Search Results (8,415)

To address the insufficient accuracy of short-term load forecasting caused by the strong randomness of distributed energy output, variable electricity consumption patterns, and complex meteorological factors, this study proposes a load forecasting method that integrates K-means scene clustering and a Transformer–BiGRU–Attention (CTBA) hybrid deep learning architecture. Different from conventional Transformer–BiGRU hybrid forecasters that train a single global predictor across all operating conditions, the proposed CTBA framework first partitions daily load curves into representative scenes and then routes each sample to a scene-specific Transformer–BiGRU–Attention predictor, thereby reducing distributional heterogeneity before temporal modeling. First, the K-means algorithm is used to perform scene clustering on historical daily load curves, and the optimal number of clusters is selected according to the silhouette coefficient and downstream prediction performance. Subsequently, the CTBA model is trained separately for each clustering subset. The Transformer encoder captures the long-range global dependencies of load sequences through the self-attention mechanism, the BiGRU module extracts local bidirectional temporal fluctuation features, and the Attention mechanism further focuses on key time nodes such as morning and evening peaks while fusing multi-source data including historical load, day-ahead electricity price, and multi-dimensional meteorological factors. Experimental results based on the German ENTSO-E power dataset show that the coefficient of determination R² of the proposed model reaches 0.9893, with MAE, RMSE, and MAPE as low as 0.0141, 0.0187, and 3.92%, respectively, which are significantly improved compared to benchmark models such as SVR, LSTM, CNN, and TCN-BiGRU. Ablation experiments further demonstrate that removing the clustering, Transformer, BiGRU, or attention layer will degrade performance, thus verifying the effectiveness and superiority of the method in short-term load forecasting and providing an accurate solution for the short-term load forecasting of power systems. Full article

With increasingly stringent restrictions on SF₆ greenhouse gas emissions, C₄F₇N-based gas mixtures have attracted considerable attention as promising alternatives for high-voltage circuit breakers; however, their relatively weaker arc-quenching capability poses significant challenges for interruption chamber design at high voltage levels. In this study, a 3.5% C₄F₇N/83.5% CO₂/13% O₂ gas mixture was used as the arc-extinguishing medium in a 550 kV environmentally friendly gas circuit breaker. Based on a magnetohydrodynamic (MHD) model considering PTFE nozzle ablation effects, systematic optimization studies were conducted on key structural parameters of the puffer-type interruption chamber, including the exhaust hole diameter, nozzle throat diameter and length, arcing contact diameter, and downstream expansion angle. Simulations under arcing times of 9.9 ms and 11.4 ms were performed to evaluate chamber pressure, axial temperature, extinction peak voltage, and post-arc conductance characteristics. The results indicate that extending the nozzle throat straight section to 70 mm, enlarging the exhaust hole, and increasing the moving contact radius can effectively enhance pressure buildup, reduce arc-core temperature, and improve dielectric recovery capability. Under the 11.4 ms arcing condition, the optimized structure achieved an extinction peak voltage of 6972.4 V and a G200 value of 0.731 mS, demonstrating substantially improved interruption performance. These findings reveal the synergistic relationship between arcing time and structural parameters and provide theoretical guidance for the engineering design of environmentally friendly high-voltage gas circuit breakers. Full article

This study presents a reproducible evaluation framework for hybrid quantum-classical neural networks (HQCNNs) in healthcare classification, rather than a new architecture. We assess a four-qubit HQCNN combining a compact classical encoder, a two-layer parameterized quantum circuit (PQC), and a classical readout (441 trainable parameters) against carefully tuned classical baselines on the Wisconsin Diagnostic Breast Cancer (WDBC) dataset under identical five-fold cross-validation. The work is framed as a single-dataset proof-of-concept: the contribution is a documented, shared-fold evaluation protocol with a parameter-matched classical control and a quantified epistemic-informativeness analysis, not a demonstration of general quantum advantage. The HQCNN reached $96.49\pm 1.96\%$ accuracy and $99.44\pm 0.60\%$ ROC-AUC. A parameter-matched classical multilayer perceptron (441 parameters) reached $95.08\pm 1.81\%$ accuracy; the HQCNN’s $+1.41$ percentage-point edge at equal capacity was not statistically significant (paired t, $p=0.056$). Across five shared folds, no HQCNN-versus-classical accuracy difference survived Holm–Bonferroni correction (all adjusted $p\ge 0.625$), so we report the HQCNN as competitive with, not superior to, strong tuned classical baselines. A multi-split depth ablation showed that circuit depth $L\in \{1,2,3\}$ had no statistically detectable effect on accuracy ($L=2$ vs. $L=3$: Wilcoxon $p=1.00$); we therefore adopt two variational layers as a practical default rather than an optimum. Under a low-noise simulator (depolarising and amplitude-damping channels, $p=0.01$), accuracy was $96.49\%$, indicating robustness only at modest uniform error rates; realistic hardware noise is higher. We additionally apply Bayesian surprise as an epistemic-informativeness heuristic—not a formal generative model—to rank which findings are most worth building on. The framework offers a reproducible, documented evaluation procedure that can support cumulative comparison of hybrid quantum-classical models in healthcare. Full article

