Preprint
Article

This version is not peer-reviewed.

Multi-Dimensional Attention for Small Object Detection in Airport Obstacle Limitation Airspace

Submitted:

31 October 2026

Posted:

03 November 2026

You are already at the latest version

Abstract
In airport clearance airspace, illegally flown unmanned aerial vehicles and birds in long-range imagery occupy extremely small pixel proportions, suffer from texture scarcity, and are readily overwhelmed by background clutter; consequently, generic detectors struggle to achieve accurate detection. To this end, this paper proposes a multi-dimensional attention collaborative optimization method that improves small-object detection performance across three complementary dimensions, feature enhancement, convolution decoupling, and loss reweighting. Initially, a multi-semantic self-attention module is developed, which employs multi-scale depthwise separable convolutions and group normalization to extract multi-semantic spatial features while guiding channel attention learning, thereby enhancing small-object response intensity. Subsequently, a receptive-field decoupled attention module is introduced, which decouples positions within the receptive field of standard convolution and independently assigns attention weights, thereby breaking the parameter-sharing limitation of convolution kernels and improving boundary regression sensitivity. Finally, a unified intersection-over-union loss is proposed, which adjusts the focus on predicted boxes of different qualities through dynamic scaling and annealing strategies, thereby strengthening the fine-grained refinement of high-quality boxes. Furthermore, the effectiveness of these three components is validated on four datasets: TIB, UB-SOD, Drone, and Flying. Notably, on TIB, the proposed method achieves 87.4% mAP50 and 38.9% mAP50-95, improving the baseline by 4.7% and 3.2%, respectively; moreover, it exhibits significant advantages over other methods.
Keywords: 
;  ;  ;  ;  

1. Introduction

Airport clearance airspace security constitutes one of the most risk-sensitive links in the civil aviation transportation system. Birds and UAVs, as two principal hazard sources threatening airport clearance airspace, must be effectively detected and identified, as this directly bears on aircraft takeoff and landing safety and airport operational order. In recent years, with the rapid development of the low-altitude economy and the growing frequency of non-cooperative UAV activities, the risk of aerial intrusion into airport clearance airspace has risen markedly. Statistics indicate that bird strikes and UAV interference have become significant contributors to flight delays, flight returns, and even safety accidents. Against this backdrop, developing an efficient and accurate object detection system for clearance airspace has become an urgent requirement for establishing airport intelligent perception and security defense systems.
With respect to technical perspective, current object detection in airport clearance airspace relies primarily on multi-source means such as radar, electro-optical (EO) systems, radio reconnaissance, and acoustic sensing. Among these, EO detection plays an irreplaceable role in short-range object recognition and classification, owing to its high spatial resolution, visual confirmation capability, and low-cost deployment advantages. Nevertheless, EO image-based object detection faces extremely stringent challenges: birds and UAVs at long range often occupy only tens or even a few pixels, with highly limited texture details, thus constituting a typical extremely small object detection problem. Meanwhile, the background complexity of clearance airspace scenarios is exceptionally high—runway lights, building edges, and cloud textures may strongly interfere with objects, rendering weak small-object features highly susceptible to being overwhelmed. Even more challengingly, although birds and UAVs differ markedly in morphology, both degrade into blurred bright spots under low-resolution imaging conditions, leading to a sharp decline in inter-class separability. Collectively, these factors cause generic object detectors to suffer from severe missed detection and false detection in airport clearance airspace scenarios.
With respect to perspective of feature extraction and representation learning, existing methods exhibit three systematic deficiencies in addressing the above challenges. First, feature coordination across multiple semantic levels is insufficient. In deep features, small objects suffer severe loss of spatial structure due to downsampling; in shallow features, spatial details are preserved but semantic discriminability is lacking. Most existing attention mechanisms model only a single semantic level or a single dimension, making it difficult to achieve deep synergy between spatial structure and channel semantics; consequently, feature representations of bird and UAV objects become blurred and inconsistent. Second, the parameter-sharing mechanism of convolution kernels limits fine-grained enhancement of weak features. Standard convolution shares weights within a sliding window, so discriminative features in small-object regions are readily diluted by background textures. Although deformable and dilated convolutions enlarge the receptive field, they do not fundamentally achieve position-wise decoupling of object and background within the window, and weak signals in small-object regions remain difficult to strengthen independently. Third, the allocation of loss weights in the boundary regression stage lacks adaptation to the localization demands of small objects. Commonly used IoU-based losses adopt a relatively uniform weighting strategy for high-quality and low-quality boxes, whereas objects with extremely small pixel proportions are highly sensitive to localization deviations; a slight shift can cause IoU to plummet, and a fixed weighting strategy can hardly guide the model toward smooth optimization in the high-precision localization stage.
To address the above issues, this paper proposes a small-object detection algorithm for airport clearance airspace based on multi-dimensional attention mechanisms. In the feature extraction stage, a Multi-Semantic Self-Attention (MSSA) module uses multi-semantic spatial structures to guide channel self-attention learning, thereby enhancing the feature discriminability of small objects under complex clearance backgrounds and alleviating semantic discrepancies among multi-scale features. In the feature enhancement stage, a Receptive Field Decoupled Attention (RFDA) module independently assigns attention weights to different positions within the receptive-field sliding window, achieving fine-grained decoupling between objects and background and suppressing interference from runway lights and ground background on tiny objects. In the loss optimization stage, a Unified-IoU (UIoU) loss dynamically adjusts the loss weights of predicted boxes with different qualities without modifying geometric measurement rules, and introduces a cosine annealing strategy and an inverse focal mechanism to achieve a smooth transition from rapid convergence to high-precision localization. MSSA and RFDA enhance the distinguishability of small objects in the feature extraction stage, while UIoU optimizes the boundary learning process in the loss function stage, thereby systematically alleviating the core difficulties of weak features, susceptibility to being overwhelmed, and localization difficulty in small-object detection within clearance airspace.
The main contributions of this paper are summarized as follows:
1. We propose a Multi-Semantic Self-Attention module. By guiding channel self-attention learning with multi-semantic spatial structures, this module achieves deep synergy between spatial and channel dimensions, thereby enhancing the capture of weak features of small objects such as birds and UAVs across different semantic levels. Moreover, it alleviates semantic discrepancies among multi-scale features and improves feature discriminability for small objects in complex clearance scenarios.
2. Furthermore, we develop a Receptive Field Decoupled Attention module. Through the unfold operation, this module decomposes the input feature map into spatial features of multiple sliding windows, employs a dual-branch aggregation module to generate position-specific attention weights, and leverages parallel depthwise separable convolution branches to enhance local structural representation. Consequently, RFDA effectively overcomes the limitation of parameter sharing in standard convolution, suppresses background interference, and markedly improves the capture of weak features of small objects.
3. In addition, we introduce a Unified-IoU loss. This loss indirectly adjusts the loss weights of boxes with varying qualities by dynamically scaling the sizes of predicted boxes and ground-truth (GT) boxes. Furthermore, it incorporates a cosine annealing strategy and an inverse focal mechanism, enabling a smooth transition from rapid convergence to high-precision localization and providing a more rational optimization pathway for boundary regression of extremely small objects.
4. Finally, the effectiveness of the proposed method is validated on datasets related to airport clearance airspace. Specifically, MSSA, RFDA, and UIoU are integrated into a unified detection framework, and ablation and comparative experiments are conducted on the TIB, Drone, and Flying datasets, as well as on UB-SOD data for airport clearance scenarios. The results demonstrate that the three modules complement each other in the two stages of feature extraction and loss optimization, thereby significantly improving the detection accuracy and localization quality for small objects such as birds and UAVs.
The remainder of this paper is organized as follows. Section 2 reviews related work and research progress on small object detection and attention mechanisms. Section 3 details the proposed MSSA, RFDA, and UIoU. Section 4 presents the experimental setup, ablation studies, and comparative results. Section 5 concludes this paper and outlines future work.

3. Proposed Method

In this section, we introduce the motivation and overview of our Multi-dimensional attention algorithm for small object detection in airport obstacle limitation airspace, and provide the details of the proposed network by several aspects. First, MSSA enhances the response intensity of small objects in feature maps along the spatial dimension and increases the probability of generating high-quality predicted boxes. Second, RFDA fundamentally overcomes the limitation imposed by convolution kernel parameter sharing on small-object feature extraction, thereby improving the sensitivity of the regression branch to boundary variations of small objects. Third, UIoU strengthens the localization of high-quality boxes in the later training stage, ensuring that the model learns more precise boundary regression from features enhanced by MSSA and RFDA.

3.1. Multi-Semantic Self-Attention Module

We propose a plug-and-play MSSA module. This module guides channel self-attention learning through multi-semantic spatial structures, thereby enhancing the feature discriminability of small objects in complex scenarios while alleviating semantic discrepancies among multi-scale features and achieving deep synergy between spatial and channel dimensions. The overall architecture is illustrated in Figure 2. Specifically, this structure operates from two aspects. At the spatial level, it leverages multi-semantic spatial structures to capture weak features of small objects at different semantic levels. At the channel level, it employs channel self-attention to enhance the global consistency of small-object features and avoid information loss caused by semantic ambiguity. Through the above operations, the proposed module effectively addresses the core challenges confronting small object detection—namely, the extremely small pixel proportions occupied by objects, limited texture details, and susceptibility to background interference—which lead to blurred feature representations and render small objects prone to being overwhelmed by large-scale objects or noise.
Figure 1. Schematic illustration of the Multi-Semantic Self-Attention architecture.
Figure 1. Schematic illustration of the Multi-Semantic Self-Attention architecture.
Preprints 241651 g001
Multi-Semantic Spatial Structure. Given an input feature map X ∈ R B × C × H × W , MSSA initially decouples it along the height and width dimensions and, through global average pooling, respectively creates two unidirectional 1D sequence structures, X H ∈ R B × C × W and X W ∈ R B × C × H . Such a decoupling operation preserves spatial structural information, thereby laying the foundation for subsequent multi-semantic extraction.
In particular, to capture the spatial distributions of small objects at different scales, we partition each sequence feature set into K independent sub-features, X H i and X W i , where the channel number of each sub-feature C/K is given by:
X H i = X H [ · , ( i − 1 ) × C K : i × C K , · ] X W i = X W [ · , ( i − 1 ) × C K : i × C K , · ]
Subsequently, we apply depthwise separable 1D convolutions with kernel sizes of 3, 5, 7, 9 to these K sub-features. This multi-scale design enables the module to simultaneously capture local details and contextual associations of small objects, effectively improving its responsiveness to weak signals from small objects. Meanwhile, we employ a lightweight shared convolution for cross-dimensional alignment, thereby enhancing feature consistency between the H and W dimensions and preventing the spatial structure fragmentation of small objects caused by dimensional decoupling:
X ˜ H i = DWConv k i K − C K ( X H i ) X ˜ W i = DWConv k i K − C K ( X W i )
Upon completing multi-semantic spatial feature extraction, we concatenate all sub-features along the channel dimension and normalize them using Group Normalization. By independently normalizing each sub-feature, Group Normalization effectively preserves the semantic independence among sub-features, thereby preventing background or other objects from interfering with small-object features. Subsequently, an initial spatial attention map is generated via the Sigmoid function.
To further enhance fine-grained perception of small-object regions, we apply a 3×3 depthwise separable convolution to the spatial attention map output by Sigmoid. This operation, while introducing minimal computational cost, locally smooths and sharpens the attention distribution, thereby strengthening spatial consistency and preventing activation fragmentation or diffusion into background regions caused by pixel sparsity in small-object areas. Finally, the convolution-refined spatial attention map is element-wise multiplied with the original input X to obtain the spatially modulated feature X s :
A t t n H = σ ( G N H K ( C o n c a t ( X ˜ H 1 , X ˜ H 2 , … , X ˜ H K ) ) ) A t t n W = σ ( G N W K ( C o n c a t ( X ˜ W 1 , X ˜ W 2 , … , X ˜ W K ) ) ) A t t n s p a t i a l = D W C o n v 3 × 3 ( A t t n H × A t t n W ) X 5 = A t t n s p a t i a l × X
The aforementioned spatial modulation process, through multi-semantic guidance and local attention refinement, enhances the spatial response intensity and consistency for small-object regions.
Channel Self-Attention Structure. Subsequently, to further improve the feature discriminability of small objects, MSSA performs channel-wise self-attention computation on X s to model inter-channel dependencies. Small objects typically lack rich texture information, and their recognition relies more heavily on cross-channel feature synergy. To this end, we first progressively compress X s by reducing the resolution from ( H , W ) to ( H ′ , W ′ ) via average pooling, yielding X p . This operation significantly reduces computational cost while preserving key spatial details; particularly for small objects, retaining their core response regions is more important than maintaining global resolution:
X p = P o o l [ ( H , W ) , ( H ′ , W ′ ) ] ( X 5 )
Subsequently, we adopt an input-aware single-head self-attention mechanism to compute attention along the channel dimension. Here, Q , K , V ∈ R B × C × N , and N = H ′ W ′ , the query, key, and value are generated via linear projection,
Q = F p r o j Q ( X p ) , K = F p r o j K ( X p ) , V = F p r o j V ( X p )
and the attention weights are computed as follows:
X a t t n = A t t n ( Q , K , V ) = S o f t m a x Q K T C V
This channel-wise self-attention effectively captures the synergistic activation patterns of small objects across different channels. Through the self-attention mechanism, these dispersed cues can be aggregated into a stronger comprehensive representation, while alleviating the semantic discrepancies introduced by multi-scale convolution and preventing semantic misalignment of small objects across features at different scales. Ultimately, global average pooling is performed on feature X a t t n , after which a Sigmoid function is employed to generate channel attention weights; these weights are subsequently element-wise multiplied with the original input X s , thereby yielding the final output of MSSA:
M S S A ( X ) = X 5 × σ ( P o o l [ ( H ′ , W ′ ) , ( 1 , 1 ) ] ( X a t t n ) )
The entire MSSA module deeply integrates multi-semantic spatial guidance with channel self-attention in a serial manner. The former captures spatial features of small objects at different semantic levels through multi-scale structures and provides precise spatial priors for the latter; the latter enhances the cross-channel consistency of small-object features via channel-wise self-attention, effectively mitigating feature blurring caused by multi-scale semantic discrepancies, thereby significantly improving the detection and segmentation performance of small objects in complex scenarios.

3.2. Receptive Field Decoupled Attention

To address the issue that standard convolution kernels share parameters within a sliding window, which causes weak features of small objects to be readily overwhelmed by background, this paper proposes a RFDA module. This module aims to independently assign attention weights to different positions within each receptive-field window, thereby achieving fine-grained decoupling between objects and background. The overall architecture is illustrated in Figure 2. Specifically, an unfold operation is first employed to decompose the input feature map into spatial features of multiple sliding windows. Subsequently, a dual-branch aggregation module utilizes average pooling and max pooling to extract global and local information of the windows, respectively, and thereby generates position-specific attention weights. Meanwhile, parallel feature transformation branches enhance local structural representation through depthwise separable convolutions. Finally, the attention weights are element-wise multiplied with the enhanced features, thereby achieving independent weighting of each position within the window. Through this process, RFDA effectively overcomes the limitation of parameter sharing in convolution kernels and significantly improves the capture of weak discriminative features in small-object regions.
Figure 2. Architecture of the Receptive Field Decoupled Attention module.
Figure 2. Architecture of the Receptive Field Decoupled Attention module.
Preprints 241651 g002
Receptive-Field Spatial Feature. To achieve independent attention for each position within the receptive field, we first convert the input feature map into receptive-field spatial features via an Unfold operation, thereby constituting the RFDA. Specifically, given an input X ∈ R C × H × W , a kernel size k × k , and a stride s, the Unfold operation unfolds the pixel values within each receptive-field window of size k × k into a vector and stacks them along the spatial dimension. The resulting tensor has a shape of ( C × k 2 ) × H ′ × W ′ , where H ′ = [ ( H − k ) / s ] + 1 , W ′ = [ ( W − k ) / s ] + 1 . Formally, for the window corresponding to output position ( u , v ) , the pixel values it contains are:
R u , v = { X ( : , u · s + i , v · s + j ) ∣ 0 ≤ i , j < k }
These values are sequentially flattened and stored along the channel dimension of the output tensor. Notably, this operation introduces no learnable parameters, and when the stride is set to s = k , the sliding windows are completely non-overlapping. Figure 3 illustrates in detail an example of extracting 2×2 receptive-field spatial features using Unfold.
Dual-Branch Aggregation Module. To more comprehensively capture the feature distribution of small objects within receptive-field windows, we design a dual-branch aggregation module to generate decoupled attention weights. One branch employs average pooling to aggregate global contextual information of the window, thereby reflecting the overall response intensity of the region where small objects reside; the other branch adopts max pooling to capture local salient peaks within the window, which is particularly critical for the weak yet prominent edge or point features of small objects. Subsequently, the features from the two branches are concatenated along the channel dimension to form a richer window-level description.
We denote the receptive-field spatial features obtained via Unfold as X r f ∈ R k 2 C × H ′ × W ′ . The objective of RFDA is to learn an attention map A r f ∈ R k 2 × H ′ × W ′ , where A r f ( i , u , v ) represents the independent weight of the i-th position within the ( u , v ) sliding window. Specifically, for each window position ( u , v ) , we compute the average-pooled feature and the max-pooled feature, respectively:
F avg ( u , v ) = 1 k 2 C ∑ c = 1 k 2 C X r f ( c , u , v ) ∈ R 1
F max ( u , v ) = max 1 ≤ c ≤ k 2 C X r f ( c , u , v ) ∈ R 1
The average-pooled and max-pooled features are concatenated along the channel dimension, thereby yielding a joint descriptor for each sliding window:
F joint ( u , v ) = Concat ( F avg ( u , v ) , F max ( u , v ) ) ∈ R 2
At this stage, F j o i n t has a shape of 2 × H ′ × W ′ . Subsequently, a lightweight grouped convolutional network with 1×1 convolutions is employed—comprising k 2 groups, 2 input channels, and k 2 output channels—to model the interrelationships among different positions within each sliding window. Each group independently learns importance weights for the corresponding positions, with no parameter sharing across groups.
A raw = Softmax g group = k 2 1 × 1 ( F joint ) ∈ R k 2 × H ′ × W ′
Softmax is then applied along the k2 dimension, such that the attention weights within each sliding window sum to 1, thereby preserving the relative scale of features.
Feature Transformation Branch. In parallel with the dual-branch aggregation branch, we design a feature transformation branch, which aims to perform depthwise separable convolution on the original input feature map to extract local structural patterns within the receptive field and enhance feature representation capability. This branch does not participate in attention weight generation; rather, it provides richer feature representations for subsequent weighted fusion.
Specifically, given an input feature map X ∈ R C × H × W , we first apply a depthwise separable convolution with a kernel size of k × k and a stride consistent with that of the Unfold operation. The separable Depthwise convolution decomposes the standard convolution into Depthwise and pointwise convolution operations, significantly reducing the number of parameters and computational cost while maintaining a comparable receptive field. This operation is capable of capturing spatial local patterns within each channel, such as discriminative features of small objects, including edges, corners, or textures.
Subsequently, we normalize the convolution results to accelerate convergence and stabilize training. A ReLU activation function is then employed to introduce nonlinearity and enhance feature discriminability. The resulting feature map is denoted as F r f , whose shape is identical to that of the receptive-field spatial feature X r f , i.e., R k 2 C × H ′ × W ′ . Formally, it is expressed as:
F r f = ReLU ( Norm ( DepthwiseConv k × k ( X ) ) ) ∈ R k 2 C × H ′ × W ′
where D e p t h w i s e C o n v k × k denotes the depthwise separable convolution. This branch introduces no additional attention parameters and improves feature quality solely through convolution and activation functions, thereby providing more robust features for subsequent fusion.
Upon obtaining the decoupled attention weights A r f ∈ R k 2 × H ′ × W ′ and the transformed features F r f ∈ R k 2 C × H ′ × W ′ , we fuse them to derive redefined output decoupled attention weights. Given that the channel dimension of A r f is k 2 , whereas that of F r f is k 2 C , we expand A r f by a factor of C along the channel dimension, rendering its dimension k 2 C × H ′ × W ′ . The expanded attention weights are then element-wise multiplied with F r f , thereby achieving independent weighting for each position within every receptive-field window.
F out = A rf ⊗ F rf
Here, ⊗ denotes the element-wise multiplication after expansion. This operation enables weak signals of small objects within a window to receive significantly higher weights, while background regions are suppressed, consequently accomplishing object-background decoupling at the feature level.
Following the above decoupled weighting, the channel dimension of F o u t remains k 2 C × H ′ × W ′ . To restore it to the output format of standard convolution and feed it into subsequent network layers, we employ a 1×1 convolution to map the number of channels to the target output channels C o u t . This 1×1 convolution linearly recombines the decoupled features, fuses information from different sliding window positions, and ultimately yields the final output feature map.
Y RFDA = Conv 1 × 1 k 2 C → C out ( F out ) ∈ R C out × H ′ × W ′
Through the above operations, RFDA not only fundamentally resolves the parameter-sharing issue of convolution kernels but also, via dual-branch pooling and depthwise separable feature transformation, significantly enhances the feature capture capability for small-object regions.

3.3. Unified-IoU Loss

We propose the Unified-IoU (UIoU) loss. Rather than modifying geometric measurement rules, this method indirectly adjusts the loss weights of boxes of different qualities by dynamically scaling the sizes of predicted boxes and ground-truth (GT) boxes. Furthermore, it incorporates a cosine annealing strategy and an inverse focal mechanism, thereby achieving a smooth transition from rapid convergence to high-precision localization.
Specifically, instead of directly multiplying the losses of predicted boxes of different qualities by coefficients, we proportionally scale both the predicted box P and the GT box G, thereby altering their IoU values and indirectly adjusting the loss values. Let P = ( x p , y p , w p , h p ) , G = ( x g , y g , w g , h g ) , and the scaling factor r > 0 . The scaled P ′ and G ′ maintain their center points unchanged, while their widths and heights are enlarged by a factor of r:
w p ′ = r w p , h p ′ = r h p , w g ′ = r w g , h g ′ = r h g
The scaled IoU is computed as:
I o U s c a l e d = | P ′ ∩ G ′ | | P ′ ∪ G ′ |
The basic regression loss is defined as:
L b o x = 1 − I o U s c a l e d
For a fixed pair of a predicted box and a GT box, when r < 1 , I o U s c a l e d < I o U o r i g , and the larger the original IoU, the more pronounced the decrease. In this case, the loss 1 − I o U s c a l e d of high-quality predicted boxes increases significantly, which is equivalent to assigning them higher weights, whereas the loss of low-quality boxes ( IoU ≈ 0 ) remains essentially unchanged. Consequently, the shrinking operation can guide the model to focus on high-quality boxes. When r > 1 , I o U s c a l e d > I o U o r i g , and the smaller the original IoU, the more pronounced the increase; thus, the loss of low-quality boxes decreases substantially and their weights are weakened, while the relative weight improvement of high-quality boxes is comparatively modest. Therefore, the enlarging operation can guide the model to focus on low-quality boxes and accelerate early-stage convergence. This is because the shrinking operation causes the relative overlap area of two boxes that originally overlapped considerably to decrease more rapidly, whereas boxes that originally do not overlap remain non-overlapping after shrinking; hence, the rate of IoU change is positively correlated with the original IoU.
From the perspective of gradient modulation, differentiating the UIoU loss with respect to the predicted-box parameters yields a gradient whose magnitude is proportional to ∂ L b o x / ∂ θ . The scaling operation indirectly influences the gradient contributions of boxes of different qualities by altering the local curvature of I o U s c a l e d ; specifically, the gradients of high-quality boxes are amplified after shrinking, whereas those of low-quality boxes are suppressed after enlarging.
To dynamically adjust the model’s attention to predicted boxes of different qualities during training, we design a cosine annealing strategy based on bounding-box attention. The scaling factor r ( t ) is defined to vary with the training epoch t as follows:
r ( t ) = r s t a r t + ( r e n d − r s t a r t ) · 1 + cos ( π t / T ) 2 , t ∈ [ 0 , T ]
where T denotes the total number of training epochs, r s t a r t > 1 , r e n d < 1 .
During the initial training stage, where r > 1 , the contribution of low-quality boxes to the gradient is suppressed, thereby preventing interference from anomalous gradients and enabling the model to rapidly converge to a reasonable initial solution. By contrast, in the later training stage, where r < 1 , the losses of high-quality boxes are amplified, and the loss function exhibits stronger convexity in the vicinity of the optimal solution, which facilitates high-precision localization of small objects and enables the model to smoothly transition from “rapid localization” to “precise localization.”
The original Focal Loss [35] excessively focuses on “hard examples” (low-confidence samples), which may trap the model in ambiguous regions caused by overlap or occlusion, consequently degrading detection accuracy at high IoU thresholds. To this end, we propose an inverse focal mechanism (Focal-inv): it increases the weights of easy-to-detect samples (high-confidence ones), thereby directing the model’s attention to predicted boxes that already perform well and further refining their boundaries. The weighting factor is defined as:
w f o c a l = ( c o n f ) γ
where conf ∈ [ 0 , 1 ] denotes the classification confidence of the predicted box, and γ is an adjustable hyperparameter that is initially set to 2 in this paper. Accordingly, this factor increases with confidence, thereby enabling high-confidence small-object predicted boxes to obtain greater gradient contributions. The total UIoU loss is defined as:
L UIoU = ( conf ) γ · ( 1 − IoU scaled )
The gradient with respect to the predicted-box parameter θ is computed as follows:
∂ L UIoU ∂ θ = ( conf ) γ · ∂ ( 1 − IoU scaled ) ∂ θ + ( 1 − IoU scaled ) · γ ( conf ) γ − 1 ∂ conf ∂ θ
The first term constitutes the dominant term, whereas the second term serves as a coupling term between confidence and localization. For low-quality predicted boxes, conf is typically small and ( conf ) γ approaches 0, so the dominant term is suppressed; although γ ( conf ) γ − 1 in the second term may be relatively large, ( 1 − I o U s c a l e d ) is close to 1 and ∂ conf / ∂ θ is generally small, resulting in limited overall contribution. Consequently, the gradients of low-quality boxes are effectively suppressed, while those of high-quality boxes are amplified, enabling the model to concentrate on optimizing boxes that already possess a certain degree of localization accuracy.

4. Experiments

4.1. Datasets

To validate the effectiveness of the proposed method, experiments are conducted on multiple datasets: 1) the TIB dataset for UAV objects over urban areas; 2) the UB-SOD dataset for object detection in airport clearance airspace, which contains two object categories, namely UAVs and birds, yet encompasses a wide variety of UAV and bird species; 3) the Drone dataset for detecting UAV, aircraft, and helicopter objects; and 4) the Flying dataset, which contains more than 40 categories of aircraft and UAV objects. The datasets employed in this paper are predominantly composed of small objects. The data splits of these datasets are described below.

4.1.1. TIB Dataset

The TIB dataset comprises 2,850 images and 2,850 objects. The size-based criterion adopted in Table 1 is defined as the ratio of the ground-truth bounding box size to the image size. According to this criterion, small objects account for 93.65% of all objects, with 2,669 such objects, as presented in Table 1.
Specifically, as presented in Table 2, under the relative-pixel criterion, the TIB dataset contains 1,501 extremely small objects, which account for 56.15% of the total number of small objects. The remaining objects are categorized into small, medium-sized, and large objects: there are 165 medium-sized objects, accounting for 5.93%, and 12 large objects, accounting for 0.42%.

4.1.2. UB-SOD Dataset

The UB-SOD dataset comprises 6,500 images and 14,979 objects. Under the relative-pixel criterion, small objects account for 86.28% of all objects, totaling 12,924 instances, as presented in Table 3. Specifically, the dataset contains two categories of no-fly-zone intrusion objects frequently encountered in airport airspace, namely UAVs and birds. Among these, there are 9,990 extremely small objects, accounting for 66.7%, as presented in Table 4 and Table 5.

4.1.3. Drone Dataset

The Drone dataset contains a total of 11,998 images and 9,975 objects. Under the relative-pixel criterion, small objects account for 3.72% of all objects, with 371 such instances, as presented in Table 6. It should be noted that, in this publicly available dataset, objects in certain images remain unannotated, consequently rendering the number of objects smaller than that of images.

4.1.4. FlyingDataset

The Flying dataset comprises 18,830 images spanning 40 categories, with a total of 24,769 objects. Under the relative-pixel criterion, small objects account for 17.36% of all objects, totaling 4,299 instances, as presented in Table 7.
To validate the effectiveness of the proposed method, experiments are conducted on the aforementioned four datasets, aiming to demonstrate its promising performance in small object detection.

4.2. Evaluation Metrics

In this paper, three evaluation metrics are primarily adopted: mAP50, mAP50-95, and FPS, which respectively denote mean average precision, mean average precision across different IoU thresholds, and frames per second. Specifically, mAP50 and mAP50-95 serve as detection accuracy metrics, with higher values indicating superior detection accuracy; by contrast, FPS functions as a speed metric, representing the number of images the model can process per second, where higher values correspond to faster detection. Given the core requirement for accuracy in small object detection within airport scenarios, this paper places particular emphasis on improving mAP50-95, so as to validate the optimization effect of the proposed algorithm on small-object localization accuracy.

4.3. Ablation Study

The proposed method is developed by improving upon the YOLOv12 model. To validate the three innovations introduced in this paper, we conduct ablation experiments on four datasets, thereby demonstrating that the proposed method yields substantial improvements across various types of small objects. Specifically, the ablation experiment on the TIB dataset is first performed, and the corresponding results are presented in Table 8.
As presented in Table 8, with respect to mAP50, the three proposed innovations yield improvements of 1.3%, 1.2%, and 1.1% over YOLOv12, respectively. Similarly, for mAP50-95, consistent improvements are also achieved, amounting to 0.4%, 0.7%, and 1.0%, respectively. From the data in the fifth and sixth rows, it is evident that the two models can mutually reinforce each other, thereby attaining superior detection performance. Naturally, the combination of the three innovations delivers the best detection performance, with mAP50 and mAP50-95 reaching 87.4% and 38.9%, respectively, representing improvements of 4.7% and 3.2% over the baseline model. Although the detection speed is slightly lower than that of YOLOv12, at 87.5 FPS, it still satisfies the requirements for real-time detection and thus does not compromise the overall performance of the model.
Ablation experiments are further conducted on the UB-SOD dataset, with the corresponding results presented in Table 9.
As presented in Table 9, for mAP50, the three proposed innovations improve YOLOv12 by 0.5%, 1.1%, and 0.4%, respectively. Likewise, consistent improvements are also achieved for mAP50-95, amounting to 0.7%, 0.8%, and 0.4%, respectively. As can be observed from the data in the fifth and sixth rows, the two models can complement each other, further enhancing detection performance. On the UB-SOD dataset, the proposed method achieves the best detection performance, with mAP50 and mAP50-95 reaching 81.4% and 41.3%, respectively, surpassing the baseline model by 2.7% and 1.2%. The detection speed is 69.5 FPS, which satisfies the requirement for real-time detection.
Ablation experiments are further conducted on the Drone dataset, with the corresponding results presented in Table 10.
As presented in Table 10, the proposed method achieves detection performance as high as 98.1% mAP50 and 78.0% mAP50-95 on the Drone dataset. Regarding mAP50, the three proposed innovations improve YOLOv12 by 0.3%, 0.2%, or remain on par with it; likewise, they also exhibit favorable performance on mAP50-95. Furthermore, on the Drone dataset, under the synergistic positive effect of multiple modules, the model’s feature acquisition capability is further enhanced, and the proposed method attains the best detection performance. The detection speed is 55.0 FPS, which satisfies the requirement for real-time detection.
Ablation experiments are further conducted on the Flying dataset, with the corresponding results presented in Table 11.
As presented in Table 11, consistent with the detection trends observed on the other three datasets, the proposed method exhibits a pronounced performance advantage over the baseline model YOLOv12, achieving 65.6% mAP50 and 55.6% mAP50-95, which correspond to improvements of 5.9% and 5.0%, respectively. Meanwhile, each individual module also yields positive effects, with mAP50 improving by 1.0%, 0.3%, and 1.2%, and mAP50-95 improving by 1.1%, 1.3%, and 2.0%, respectively. Nevertheless, the detection speed is suboptimal, amounting to only half that of the original model.

4.4. Performance Comparison with Existing Approaches

Moreover, to demonstrate the superiority of the proposed method, we compare it with other recently developed methods on the TIB dataset, thereby verifying its practical applicability. The comparison results are presented in Table 12. Notably, for those methods that do not report the mAP50-95 metric, the corresponding results are adopted from references [36] and [37]. As can be observed, the proposed method still exhibits a pronounced advantage over existing mainstream methods.
To more intuitively examine the detection performance of the proposed method on small objects, we select several representative images for visualization, thereby demonstrating its effectiveness. The detection results are presented in Figure 4. As can be observed from Figure 4, the proposed method achieves superior detection performance for objects with complex backgrounds, extremely small sizes, and blurred edges. Although YOLO26 can also correctly detect objects, the predicted boxes generated by the proposed method align more closely with the true extent of the objects and yield higher prediction confidence. For instance, in the third column, the ground-truth boxes do not tightly enclose the objects during annotation; nevertheless, the proposed method can rectify the positions of the predicted boxes, thereby yielding improved detection results.
To validate the effectiveness of the proposed method for illegally flown UAVs and bird objects in airport clearance airspace, we present the detection results on the UB-SOD airport dataset, as illustrated in Figure 5. The first and third rows correspond to the original images to be detected, whereas the second and fourth rows present the corresponding detection results. The UAV objects and bird objects in the figure are all small or extremely small objects. Moreover, under conditions of blurred and complex backgrounds, the proposed method still achieves favorable detection performance. Even when dense bird flocks are present in the images, the proposed method maintains robust performance, further verifying its applicability.

5. Conclusions

In this paper, to address the challenges of small pixel proportions, scarce textures, and strong background interference in the detection of illegally flown UAVs and small bird objects within airport clearance airspace, we propose a multi-dimensional attention collaborative optimization method. This method optimizes detection from three aspects: feature enhancement, convolution decoupling, and loss reweighting. Specifically, MSSA guides channel attention learning through multi-semantic spatial features, thereby enhancing the response intensity of small objects. RFDA decouples positions within the receptive field of standard convolution and independently assigns attention weights, thereby breaking through the parameter-sharing limitation and enhancing boundary regression sensitivity. UIoU adjusts the attention to predicted boxes of different qualities via dynamic scaling and an annealing strategy, thereby strengthening the fine refinement of high-quality boxes. Ablation experiments validate the complementarity of the three modules across the feature extraction and loss optimization stages, demonstrating the effectiveness and robustness of this collaborative framework in densely populated small-object scenarios. Experiments on four datasets—TIB, UB-SOD, Drone, and Flying—demonstrate that the proposed method achieves significant performance improvements in small object detection tasks within airport clearance airspace.
Future work will proceed along two directions. The first involves exploring the deployment efficiency of MSSA and RFDA in lightweight networks to satisfy the engineering requirements of real-time detection in airport clearance airspace. The second focuses on extending the method to infrared and visible-light multimodal fusion scenarios to further enhance detection robustness under complex meteorological conditions.

Author Contributions

Conceptualization, S.L. and G.G.; methodology, S.L.; software, S.M.; validation, S.L., and G.G.; formal analysis, S.M. and X.J.; investigation, G.Z. and H.Z.; resources, Y.X.; data curation, S.M. and X.J.; writing—original draft preparation, S.L.; writing—review and editing, S.M.; visualization, S.L.; supervision, G.G.; project administration, S.L.; funding acquisition, S.L. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China under Grant 62301623, the Natural Science Youth Foundation of Zhongyuan University of Technology under Grant K2025YB003, Henan Provincial Science and Technology R&D Plan Joint Fund under Grant 5321260028, and Henan Provincial Science and Technology Research Project under Grant 262102210132.

Data Availability Statement

The data supporting the findings of this study are available from the cor responding author upon reasonable request, subject to applicable ethical and institutional restrictions.

Acknowledgments

We would like to express our sincere gratitude to our colleagues who provided constructive suggestions and valuable feedback during this study. Their insightful comments and thoughtful discussions greatly helped us improve the research design and refine the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
UAV Unmanned Aerial Vehicle
EO Electro-Optical
MSSA Multi-Semantic Self-Attention
RFDA Receptive Field Decoupled Attention
UIoU Unified-IoU
GT Ground-Truth
FPN Feature Pyramid Network
DIoU Distance IoU
CIoU Complete IoU

References

  1. Wang X, Wang A, Yi J, et al. Small object detection based on deep learning for remote sensing: A comprehensive review. Remote Sensing 2023, 15(13), 3265. [CrossRef]
  2. Dalal N, Triggs B. Histograms of oriented gradients for human detection//2005 IEEE computer society conference on computer vision and pattern recognition (CVPR’05). IEEE, 2005, 1: 886-893.
  3. Felzenszwalb P F, Girshick R B, McAllester D, et al. Object detection with discriminatively trained part-based models. IEEE transactions on pattern analysis and machine intelligence 2009, 32(9), 1627-1645. [CrossRef] [PubMed]
  4. Bosquet B, Mucientes M, Brea V M. STDnet: Exploiting high resolution feature maps for small object detection. Engineering Applications of Artificial Intelligence 2020, 91, 103615. [CrossRef]
  5. Cao J, Bao W, Shang H, et al. GCL-YOLO: A GhostConv-based lightweight yolo network for UAV small object detection. Remote Sensing 2023, 15(20), 4932. [CrossRef]
  6. Sun S, Mo B, Xu J, et al. Multi-YOLOv8: An infrared moving small object detection model based on YOLOv8 for air vehicle. Neurocomputing 2024, 588, 127685. [CrossRef]
  7. Lin T Y, Dollár P, Girshick R, et al. Feature pyramid networks for object detection//2017 IEEE conference on computer vision and pattern recognition (CVPR). IEEE, 2017: 936-944.
  8. Zhang H, Li M, Miao D, et al. Construction of a feature enhancement network for small object detection. Pattern Recognition 2023, 143, 109801. [CrossRef]
  9. Zhang Y, Zhang H, Huang Q, et al. DsP-YOLO: An anchor-free network with DsPAN for small object detection of multiscale defects. Expert Systems with Applications 2024, 241, 122669. [CrossRef]
  10. Miri Rekavandi A, Rashidi S, Boussaid F, et al. Transformers in small object detection: A benchmark and survey of state-of-the-art. ACM Computing Surveys 2025, 58(3), 1-33. [CrossRef]
  11. Ni J, Zhu S, Tang G, et al. A small-object detection model based on improved YOLOv8s for UAV image scenarios. Remote Sensing 2024, 16(13), 2465. [CrossRef]
  12. Chen T, Ye Z, Tan Z, et al. MiM-ISTD: Mamba-in-mamba for efficient infrared small-target detection. IEEE Transactions on Geoscience and Remote Sensing 2024, 62, 1-13. [CrossRef]
  13. Zhang Y, Li Z, Siddique A, et al. Infrared small target detection based on interpretation weighted sparse method. IEEE Transactions on Geoscience and Remote Sensing 2025, 63, 1-15. [CrossRef]
  14. Jia G, Cheng Y, Chen T. IRGraphSeg: Infrared small target detection based on hierarchical GNN. IEEE Geoscience and Remote Sensing Letters 2024, 21, 1-5. [CrossRef]
  15. Niu Z, Zhong G, Yu H. A review on the attention mechanism of deep learning. Neurocomputing 2021, 452, 48-62. [CrossRef]
  16. Hu J, Shen L, Sun G. Squeeze-and-excitation networks. Proceedings of the IEEE conference on computer vision and pattern recognition. 2018: 7132-7141.
  17. Woo S, Park J, Lee J Y, et al. Cbam: Convolutional block attention module. European conference on computer vision. Cham: Springer International Publishing, 2018: 3-19.
  18. Wang Q, Wu B, Zhu P, et al. ECA-Net: Efficient channel attention for deep convolutional neural networks//2020 IEEE/CVF conference on computer vision and pattern recognition (CVPR). IEEE, 2020: 11531-11539.
  19. Dosovitskiy A. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv 2020, 2010, 11929.
  20. Wu F, Sun Y, Wang P, et al. Enhanced u-net for lesion segmentation in whole-slide images: Integrating attention mechanisms and multi-scale feature extraction. Ain Shams Engineering Journal 2025, 16(12), 103748. [CrossRef]
  21. Qureshi S A, Sadiq T, Usman A, et al. SAlexNet: Superimposed AlexNet using residual attention mechanism for accurate and efficient automatic primary brain tumor detection and classification. Results in Engineering 2025, 25, 104025. [CrossRef]
  22. Xu K, Ba J, Kiros R, et al. Show, attend and tell: Neural image caption generation with visual attention[C]//International conference on machine learning. PMLR, 2015: 2048-2057.
  23. Mnih V, Heess N, Graves A, et al. Recurrent models of visual attention. Advances in neural information processing systems 2014, 27.
  24. Lu J, Yang J, Batra D, et al. Hierarchical question-image co-attention for visual question answering. Advances in neural information processing systems 2016, 29.
  25. Lin Z, Feng M, Santos C N, et al. A structured self-attentive sentence embedding. arXiv preprint arXiv 2017, 1703, 03130.
  26. Yang Z, Yang D, Dyer C, et al. Hierarchical attention networks for document classification//Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics: Human language technologies. 2016: 1480-1489.
  27. Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. Advances in neural information processing systems 2017, 30.
  28. Shen T, Zhou T, Long G, et al. Disan: Directional self-attention network for rnn/cnn-free language understanding//Proceedings of the AAAI conference on artificial intelligence. 2018, 32(1).
  29. Yu J, Jiang Y, Wang Z, et al. UnitBox: An advanced object detection network//Proceedings of the 24th ACM International Conference on Multimedia. 2016: 516-520.
  30. Rezatofighi H, Tsoi N, Gwak J Y, et al. Generalized intersection over union: A metric and a loss for bounding box regression[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2019: 658-666.
  31. Zheng Z, Wang P, Liu W, et al. Distance-IoU loss: Faster and better learning for bounding box regression//Proceedings of the AAAI Conference on Artificial Intelligence. 2020, 34(07): 12993-13000. [CrossRef]
  32. Gevorgyan Z. SIoU loss: More powerful learning for bounding box regression. arXiv preprint arXiv 2022, 2205, 12740.
  33. Zhang Y F, Ren W, Zhang Z, et al. Focal and efficient IOU loss for accurate bounding box regression. Neurocomputing 2022, 506, 146-157. [CrossRef]
  34. Tong Z, Chen Y, Xu Z, et al. Wise-IoU: Bounding box regression loss with dynamic focusing mechanism. arXiv preprint arXiv 2023, 2301, 10051.
  35. Lin T Y, Goyal P, Girshick R, et al. Focal loss for dense object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 2020, 42(2), 318-327. [CrossRef] [PubMed]
  36. Sun H, Yang J, Shen J, et al. TIB-Net: Drone detection network with tiny iterative backbone. IEEE Access 2020, 8, 130697-130707. [CrossRef]
  37. Zhou X, Yang G, Chen Y, et al. VDTNet: A high-performance visual network for detecting and tracking of intruding drones. IEEE Transactions on Intelligent Transportation Systems 2024, 25(8), 9828-9839. [CrossRef]
  38. Jocher, G., Chaurasia, A., Qiu, J., & Others. (2023). Ultralytics YOLOv8: Real-time object detection and segmentation. Ultralytics, Tech. Rep. YOLOv8-V1.0.
  39. Wang A, Chen H, Liu L, et al. Yolov10: Real-time end-to-end object detection. Advances in neural information processing systems 2024, 37, 107984-108011. [CrossRef]
  40. Khanam R, Hussain M. Yolov11: An overview of the key architectural enhancements. arXiv preprint arXiv 2024, 2410, 17725.
  41. Tian Y, Ye Q, Doermann D. Yolov12: Attention-centric real-time object detectors (2025). arXiv preprint arXiv 2026, 2502, 12524.
  42. Jocher G, Qiu J, Liu M, et al. Ultralytics YOLO26: Unified real-time end-to-end vision models. arXiv preprint arXiv 2026, 2606, 03748.
Figure 3. Procedure of extracting spatial features via Unfold.
Figure 3. Procedure of extracting spatial features via Unfold.
Preprints 241651 g003
Figure 4. Comparison of detection results between the proposed method and other methods on the TIB dataset.
Figure 4. Comparison of detection results between the proposed method and other methods on the TIB dataset.
Preprints 241651 g004
Figure 5. Detection results of the proposed method on the UB-SOD dataset.
Figure 5. Detection results of the proposed method on the UB-SOD dataset.
Preprints 241651 g005
Table 1. Distribution of target sizes in the TIB dataset.
Table 1. Distribution of target sizes in the TIB dataset.
Dataset Criterion(%) Object number Proportion (%) Scope of application
TIB ≤ 0.12 2673 93.79 Small
≤ 0.48 165 5.79 Middle
> 0.48 12 0.42 Large
Table 2. Distribution of uavs pixels in the TIB dataset.
Table 2. Distribution of uavs pixels in the TIB dataset.
Category Absolute pixels Relative
pixels(%)
Object number Proportion (%) Cumulative
proportion(%)
Scope
of application
UAV ≤ 100 ≤ 0.0048 49 1.72 1.72 Tiny
> 100 , ≤ 400 ≤ 0.0193 1091 38.28 40.00
> 400 , ≤ 900 ≤ 0.0434 361 12.67 52.67
> 900 , ≤ 1600 ≤ 0.0772 591 20.74 73.40 Small
> 1600 , ≤ 2500 ≤ 0.12056 581 20.39 93.79
> 2500 , ≤ 10000 ≤ 0.48225 165 5.79 99.58 Middle
> 10000 , ≤ 40000 ≤ 1.92901 12 0.42 100.00 Large
> 40000 > 1.92901 0 0.00 100.00
Table 3. Distribution of target sizes in the UB-SOD dataset.
Table 3. Distribution of target sizes in the UB-SOD dataset.
Dataset Criterion(%) Object number Proportion (%) Scope of application
UB-SOD ≤ 0.12 12924 86.28 Small
≤ 0.48 1635 10.92 Middle
> 0.48 420 2.80 Large
Table 4. Distribution of UAVs pixels in the UB-SOD dataset.
Table 4. Distribution of UAVs pixels in the UB-SOD dataset.
Category Absolute pixels Relative
pixels(%)
Object number Proportion (%) Cumulative
proportion(%)
Scope of application
UAV ≤ 100 ≤ 0.0048 278 3.96 3.96 Tiny
> 100 , ≤ 400 ≤ 0.0193 867 12.34 16.30
> 400 , ≤ 900 ≤ 0.0434 1208 17.20 33.49
> 900 , ≤ 1600 ≤ 0.0772 2049 29.17 62.66 Small
> 1600 , ≤ 2500 ≤ 0.12056 651 9.27 71.93
> 2500 , ≤ 10000 ≤ 0.48225 1564 22.26 94.19 Middle
> 10000 , ≤ 40000 ≤ 1.92901 406 5.78 99.97 Large
> 40000 > 1.92901 2 0.03 100.00
Table 5. Distribution of birds pixels in the UB-SOD dataset.
Table 5. Distribution of birds pixels in the UB-SOD dataset.
Category Absolute pixels Relative
pixels(%)
Object number Proportion (%) Cumulative
proportion(%)
Scope of application
Bird ≤ 100 ≤ 0.0048 217 2.73 2.73 Tiny
> 100 , ≤ 400 ≤ 0.0193 4859 61.09 63.82
> 400 , ≤ 900 ≤ 0.0434 2561 32.20 96.01
> 900 , ≤ 1600 ≤ 0.0772 173 2.18 98.19 Small
> 1600 , ≤ 2500 ≤ 0.12056 61 0.77 98.96
> 2500 , ≤ 10000 ≤ 0.48225 71 0.89 99.85 Middle
> 10000 , ≤ 40000 ≤ 1.92901 12 0.15 100.00 Large
> 40000 > 1.92901 0 0.00 100.00
Table 6. Distribution of uavs pixels in the Drone dataset.
Table 6. Distribution of uavs pixels in the Drone dataset.
Dataset Criterion(%) Object number Proportion (%) Scope of application
Drone ≤ 0.12 371 3.72 Small
≤ 0.48 5961 59.76 Middle
> 0.48 3643 36.52 Large
Table 7. Distribution of target sizes in the Flying dataset.
Table 7. Distribution of target sizes in the Flying dataset.
Dataset Criterion(%) Object number Proportion (%) Scope of application
Flying ≤ 0.12 4299 17.36 Small
≤ 0.48 4421 17.85 Middle
> 0.48 16049 64.79 Large
Table 8. Ablation experiments on the TIB dataset.
Table 8. Ablation experiments on the TIB dataset.
Improve MSSA RFDA UIOU mAP50 mAP50-95 FPS
1 YOLOv12 82.7 35.7 131.5
2 ✔ 84.0 36.1 73.5
3 ✔ 83.9 36.4 84.5
4 ✔ 83.8 36.7 166.0
5 ✔ ✔ 85.7 37.4 89.5
6 ✔ ✔ 85.2 37.0 83.5
7 ✔ ✔ ✔ 87.4 38.9 87.5
Table 9. Ablation experiments on the UB-SOD dataset.
Table 9. Ablation experiments on the UB-SOD dataset.
Improve MSSA RFDA UIOU mAP50 mAP50-95 FPS
1 YOLOv12 78.7 40.1 73.5
2 ✔ 79.2 40.8 85.0
3 ✔ 79.8 40.9 83.5
4 ✔ 79.1 40.5 73.5
5 ✔ ✔ 80.7 41.2 89.5
6 ✔ ✔ 80.7 41.0 83.5
7 ✔ ✔ ✔ 81.4 41.3 69.5
Table 10. Ablation experiments on the Drone dataset.
Table 10. Ablation experiments on the Drone dataset.
Improve MSSA RFDA UIOU mAP50 mAP50-95 FPS
1 YOLOv12 97.6 78.0 96.0
2 ✔ 97.9 78.1 53.0
3 ✔ 97.8 78.0 23.0
4 ✔ 97.6 78.2 96.0
5 ✔ ✔ 98.0 77.2 55.0
6 ✔ ✔ 97.8 77.8 47.0
7 ✔ ✔ ✔ 98.1 78.0 55.0
Table 11. Ablation experiments on the Flying dataset.
Table 11. Ablation experiments on the Flying dataset.
Improve MSSA RFDA UIOU mAP50 mAP50-95 FPS
1 YOLOv12 59.7 50.6 121.5
2 ✔ 60.7 51.7 56.5
3 ✔ 60.0 51.9 58.5
4 ✔ 62.9 52.6 112
5 ✔ ✔ 63.5 53.2 67.5
6 ✔ ✔ 64.7 54.2 59.0
7 ✔ ✔ ✔ 65.6 55.6 67.5
Table 12. Distribution of target sizes in the UB-SOD dataset.
Table 12. Distribution of target sizes in the UB-SOD dataset.
Methods mAP50 mAP50-95 FPS
Faster Rcnn-MobileNet 67.5 \ 8.0
Casacade Rcnn-MobileNet 78.0 \ 6.1
EXTD-MobileFaceNet 85.1 \ 3.6
Yolov2 67.1 \ 33.2
Yolov3 84.9 \ 15.2
Yolov4 86.0 \ 52.6
Yolov4s-Tiny 26.7 \ 45.5
Yolov5 86.2 38.7 128.2
Yolov6n 86.5 \ 76.9
Yolov7-Tiny 85.8 \ 125.0
Yolov8 [38] 80.3 31.7 135.2
Yolov10 [39] 72.6 29.3 78.2
Yolov11 [40] 79.7 31.2 90.9
Yolov12 [41] 82.7 35.7 65.8
Yolo26 [42] 83.2 34.6 82.0
Ours 87.4 38.9 87.5
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.