Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig1.

The network framework of the proposed model.

represents the foggy image, is the dehazed image of , is the clear image corresponding to , and is the final object detection result.

More »

Fig1 Expand

Fig 2.

The patchwise contrastive learning method based on PatchNCE loss.

More »

Fig 2 Expand

Fig 3.

The improved network diagram of YOLOv9s.

More »

Fig 3 Expand

Fig 4.

The efficient multi-scale attention.

More »

Fig 4 Expand

Fig 5.

IoU measures the overlap between predicted bounding boxes and ground truth boxes.

It is defined as the intersection area divided by the union area of the two boxes. WIoU implements a dynamic, non-monotonic focusing mechanism by IoU.

More »

Fig 5 Expand

Table 1.

Results on the hazed COCO2017 dataset.

More »

Table 1 Expand

Table 2.

Results on the hazed RTTS dataset.

More »

Table 2 Expand

Fig 6.

The confusion matrix of our model in the hazed COCO2017 dataset.

More »

Fig 6 Expand

Table 3.

Model performance.

More »

Table 3 Expand

Table 4.

Image dehazing.

More »

Table 4 Expand

Table 5.

Influence of PatchLoss.

More »

Table 5 Expand

Table 6.

Comparison among different attention mechanisms.

More »

Table 6 Expand

Table 7.

Influence of EMA attention.

More »

Table 7 Expand

Table 8.

Influence of Wise-IoU.

More »

Table 8 Expand

Fig 7.

Training results under different parameters.

More »

Fig 7 Expand