Red–Green–Blue–Thermal (RGBT) object detection is critical for robust all-weather perception. However, deploying dual-stream networks on resource-constrained edge devices is severely hindered by insufficiently adaptive multimodal fusion, the loss of small-object features during downsampling, and substantial computational overhead. To address these challenges, we propose GEM-YOLO, a real-time and lightweight RGBT detector. Specifically, an Adaptive Multimodal Gated Fusion Mechanism (GFM) is designed to dynamically calibrate modality weights and suppress noise. Furthermore, Space-to-Depth (SPD) convolutions are integrated into the backbone to achieve lossless downsampling, preventing the feature collapse of small targets. Finally, a lightweight Ghost-Neck is constructed using Ghost modules and GSConv to eliminate computational redundancy. Extensive experiments on the Forward-Looking Infrared (FLIR) and Multi-Modal Multispectral Fusion Dataset (M3FD) datasets demonstrate the effectiveness of the proposed method. With only 7.58 Giga Floating-Point Operations (GFLOPs) and 3.44 million parameters (M), GEM-YOLO reduces the computational cost by 18.6% relative to the dual-stream YOLOv11n baseline. Concurrently, it achieves competitive mean Average Precision at IoU = 0.5 (mAP@50) scores of 82.8% and 69.0% on FLIR and M3FD, respectively, with more evident gains on small-target localization. In practice, GEM-YOLO maintains competitive detection performance while keeping computational overhead low, making it promising for real-time multispectral perception on resource-constrained edge platforms.

In security applications, visible-light pedestrian detectors are highly sensitive to changes in illumination and fail under low-light or nighttime conditions, while infrared sensors, though resilient to lighting, often produce blurred object boundaries that hinder precise localization. To address these complementary limitations, we propose a practical multimodal pipeline—Adaptive Energy–Gradient–Contrast (EGC) Fusion with AIFI-YOLOv12—that first fuses infrared and low-light visible images using per-pixel weights derived from local energy, gradient magnitude and contrast measures, then detects pedestrians with an improved YOLOv12 backbone. The detector integrates an AIFI attention module at high semantic levels, replaces selected modules with A2C2f blocks to enhance cross-channel feature aggregation, and preserves P3–P5 outputs to improve small-object localization. We evaluate the complete pipeline on the LLVIP dataset and report Precision, Recall, mAP@50, mAP@50–95, GFLOPs, FPS and detection time, comparing against YOLOv8, YOLOv10–YOLOv12 baselines (n and s scales). Quantitative and qualitative results show that the proposed fusion restores complementary thermal and visible details and that the AIFI-enhanced detector yields more robust nighttime pedestrian detection while maintaining a competitive computational profile suitable for real-world security deployments.

Visible–thermal object detection benefits from the complementary properties of RGB and thermal imagery, but repeated cross-modal fusion can increase model complexity, particularly in lightweight detectors. This paper proposes SDR-YOLO, a scale-selective detector designed to make better use of shallow spatial details without adding an extra prediction scale. A P2-guided Cross-modal Detail Enhancement module uses visible and thermal P2 features as auxiliary detail sources and injects the resulting residual into the P3 fused feature. P4 and P5 retain simple concatenation-based fusion, while a Lightweight Shared Convolutional Detection Head reduces redundant prediction parameters across scales. Experiments are conducted on DroneVehicle and M3FD. On DroneVehicle, SDR-YOLO achieves 80.4% mAP@50 and 56.7% mAP@50:95, improving both metrics by 1.0 percentage point over YOLO11s, with 14.06 M parameters and 32.51 GFLOPs; under workstation profiling conditions, it reaches 197.62 ± 1.57 FPS. On M3FD, the model obtains 82.5% mAP@50 and 55.6% mAP@50:95, corresponding to changes of 0.2 and −0.1 percentage points relative to YOLO11s. Since these differences are comparable to the run-to-run variation observed across three training seeds, they are treated as numerical trends rather than statistically established improvements. On the Jetson Orin Nano, TensorRT FP16 deployment achieves 45.78 ± 0.16 FPS, with a latency of 21.84 ± 0.08 ms and peak additional unified-memory usage of 590.4 ± 5.1 MiB, although YOLO11s-RGBT runs faster on this platform. Overall, the proposed design provides a clearer benefit in dense UAV-view scenes while maintaining comparable accuracy and slightly lower model complexity on M3FD, although its runtime efficiency remains hardware-dependent.

Detecting infrared moving targets in low-altitude remote sensing scenes remains challenging due to strong clutter, scale inconsistency, and residual interference. Because these factors are often coupled in complex scenes and cannot be handled effectively by a single operation, a three-stage progressive Decoupled Separation, Enhancement, and Purification (DSEP) framework is proposed. The method integrates edge-preserving background decoupling, scale-consistent spatial screening, and residual-response purification into a non-iterative feedforward pipeline. Experiments on six representative self-collected infrared sequences and six selected scenes from the public SIRST dataset suggest that DSEP produces relatively compact and spatially continuous target responses while suppressing background interference. On the self-collected dataset, the method can achieve SCRG and BSF values up to 10.61 and 7.38, respectively, with a processing time of 0.009–0.016 s per frame. Compared with representative spatial filtering, local contrast, and low-rank decomposition methods, DSEP shows a relatively favorable balance between detection performance and low-latency processing efficiency. Although the performance gain becomes smaller in some SIRST scenes, the proposed method still shows generally stable detection performance across the evaluated scenes.