Background/Objectives: With the increasing incidence of drug counterfeiting and the emergence of personalized medicine, the need for unique marking of solid dosage forms, e.g., tablets, has attracted considerable interest in the current research and development landscape. Besides traditional printing methods, laser marking offers several advantages, as it eliminates the need for organic solvents and enables the generation of precise patterns. However, laser exposure may raise safety concerns regarding the stability of photosensitive drugs in the irradiated dosage forms. Therefore, the aim of the present study was to test the photoprotective effect of titanium dioxide (TiO₂) and its various alternatives, e.g., talc, calcium carbonate (CaCO₃), zinc oxide (ZnO), and black iron oxide (Fe₃O₄), alongside a ready-to-use reference formulation, Opadry^® Brown, which contains TiO₂ (titanium-containing, TC) on nifedipine, a light-sensitive model drug. Methods: Laser marking or short-term laser ablation at different wavelengths (193 nm, 248 nm, 532 nm, and 781 nm) was applied to different coating formulations. As a positive control, prolonged exposure to daylight was applied. The properties and photostability of these formulations were evaluated using several analytical methods (i.e., surface profilometry, Raman spectroscopy, and high-performance liquid chromatography (HPLC)). Results: The TiO₂, ZnO, Fe₃O₄, and Opadry^® TC Brown coatings maintained their color during the long-term study under all conditions. Furthermore, the prepared formulations exhibited different ablation depths and morphological changes depending on the coating and laser type. HPLC measurements confirmed significant differences in the protective ability of various pigments against sunlight and different types of lasers. Nevertheless, the obtained Raman spectra were not in complete agreement with HPLC results, which can be attributed to spectral overlap between key nifedipine degradation markers and excipient signals in the tablet core. Conclusions: Overall, laser treatment of tablets containing photosensitive drugs may induce API decomposition; however, this effect can be minimized or avoided by careful selection of the appropriate combination of laser type and photoprotective pigment. Under the applied experimental conditions, Ti:Sa laser treatment was associated with the lowest degree of nifedipine degradation among all formulations, while ZnO-containing coatings demonstrated the most consistent photoprotective performance against the majority of the tested laser types, while Fe₃O₄-containing coatings provided superior protection during prolonged sunlight exposure and Nd:YAG laser irradiation. Full article

The Nuri deposit is currently the only Cu-W-Mo deposit in the Gangdese metallogenic belt, Tibet, China, that contains large-scale tonnages for both Cu and WO₃ resources, accompanied by a medium-scale Mo resources. Previous studies have suggested the potential presence of concealed porphyry-type orebodies at depth, yet effective exploration tools for verifying this hypothesis remain lacking. In this study, microscopic identification, electron probe microanalysis (EPMA), and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) were integrated to investigate the mineral chemical characteristics of chlorite from the Nuri deposit. The aim was to evaluate the effectiveness of chlorite geochemistry as an exploration vector for predicting deep concealed porphyry orebodies and to establish corresponding exploration indicators. Chlorite in the deposit can be genetically classified into metasomatic (Chl-I) and hydrothermal (Chl-II) types. Both types are Mg-rich varieties, indicating formation under conditions of low oxygen fugacity and low pH. With decreasing vertical distance to the orebody and toward the southeast direction of the exploration section, the contents of Ti (10–950 ppm) and V (50–820 ppm), as well as the Ti/Sr, Ti/Mn, Ti/Li, and V/Li ratios, progressively increase. In contrast, the concentrations of Li (36–390 ppm), Mn (1270–6730 ppm), Sr (1–510 ppm), and Zn (110–1100 ppm) systematically decrease. These systematic compositional variations demonstrate that chlorite geochemistry is an effective exploration tool in the Nuri mining area and suggest the presence of a concealed mineralization center or porphyry orebody beneath the interval from ZK4501 to ZK4502. Full article

