. Wang, W. Wang, Y. Yuan, D. Zhang, et al., VisDrone-DET2021: The vision meets drone object detection challenge results, in 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), (2021), 2847–2854. https://doi.org/10.1109/ICCVW54120.2021.00319
    [31] C. Liu, D. G. Yang, L. Tang, X. Zhou, Y. Deng, A lightweight object detector based on spatial-coordinate self-attention for UAV aerial images, Remote Sens., 15 (2022), 83. https://doi.org/10.3390/rs15010083 doi: 10.3390/rs15010083
    [32] H. J. Nie, H. L. Pang, M. Y. Ma, R. K. Zheng, A lightweight remote sensing small target image detection algorithm based on improved YOLOv8, Sensors, 24 (2024), 2952. https://doi.org/10.3390/s24092952 doi: 10.3390/s24092952
  • This article has been cited by:

    1. Sun Shin Ahn, Ravikumar Bandaru, Young Bae Jun, Imploring interior GE-filters in GE-algebras, 2021, 7, 2473-6988, 855, 10.3934/math.2022051
  • Reader Comments
  • © 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(1257) PDF downloads(114) Cited by(2)

Figures and Tables

Figures(12)  /  Tables(5)

Other Articles By Authors

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog