Yolo-C:基于单阶段网络的X光图像违禁品检测 
下载： 1124次
Yolo-C: One-Stage Network for Prohibited Items Detection Within X-Ray Images
中国民航大学天津市智能信号与图像处理重点实验室, 天津 300300
图 & 表
图 1. Yolov3网络结构图
Fig. 1. Yolov3 network structure
下载图片 查看原文
图 2. 复合骨干网络结构图
Fig. 2. Composite backbone network structure
下载图片 查看原文
图 3. 特征增强模块
Fig. 3. Feature augment block
下载图片 查看原文
图 4. 单阶段双网络目标检测算法Yolo-C网络结构图
Fig. 4. Yolo-C network structure of one-stage dual-network object detection algorithm
下载图片 查看原文
图 5. GDXray数据集随机选取的X光图像
Fig. 5. Random X-ray images of GDXray dataset
下载图片 查看原文
图 6. SIXray数据集随机选取的X光图像
Fig. 6. Random X-ray images of SIXray dataset
下载图片 查看原文
图 7. 违禁物品样例展示
Fig. 7. Sample display of prohibited items
下载图片 查看原文
图 8. 网络训练过程对比
Fig. 8. Comparison graph of network training process
下载图片 查看原文
图 9. IoU定义及计算示意图
Fig. 9. IoU definition and calculation diagram
下载图片 查看原文
图 10. 实验3, 5和7的检测结果
Fig. 10. Detection results of experiments 3, 5 and 7
下载图片 查看原文
表 1Yolo-C参数
Table1. Parameters of Yolo-C
| Type | Layer | Filter number | Size | Output |
|---|
| DBL | Conv | 32 | 3×3 | 416×416 | | Res1 | Conv2 | 64 | 3×3, 1×1 | 208×208 | | — | Upsample | 32 | 3×3 | 416×416 | | Res1' | Conv2 | 64 | 3×3, 1×1 | 208×208 | | Res2 | Conv4 | 128 | 3×3, 1×1 | 104×104 | | — | Upsample | 64 | 3×3 | 208×208 | | Res2' | Conv4 | 128 | 3×3, 1×1 | 104×104 | | Res8 | Conv16 | 256 | 3×3, 1×1 | 52×52 | | — | Upsample | 128 | 3×3 | 104×104 | | Res8' | Conv16 | 256 | 3×3, 1×1 | 52×52 | | Res8 | Conv16 | 512 | 3×3, 1×1 | 26×26 | | — | Upsample | 256 | 3×3 | 52×52 | | Res8' | Conv16 | 512 | 3×3, 1×1 | 26×26 | | Res4 | Conv8 | 1024 | 3×3, 1×1 | 13×13 | | — | Upsample | 512 | 3×3 | 26×26 | | Res4' | Conv8 | 1024 | 3×3, 1×1 | 13×13 | | DBL | Conv | 30 | 3×3 | 13×13 | | Head | — | — | — | — | | FAB | Upsample | | 3×3, 1×1 | 26×26 | | DBL | Conv | 30 | 3×3 | 26×26 | | Head | — | — | — | — | | FAB | Upsample | | 3×3, 1×1 | 52×52 | | DBL | Conv | 30 | 3×3 | 52×52 | | Head | — | — | — | — |
|
查看原文
表 2不同网络复杂度分析
Table2. Analysis of different network complexity
| Model | Backbone | FAB | FLOPs | Params/106 | |
|---|
| Yolov3 | DarkNet-53 | | 32.77 | 61.55 | | | Ours | DarkNet-53 | √ | 35.58 | 64.65 | | | DarkNet-C | √ | 61.85 | 105.93 | |
|
查看原文
表 3在SIXray_OD数据集上对Yolo-C网络进行消融实验
Table3. Ablation experiments on the Yolo-C network based on SIXray_OD dataset
| No. | Model | Backbone | FAB | AP for gun /% | AP for knife /% | AP for pliers /% | AP for wrench /% | AP for scissor /% | mAP /% | Detection rate /(frame·s-1) |
|---|
| 1 | SSD | ResNet50 | | 85.69 | 72.44 | 51.21 | 60.62 | 41.99 | 62.39 | 56 | | 2 | FASF | ResNet101 | | 82.77 | 70.23 | 48.6 | 57.7 | 38.7 | 59.6 | 48 | | 3 | Yolov3 | DarkNet-53 | | 88.67 | 76.53 | 52.49 | 61.41 | 42.6 | 64.34 | 57 | | 4 | Faster-RCNN | ResNet101 | | 93.83 | 83.74 | 58.85 | 71.58 | 54.1 | 72.18 | 10 | | 5 | Ours | DarkNet-53 | √ | 90.1 | 79.57 | 55.67 | 64.81 | 51.00 | 68.23 | 55 | | 6 | DarkNet-C | | 91.6 | 82.1 | 58.7 | 69.7 | 54.4 | 71.10 | 42 | | 7 | DarkNet-C | √ | 93.14 | 83.12 | 60.18 | 73.82 | 58.14 | 73.68 | 40 |
|
查看原文
郭守向, 张良. Yolo-C:基于单阶段网络的X光图像违禁品检测[J]. 激光与光电子学进展, 2021, 58(8): 0810003. Shouxiang Guo, Liang Zhang. Yolo-C: One-Stage Network for Prohibited Items Detection Within X-Ray Images[J]. Laser & Optoelectronics Progress, 2021, 58(8): 0810003.
PDF全文