Purpose-aware AI systems are increasingly deployed in safety-critical, multi-agent, and human-facing environments, where they must transform heterogeneous data into timely, explainable, and goal-aligned decisions under uncertainty. Existing architectures often couple perception, reasoning, communication, and security only at the pipeline level. This creates a research gap in unified semantic transformation, purpose-oriented judgment, bounded imperfection handling, and semantic self-protection. To address this gap, this paper proposes a DIKWP+BUG semantic–cognitive reference architecture for artificial-consciousness-oriented computing, without claiming definitive artificial consciousness. The architecture represents cognition through the Data–Information–Knowledge–Wisdom–Purpose (DIKWP) model and uses BUG theory to model bounded approximation, incomplete evidence, and confidence miscalibration in cross-dimensional reasoning. The model is mapped to an Artificial Consciousness Processing Unit (ACPU) reference substrate, an Artificial Consciousness Operating System (ACOS), a DIKWP semantic communication subsystem, and a concept–semantic fused security subsystem. The components are implemented through runtime emulation and evaluated in smart-city governance, autonomous-driving, and medical-triage simulations. Compared with selected baselines, the prototype increased cognitive throughput from $4.5\mathrm{k}$ to $7.8\mathrm{k}$ logged events, reduced perception–action latency from $340\mathrm{ms}$ to $120\mathrm{ms}$, reduced CPU utilization from $95\%$ to $68\%$, lowered smart-city congestion duration by $30\%$, improved emergency response time by approximately $40\%$, achieved 0 collisions versus approximately $2/10$ baseline IoV runs, and improved medical-triage accuracy from $85\%$ to $92\%$. These online-runtime results provide initial feasibility evidence under controlled simulation conditions; they do not include offline model-preparation costs and therefore should not be interpreted as end-to-end lifecycle speedups. Matched-compute ablation, statistical benchmarking, hardware prototyping, and real-world validation remain future work. Full article

Driving behavior recognition plays a crucial role in intelligent driving systems and road traffic safety. Due to insufficient driving experience and limited ability to allocate visual attention, novice drivers are considered a high-risk group for traffic accidents. Existing approaches primarily focus on experienced drivers and rely on single-modal eye-tracking data, making it difficult to model spatial attention distributions and long-term temporal dependencies simultaneously. Moreover, these methods are often affected by modality asynchrony during multimodal fusion, further limiting performance gains. To address these challenges, this study proposes a novice driver behavior recognition method based on bimodal eye-tracking features and a gated cross-modal attention fusion (GCMAF) mechanism. The model adopts a spatial–temporal dual-branch architecture. The spatial branch employs ResNet34 to extract eye-tracking heatmap features to represent the visual attention distribution. In contrast, the temporal branch integrates a 1D-CNN with the Mamba model to capture local dynamic patterns and long-range temporal dependencies. In the fusion stage, the GCMAF module is introduced to enhance cross-modal interactions, and a gating mechanism is further used to adaptively adjust modality weights, thereby mitigating the adverse effects of modality asynchrony. To validate the effectiveness and generalization ability of the proposed method, repeated experiments and five-fold cross-validation are conducted. The results demonstrate that the model achieves an average classification accuracy of 93.86% across four driving behavior categories, with standard deviations below 0.3%. Compared with baseline methods, paired t-test results show that the performance improvement is statistically significant (p < 0.01). Ablation studies further confirm the independent contribution of each component. Overall, the proposed method outperforms existing approaches in terms of accuracy and stability, providing effective support for driving behavior assessment and proactive safety warning systems. Full article

Fringe projection profilometry (FPP) is widely used for 3D reconstruction, but conventional single-view FPP systems suffer from inherent occlusions and shadow regions, leading to incomplete surface recovery. In this study, we propose CVA-Net, an end-to-end deep learning framework with cross-view attention (CVA) that directly reconstructs dense depth maps from multi-view fringe patterns. CVA-Net simultaneously processes four fringe images acquired from orthogonal projection directions and leverages a CVA module to explicitly model inter-view dependencies, enabling adaptive fusion of complementary information. A 3D U-Net backbone with attention gates, atrous spatial pyramid pooling (ASPP), and an auxiliary parameter estimation branch further enhances reconstruction accuracy and structural consistency via multitask learning. To support Sim2Real network training, we build a Blender-based digital twin of a multi-view FPP system and generate a large-scale synthetic dataset with perfect ground truth. Extensive experiments on both synthetic and real-world objects demonstrate that CVA-Net significantly outperforms state-of-the-art single-view methods. With a symmetric four-view configuration and fringe period of 8, CVA-Net achieves an MAE of 0.0359 mm, an MSE of 0.0379 mm² and an RMSE of 0.1947 mm, reducing the MAE, MSE, and RMSE by 32.8%, 54.1%, and 32.2%, respectively, compared to the best single-view competitor. Ablation studies validate the contribution of each architectural component, while real-system experiments demonstrate the feasibility of transferring a network trained purely on synthetic data to practical FPP measurements without domain adaptation. Although further improvements are required to enhance reconstruction accuracy under real imaging conditions, the proposed framework provides an effective initial step toward bridging the gap between digital-twin-based training and real-world multi-view FPP applications. CVA-Net provides a robust, occlusion-aware solution for multi-view FPP reconstruction. Full article

The large-scale integration of distributed photovoltaics (DPVs) into the distribution network exacerbates voltage fluctuations and substantially increases network losses. To improve the voltage quality and economic efficiency of distribution networks, a Volt/Var optimization (VVO) model is established. Coordinating multiple heterogeneous devices, the model aims to minimize the total voltage deviation, the total network losses, and the regulation cost of discrete equipment simultaneously. Considering multi-constraint coupling characteristics, a quantitative method is proposed to evaluate the reactive power regulation potential of DPVs under intricate operating conditions. Then, the multi-strategy integrated rime optimization algorithm (MSIRIME) is utilized for the model solution. Fuch chaotic mapping generates uniformly distributed and ergodic initial populations. A dual-branch search mechanism combining the snow ablation optimizer with the rime optimization significantly enhances global exploration capabilities. The guided learning strategy balances exploration and exploitation for high-dimensional VVO, preventing local optima. Case tests on a modified IEEE 33-bus system demonstrate that the proposed model exhibits excellent effectiveness and robustness. Moreover, MSIRIME exhibits better optimization performance than some classic and recently proposed strategies, reducing the average network losses and voltage deviation over 30 independent runs by at least 5.87% and 52.22%, respectively, relative to those of the compared methods. Full article

Water quality monitoring is a fundamental element of sustainable aquaculture management, as changes in parameters of physicochemical and biological properties directly affect the health, growth performance, and productivity of the aquaculture systems. Although traditional machine learning (ML) methods have demonstrated effectiveness in water quality classification, their performance often depends on large amounts of labeled data, which can be challenging and expensive to collect in real-world aquaculture environments. This study explores a few-shot learning (FSL) framework for data-efficient water quality classification under limited supervision to address this limitation. Several FSL models, including prototypical networks (ProtoNet), Siamese Networks, and Matching Networks were developed and evaluated in a comparative experimental framework against the traditional machine learning classifiers logistic regression, random forest, support vector machine and extreme gradient boosting. Low-data learning scenarios were simulated using a structured episodic evaluation approach. Experimental results demonstrate FSL techniques outperform traditional machine learning methods across all evaluated scenarios. Among the tested methods, ProtoNet achieved the highest performance, attaining an accuracy of 94.46% and an ROC-AUC score of 98.65%, indicating superior discriminative capability and robustness. Siamese Networks also demonstrated competitive performance under highly constrained data conditions. Furthermore, latent-space visualization, confusion matrix analysis, paired t-test statistical analysis, and ablation studies confirmed that episodic meta-learning enables the learning of highly discriminative latent representations with strong generalization capability under limited labeled data conditions. The findings highlight that FSL provides a robust and scalable framework for intelligent water quality classification in aquaculture systems, particularly in scenarios where labeled data are scarce, offering significant potential for sustainable aquaculture monitoring applications. Full article

