YOLOv5改进（一）--轻量化YOLOv5s模型

1、前言

在边缘设备的场景下，目前的YOLOv5s，虽然能够快速实现目标检测，但是运行速度依旧稍慢点，本文在牺牲一点精度前提下，提高目标检测速度，即轻量化YOLOv5s模型，并部署到边缘设备上，可以在CPU上达到实时的检测效果，满足业务的性能需求。

2、轻量化模型结构：

轻量化思路：

1、改进锚框，将对应的锚框全部减半

2、将yolov5s的模型的channels通道数全部都减少一半。

3、训练时，输入图片大小为320，即从 640x640 变为 320x320

原始yolov5s.yaml模型结构

# YOLOv5 🚀 by Ultralytics, GPL-3.0 license

# Parameters
nc: 80  # number of classes
depth_multiple: 0.33  # model depth multiple
width_multiple: 0.50  # layer channel multiple
anchors:
  - [10,13, 16,30, 33,23]  # P3/8
  - [30,61, 62,45, 59,119]  # P4/16
  - [116,90, 156,198, 373,326]  # P5/32

# YOLOv5 v6.0 backbone
backbone:
  # [from, number, module, args]
  [[-1, 1, Conv, [64, 6, 2, 2]],  # 0-P1/2
   [-1, 1, Conv, [128, 3, 2]],  # 1-P2/4
   [-1, 3, C3, [128]],
   [-1, 1, Conv, [256, 3, 2]],  # 3-P3/8
   [-1, 6, C3, [256]],
   [-1, 1, Conv, [512, 3, 2]],  # 5-P4/16
   [-1, 9, C3, [512]],
   [-1, 1, Conv, [1024, 3, 2]],  # 7-P5/32
   [-1, 3, C3, [1024]],
   [-1, 1, SPPF, [1024, 5]],  # 9
  ]

# YOLOv5 v6.0 head
head:
  [[-1, 1, Conv, [512, 1, 1]],
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 6], 1, Concat, [1]],  # cat backbone P4
   [-1, 3, C3, [512, False]],  # 13

   [-1, 1, Conv, [256, 1, 1]],
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 4], 1, Concat, [1]],  # cat backbone P3
   [-1, 3, C3, [256, False]],  # 17 (P3/8-small)

   [-1, 1, Conv, [256, 3, 2]],
   [[-1, 14], 1, Concat, [1]],  # cat head P4
   [-1, 3, C3, [512, False]],  # 20 (P4/16-medium)

   [-1, 1, Conv, [512, 3, 2]],
   [[-1, 10], 1, Concat, [1]],  # cat head P5
   [-1, 3, C3, [1024, False]],  # 23 (P5/32-large)

   [[17, 20, 23], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5)
  ]

轻量化yolov5s-320.yaml结构

# YOLOv5 🚀 by Ultralytics, GPL-3.0 license

# Parameters
nc: 10  # number of classes
depth_multiple: 0.33  # model depth multiple
width_multiple: 0.50  # layer channel multiple
anchors:
  - [5,6,  8,15,  16,11]  # P3/8
  - [15,30,  31,22,  29,59]  # P4/16
  - [58,45,  78,99,  186,163]  # P5/32

# YOLOv5 v6.0 backbone
backbone:
  # [from, number, module, args]
  [[-1, 1, Conv, [32, 6, 2, 2]],  # 0-P1/2
   [-1, 1, Conv, [64, 3, 2]],  # 1-P2/4
   [-1, 3, C3, [64]],
   [-1, 1, Conv, [128, 3, 2]],  # 3-P3/8
   [-1, 6, C3, [128]],
   [-1, 1, Conv, [256, 3, 2]],  # 5-P4/16
   [-1, 9, C3, [256]],
   [-1, 1, Conv, [512, 3, 2]],  # 7-P5/32
   [-1, 3, C3, [512]],
   [-1, 1, SPPF, [512, 5]],  # 9
  ]

# YOLOv5 v6.0 head
head:
  [[-1, 1, Conv, [256, 1, 1]],
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 6], 1, Concat, [1]],  # cat backbone P4
   [-1, 3, C3, [256, False]],  # 13

   [-1, 1, Conv, [128, 1, 1]],
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 4], 1, Concat, [1]],  # cat backbone P3
   [-1, 3, C3, [128, False]],  # 17 (P3/8-small)

   [-1, 1, Conv, [128, 3, 2]],
   [[-1, 14], 1, Concat, [1]],  # cat head P4
   [-1, 3, C3, [256, False]],  # 20 (P4/16-medium)

   [-1, 1, Conv, [256, 3, 2]],
   [[-1, 10], 1, Concat, [1]],  # cat head P5
   [-1, 3, C3, [512, False]],  # 23 (P5/32-large)

   [[17, 20, 23], 1, Detect, [nc, anchors]],  # Detect(P3, P4, P5)
  ]

模型查看

# 查看模型结构可以运行
python models/yolo.py --cfg yolov5s-320.yaml

# 训练时候设置--img 320
python train.py --batch-size 16 --epochs 100 --img 320 --cfg models/yolov5s-320.yaml --data data/traffic.yaml --weights weights/yolov5s.pt --device 0 > myout.file 2>&1 &

3、模型对比

模型	input-size	params(M)	GFLOPs	precision	recall	mAP_0.5	mAP_0.5:0.95	模型大小
YOLOv5s	640×640	7.04	15.8	0.987	0.99	0.993	0.828	14.4MB
YOLOv5s-320	320x320	1.77	4.2	0.895	0.992	0.912	0.749	3.9MB

4、训练结果图

这是训练epoch的可视化图，可以看到mAP随着Epoch训练，逐渐提高

这是每个类别的F1-Score分数

这是模型的PR曲线

这是混淆矩阵

5、目标检测文章

1. YOLOv5s网络模型讲解(一看就会)
2. 生活垃圾数据集（YOLO版）
3. YOLOv5如何训练自己的数据集
4. 双向控制舵机（树莓派版）
5. 树莓派部署YOLOv5目标检测（详细篇）
6. YOLO_Tracking 实践 （环境搭建 & 案例测试）
7. 目标检测：数据集划分 & XML数据集转YOLO标签