Training results under different parameters. حسب Yang Zhang (30734)

الموضوعات:

IoU measures the overlap between predicted bounding boxes and ground truth boxes. حسب Yang Zhang (30734)

الموضوعات:

The patchwise contrastive learning method based on PatchNCE loss. حسب Yang Zhang (30734)

الموضوعات:

The network framework of the proposed model. حسب Yang Zhang (30734)

الموضوعات:

The confusion matrix of our model in the hazed COCO2017 dataset. حسب Yang Zhang (30734)

الموضوعات:

The efficient multi-scale attention. حسب Yang Zhang (30734)

الموضوعات:

The improved network diagram of YOLOv9s. حسب Yang Zhang (30734)

الموضوعات:

Structure diagram of the CGLU module. حسب Qian Shen (89639)

الموضوعات: "…existing mainstream algorithms…"

SPPF-LSKA module structure. حسب Qian Shen (89639)

الموضوعات: "…existing mainstream algorithms…"

The improved CSW-YOLO model framework. حسب Qian Shen (89639)

الموضوعات: "…existing mainstream algorithms…"

Thermodynamic diagram visualization results after adding the LSKA attention mechanism. حسب Qian Shen (89639)

الموضوعات: "…existing mainstream algorithms…"

Visualization result of yolov8 heatmap. حسب Qian Shen (89639)

الموضوعات: "…existing mainstream algorithms…"

Block diagram of PConv and FasterNetBlock modules. حسب Qian Shen (89639)

الموضوعات: "…existing mainstream algorithms…"

Ablation study results. حسب Beigeng Zhao (21637197)

الموضوعات:

Comparison of the YOLO11 baseline and our optimized model at the “s” scale. حسب Beigeng Zhao (21637197)

الموضوعات:

Sample counts and area statistics for object categories in the SeaDronesSee dataset. حسب Beigeng Zhao (21637197)

الموضوعات:

Performance comparison between the proposed method and YOLO-based state-of-the-art methods on the SeaDronesSee dataset under varying model complexities. حسب Beigeng Zhao (21637197)

الموضوعات:

Comparison with YOLO-based state-of-the-art methods. حسب Beigeng Zhao (21637197)

الموضوعات:

Overall rankings of single-objective optimization algorithms. حسب Máté Mohácsi (20469514)

The computational complexity of each algorithm. حسب Emad S. Hassan (17775798)

"…By exploiting the channel’s frequency selectivity, the power allocation algorithms adjust the modulation type and power distribution for each sub-carrier dynamically. …"