Infrared and visible image fusion-based object detection is critical for robust environmental perception under adverse conditions, yet existing methods still suffer from insufficient modeling of modality discrepancies and limited adaptivity in their fusion mechanisms. This work proposes an asymmetric spatial–frequency fusion network, AsyFusionNet. The network adopts an asymmetric dual-branch backbone that extends the RGB branch to P5 while truncating the infrared branch at P4, thereby better aligning with the physical characteristics of the two modalities, enhancing feature complementarity, and enabling fine-grained modeling of modality differences. On top of this backbone, a local–global attention fusion (LGAF) module is introduced to model local and global attention in parallel and reorganize them through lightweight convolutions, achieving joint spatial–channel selective enhancement. Modality-specific feature enhancement is further realized via a hierarchical attention module (HAM) in the RGB branch, which employs dynamic kernel selection to emphasize multi-level texture details, and a fourier spatial spectral modulation (FS2M) module in the infrared branch, which more effectively captures global thermal radiation patterns. Extensive experiments on the $M^3$FD and VEDAI datasets demonstrate that AsyFusionNet attains $86.3\%$ and $54.1\%$${\mathrm{mAP}}_{50}$, respectively, surpassing the baseline by $8.8$ and $6.4$ points (approximately $11.4\%$ and $13.4\%$ relative gains) while maintaining real-time inference speed.

In unmanned aerial vehicle (UAV)-based small object detection, enhancing object perception under complex backgrounds remains a critical challenge for current detection models. Owing to the small scale, low pixel occupancy, and cluttered backgrounds of UAV small objects, their discriminative features are prone to attenuation in deep networks, which limits multi-scale feature fusion and consequently increases detection difficulty. To resolve these issues, this paper presents a Detail-Aware Multi-scale Fusion Network (DMF-Net) for UAV small object detection, consisting of two core modules: the Multi-Branch Detail-Enhanced Module (MBDE) and the Dual-Attention Fusion (DAF) Module. First, during feature extraction, a multi-branch detail enhancement module with serial convolutions in each branch and residual connections is introduced to strengthen high-frequency details and local textures while preserving semantic consistency. Second, at the feature fusion stage, a dual-attention-based feature fusion module with a symmetric interaction structure is designed to dynamically evaluate the significance of features at different scales via adaptive attention mechanisms, enabling symmetric cross-scale interaction and fine-grained feature complementarity. The experimental results obtained on the challenging VisDrone and TinyPerson datasets confirm that DMF-Net outperforms existing state-of-the-art detection methods in terms of the accuracy of small object detection, while maintaining high inference efficiency. On the TinyPerson dataset, DMF-Net improves AP by 1.4%, AP_50 by 4.1%, and AP_75 by 0.6% compared with YOLO11n, while maintaining 97.2 FPS. Furthermore, it shows promising performance in complex backgrounds and densely populated scenarios.

Object detection in synthetic aperture radar (SAR) imagery is challenged by speckle noise, background clutter, and large variations in target scale. We present ASCE-SAR, a YOLO11-based detector that coordinates adaptive shift-wise downsampling with frequency-spatially modulated multi-scale fusion. In the backbone, the Adaptive Downsampling with Shift-Wise Convolution (ADSC) block combines structure-preserving downsampling and enlarged receptive-field modeling. In the neck, Dynamic Shift-wise Feature Fusion (DSFF) uses content-aware resampling and frequency-spatial modulation to improve cross-scale feature alignment and weak-target representation. On the OGSOD-1.0 and HRSID test sets, ASCE-SAR achieves AP values of 59.9% and 67.9%, respectively, corresponding to gains of 7.1 and 1.8 percentage points over the YOLO11 Baseline under the same evaluation protocol. Under synthetic Gamma speckle perturbations on HRSID, the full model retains a 0.8–2.0 percentage-point mAP\(_{50{:}95}\) advantage over the Baseline across the clean condition and the evaluated noise levels. The full model contains 3.17 M parameters and requires 14.94 GFLOPS; its profiling results are reported with forward latency and peak memory to make the accuracy-cost trade-off explicit. These results indicate improved detection accuracy and noise-perturbed performance, accompanied by a measurable computational cost.