Remote sensing object detection remains challenging because objects often appear with large scale variation, dense spatial layouts, and strong interference from complex geographical backgrounds. To address these coupled difficulties, we propose EDM-Net, an end-to-end multi-scale detector that organizes feature processing into three coordinated stages: adaptive extraction, intra-scale interaction, and cross-scale fusion. First, an efficient sparse mixture-of-experts (ES-MoE) module is embedded in the backbone to allocate scale-specific convolutional experts according to scene-level feature responses, providing a more adaptive feature basis than a single static extraction path. Second, a dynamic mixing intra-scale feature interaction (DMIFI) module is introduced into the Transformer encoder. This module combines global self-attention with dynamic spatial mixing, thereby preserving long-range context while reintroducing local two-dimensional inductive bias for dense and small objects. Third, a multi-scale synergistic attention fusion (MSAF) module aligns adjacent feature levels through parallel local and global attention branches and structural re-parameterization, reducing semantic dilution during feature aggregation. Comprehensive experiments on three large-scale remote sensing benchmark datasets, DIOR, NWPU VHR-10, and RSOD, demonstrate that EDM-Net consistently improves over the re-trained RT-DETR-R18 baseline under the same experimental protocol, attaining mAP50 scores of 83.7%, 95.6%, and 95.8% respectively. Additional ablation and scale-specific analyses indicate that the three modules contribute complementary gains, especially for small and densely distributed objects. These results suggest that coordinated extraction, interaction, and fusion can improve remote sensing object detection under complex scale and background conditions. Full article

Aircraft taxiing speed control along predefined airport surface routes is a constrained single-aircraft longitudinal control problem involving heterogeneous route geometry, action smoothness, and terminal velocity feasibility. Existing learning-based taxiing controllers commonly use a unified policy for the whole route, which may be insufficient for handling straight-segment propulsion, curved-segment speed regulation, and action discontinuities near straight–curve transitions. This paper proposes SegCoord-Taxi, a geometry-aware segmented deep reinforcement learning framework for taxiing speed control. The route is decomposed into straight segments, curved segments, and transition boundary zones. A Straight-Segment Policy (SSP) and a Curved-Segment Policy (CSP) generate geometry-dependent base acceleration commands, a Switch Residual Adapter (SRA) provides local residual correction near transition regions, and a Route-Level Feasibility Projection (RFP) maps the coordinated action into an executable acceleration satisfying route-level feasibility constraints. Experiments on departure taxiing routes at Chengdu Tianfu International Airport (ZUTF) included baseline comparison, ablation analysis, projection diagnostics, sensitivity analysis, and a trajectory-level case study. On the evaluated ZUTF case-study routes, SegCoord-Taxi achieves the lowest final velocity on the test set, 0.336 ± 0.017 m/s, compared with 0.732 ± 0.061 m/s for the unified Proximal Policy Optimization (PPO) controller and 0.586 m/s for the curvature-aware constrained optimizer. The complete framework also reduces switch action jump from 1.022 ± 0.017 m/s² to 0.429 ± 0.004 m/s² in the ablation study. These results indicate improved terminal feasibility and transition-region smoothness in the evaluated single-airport case-study setting under an explicit efficiency–smoothness–feasibility trade-off. Future work will extend the framework to multi-aircraft and multi-airport settings under operational uncertainty. Full article

Background/Objectives: Deep learning models for digital mammography sensor data are increasingly deployed across hospitals using different X-ray detector technologies and patient populations. Whether models trained on one sensor platform and population maintain accuracy when transferred to another has not been tested for the latest generation of mammography-specific foundation models under one controlled protocol. Methods: We fine-tuned five backbone architectures (ResNet-50, DINOv2-B14, Rad-DINO, Mammo-CLIP B5, and Mammo-FM) on CBIS-DDSM (film-digitized, USA, n = 714 validation) with three seeds, ablated a density-aware focal loss across three auxiliary weights, and evaluated transfer to three external sensor cohorts: CMMD (full-field digital, China, n = 1032), DMID (mixed digital, India, n = 509), and MIAS (film-digitized, UK, n = 322). Significance used paired DeLong z-tests with Benjamini–Hochberg FDR correction; temperature scaling tested post hoc recalibration at all transfer targets. Results: Within this single-source three-seed evaluation, ResNet-50 outperformed all four foundation models on CBIS-DDSM (AUC 0.867 vs. 0.847, 0.846, 0.813, and 0.703; all gaps p_adj < 0.05). The density-aware focal loss degraded both AUC and calibration at every weight tested. At transfer, every model lost 0.165 to 0.320 AUC points relative to in-distribution performance, with sensitivity at 95% specificity collapsing from 0.31 to 0.47 in-distribution to 0.11 to 0.22 across the three external targets. A per-seed Stouffer meta-analysis confirms that Mammo-CLIP B5 and Mammo-FM significantly outperformed ResNet-50 on DMID and Mammo-CLIP on CMMD, after BH-FDR; MIAS comparisons remained directional only. In the extremely dense subgroup (BI-RADS D4), Mammo-FM reached AUC 0.870 versus ResNet-50 at 0.842, a directional observation whose 95% CIs overlap heavily at the n = 140 sample size and which we do not interpret as a statistically supported advantage. Conclusions: In this single training-source, three-seed protocol, mammography-specific pretraining did not deliver the in-distribution AUC premium reported in the originating papers, and no architecture reached a level at which transfer deployment without local validation would be defensible. We frame these as observations specific to the present protocol rather than as broader conclusions about foundation models for mammography classification. The findings argue for sensor-stratified and population-stratified external validation and for local recalibration as practical prerequisites before clinical use. Code and weights are released under MIT license. Full article

