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

Fig 1.

Roadmap for retail product object detection using lightweight methods.

More »

Fig 1 Expand

Fig 2.

Overall structure diagram of YOLOv8n-improved model for the retail product object detection.

More »

Fig 2 Expand

Fig 3.

Structure diagram of DWConv.

More »

Fig 3 Expand

Fig 4.

Network structure of CBAM module.

More »

Fig 4 Expand

Fig 5.

Structure diagram of CSPHet bottleneck and CSPHet-CBAM.

More »

Fig 5 Expand

Fig 6.

Structure diagram of ADown.

It is used in YOLOv8n backbone to replace Conv.

More »

Fig 6 Expand

Fig 7.

Schematic of channel pruning.

More »

Fig 7 Expand

Fig 8.

The distribution of product categories and their quantities in the Locount dataset.

More »

Fig 8 Expand

Table 1.

Experimental configuration.

More »

Table 1 Expand

Fig 9.

The training curves of metrics in YOLOv8n-improved model.

More »

Fig 9 Expand

Table 2.

Ablation experiment results on Locount dataset.(“D” denotes DWConv, “CC” denotes CSPHet-CBAM, and “A” denotes ADown).

More »

Table 2 Expand

Table 3.

Model parameters and performance metrics under different channel pruning rates.

More »

Table 3 Expand

Fig 10.

Changes in the number of channels in the YOLOv8n-improved model at a pruning rate of 0.5.

More »

Fig 10 Expand

Fig 11.

The distribution of Gamma coefficients () of the BN layers before YOLOv8n-improved model’s sparse training.

More »

Fig 11 Expand

Fig 12.

Precision-Recall curve of LSR-YOLO model.

More »

Fig 12 Expand

Table 4.

Performance comparison of the improved model before and after pruning.

More »

Table 4 Expand

Table 5.

Comparison of different object detection algorithms on Locount dataset.

More »

Table 5 Expand

Fig 13.

Comparison of heatmaps on retail products for different object detection models.

More »

Fig 13 Expand

Table 6.

Ablation experiment results on COCO2017 dataset.(“D” denotes DWConv, “CC” denotes CSPHet-CBAM, and “A” denotes ADown).

More »

Table 6 Expand

Table 7.

Comparison of detection performance between LSR-YOLO and other models on COCO2017 dataset.

More »

Table 7 Expand

Fig 14.

Comparison of heatmaps for different object detection models on COCO scenes.

More »

Fig 14 Expand

Table 8.

The 5-fold cross validation result of LSR-YOLO on Locount dataset.

More »

Table 8 Expand

Fig 15.

Examples of negative samples for product detection with multiple object occlusions.

More »

Fig 15 Expand

Fig 16.

Examples of negative samples for product detection under low-light conditions.

More »

Fig 16 Expand