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

Table 1.

Summary of different methods from the latest literature.

More »

Table 1 Expand

Fig 1.

YOLOv5 network architecture.

More »

Fig 1 Expand

Fig 2.

CA attention mechanism.

More »

Fig 2 Expand

Fig 3.

CBAM attention mechanism.

More »

Fig 3 Expand

Fig 4.

ECA attention mechanism.

More »

Fig 4 Expand

Fig 5.

SE attention mechanism.

More »

Fig 5 Expand

Fig 6.

Ghost module network structure.

More »

Fig 6 Expand

Fig 7.

FPN, PAN, and BiFPN structures.

More »

Fig 7 Expand

Fig 8.

Schematic diagram of EIOU-loss.

More »

Fig 8 Expand

Fig 9.

Image labeling.

(a) belt, (b) helmet and trunk, (c) incorrect helmet and vest, and (d) smoking and head.

More »

Fig 9 Expand

Table 2.

Human Unsafe Behavior Dataset Distribution.

More »

Table 2 Expand

Fig 10.

Data enhancement.

More »

Fig 10 Expand

Table 3.

Insecure state data set distribution of objects.

More »

Table 3 Expand

Fig 11.

Improved attention module.

More »

Fig 11 Expand

Table 4.

Experimental results of the first improvement of the attention mechanism.

More »

Table 4 Expand

Table 5.

Experimental results of the second improvement of the attention mechanism.

More »

Table 5 Expand

Table 6.

Experimental results of the third improvement of attention mechanisms.

More »

Table 6 Expand

Fig 12.

Conv module replaced with GhostConv module.

More »

Fig 12 Expand

Fig 13.

C3 module replaced with C3Ghost module.

More »

Fig 13 Expand

Fig 14.

Mixing the first two improvements.

More »

Fig 14 Expand

Table 7.

Ghost improvement experiment results.

More »

Table 7 Expand

Table 8.

Results of the first group of experiments.

More »

Table 8 Expand

Table 9.

Experimental results of the second group.

More »

Table 9 Expand

Table 10.

Results of the third group of experiments.

More »

Table 10 Expand

Fig 15.

The way the first group of experiments was improved.

More »

Fig 15 Expand

Fig 16.

The way the second group of experiments was improved.

More »

Fig 16 Expand

Fig 17.

The way the third group of experiments was improved.

More »

Fig 17 Expand

Fig 18.

BiFPN (concat) improvement approach.

More »

Fig 18 Expand

Table 11.

BiFPN improvement experimental results.

More »

Table 11 Expand

Table 12.

Experimental results of loss function improvement.

More »

Table 12 Expand

Table 13.

Results of NMS improvement experiments.

More »

Table 13 Expand

Table 14.

Results of ablation experiments.

More »

Table 14 Expand

Fig 19.

YOLO-CGBSE network architecture.

More »

Fig 19 Expand

Fig 20.

P curve, R curve, PR curve and F1 curve.

More »

Fig 20 Expand

Table 15.

Model comparison results.

More »

Table 15 Expand

Fig 21.

Comparison of detection results.

The first column of images is the image to be detected, the second column of images is the original YOLOv5n detection effect image, the third column of images is the YOLO-CGBSE detection effect image, Fig (a) is the UAV viewpoint image, Fig (b) is the high-resolution image, Fig (c) is the indoor operation image, Fig (d) is the high altitude operation image, and Fig (e) is the occluded high altitude operation image.

More »

Fig 21 Expand

Table 16.

Object unsafe state model performance.

More »

Table 16 Expand

Fig 22.

Detection effect of unsafe state of object.

The first column of pictures is the picture to be detected, the second column of pictures is the original YOLOv5n detection effect, the third column is the YOLO-M detection effect, (a) is the picture of the safety net, (b) is the picture of the scaffolding perforation.

More »

Fig 22 Expand

Fig 23.

System framework diagram.

More »

Fig 23 Expand

Fig 24.

Detection system display.

More »

Fig 24 Expand

Fig 25.

Identify the system’s detection results.

More »

Fig 25 Expand