Accurate and robust detection of leaf diseases is a key enabler for precision agriculture and large-scale crop health monitoring. Despite the strong generalization of modern one-stage detectors (e.g., YOLOv8), two domain-specific challenges remain: (i) weak or blurry lesion boundaries hinder precise localization, and (ii) low color contrast between diseased and healthy tissues forces models to rely on subtle texture patterns rather than salient shapes. To tackle these challenges, we reframe the core agricultural disease detection task as the identification of “asymmetric morphological anomalies” and propose a domain-tailored enhancement framework. First, we introduce an Edge Enhancement Module (EEM) that explicitly strengthens boundary-aware representations. Inspired by the natural symmetry of healthy leaves, our EEM is specifically designed to capture symmetry-breaking boundary discontinuities and localized asymmetric edges caused by disease lesions. Our method enhances edge and texture cues that are indicative of disease lesions, which often exhibit local asymmetries and boundary discontinuities. The EEM includes a Differential Normalized Pooling Block (DNPB) that highlights edge responses through discrepancies between max pooling and average pooling, which also models cross-group edge correlations. Second, the Lightweight Texture-Sensitive Feature Enhancement (LTSFE) mechanism amplifies texture-discriminative channels under low-contrast conditions by leveraging complementary global statistics and efficient channel mixing, all with negligible computational overhead. We evaluated our method on a self-constructed dataset of 106,434 images with 225,640 annotations covering diverse crops. Experiments show that the proposed method achieves state-of-the-art accuracy (81.54% mAP@0.5:0.95) while maintaining real-time inference (142 FPS), consistently outperforming strong baselines. Ablations confirm the effectiveness and complementarity of EEM and LTSFE, demonstrating that domain-specific architectural design, inspired by biological symmetry, can substantially improve agricultural vision systems. Full article

Unmanned Aerial Vehicle (UAV) image object detection holds significant application value in the low-altitude economy, traffic monitoring, intelligent agriculture, and disaster rescue. However, due to the top-down perspective, UAV images typically suffer from challenges such as small target scales, dense object distribution, severe occlusions, and complex backgrounds. These issues often limit the recall and localization accuracy of general-purpose detectors when they are directly applied to UAV small-object detection scenarios. To address these aforementioned challenges, this paper proposes an Adaptive Dynamic Aggregation YOLO network, termed ADA-YOLO. The novelty of ADA-YOLO lies in its highly efficient combinatorial design specifically tailored for UAV small object detection, while retaining the efficient backbone of YOLOv8, we systematically reconstruct the neck and detection head to improve accuracy. Specifically, a high-resolution P2 detection branch is incorporated to construct a P2–P5 multi-scale prediction structure. Furthermore, the lightweight DySample dynamic upsampling module is adopted to replace traditional upsampling methods, and an Adaptive Spatial Feature Fusion (ASFF) mechanism is introduced to alleviate semantic conflicts and noise interference during multi-scale feature fusion. This synergistic combination explicitly addresses multi-scale representation challenges and enhances small-object detection performance in complex scenes. Comparative experiments with the baseline YOLOv8n on the VisDrone2019 dataset demonstrate that ADA-YOLO achieves an improvement of 11.3% in mAP@0.5 and 8.2% in mAP@0.5:0.95. The improved model achieves these performance gains with a modest parameter increase and acceptable computational complexity. Finally, ablation experiments further validate the effectiveness of each individual module and their synergistic gains. Full article