YOLOv5增加小目标检测层
采用增加小目标检测层的方式来使YOLOv5能够检测小目标，只需要修改models下的yaml文件中的内容即可。
主要改变如下：
原yaml：
# parameters
nc: 80 # number of classes
depth_multiple: 0.33 # model depth multiple
width_multiple: 0.50 # layer channel multiple
# anchors
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 backbone
backbone:
# [from, number, module, args]
[[-1, 1, Focus, [64, 3]], # 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, 9, 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, 1, SPP, [1024, [5, 9, 13]]],
[-1, 3, C3, [1024, False]], # 9
]
# YOLOv5 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)
]
改变后的yaml：
# parameters
nc: 10 # number of classes
depth_multiple: 0.33 # model depth multiple
width_multiple: 0.50 # layer channel multiple
# anchors
anchors:
- [5,6, 8,14, 15,11]
- [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 backbone
backbone:
# [from, number, module, args]
[[-1, 1, Focus, [64, 3]], # 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, 9, 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, 1, SPP, [1024, [5, 9, 13]]],
[-1, 3, C3, [1024, False]], # 9
]
# YOLOv5 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, [512, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 4], 1, Concat, [1]], # cat backbone P3
[-1, 3, C3, [512, False]], # 17 (P3/8-small)
[-1, 1, Conv, [256, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 2], 1, Concat, [1]], # cat backbone P3
[-1, 3, C3, [256, False]], # 17 (P3/8-small)
[-1, 1, Conv, [256, 3, 2]],
[[-1, 18], 1, Concat, [1]], # cat head P4
[-1, 3, C3, [256, False]], # 20 (P4/16-medium)
[-1, 1, Conv, [256, 3, 2]],
[[-1, 14], 1, Concat, [1]],
[-1, 3, C3, [512, False]],
[-1, 1, Conv, [512, 3, 2]],
[[-1, 10], 1, Concat, [1]], # cat head P5
[-1, 3, C3, [1024, False]], # 23 (P5/32-large)
[[21, 24, 27, 30], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5)
]
主要改变了两个地方：anchors和head
（1）anchors
原yaml：
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
改变后yaml：
anchors:
- [5,6, 8,14, 15,11]
- [10,13, 16,30, 33,23] # P3/8
- [30,61, 62,45, 59,119] # P4/16
- [116,90, 156,198, 373,326] # P5/32（2）head
原yaml：
head:
[[-1, 1, Conv, [512, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 6], 1, Concat, [1]], # 第一段
[-1, 3, C3, [512, False]],
[-1, 1, Conv, [256, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 4], 1, Concat, [1]], # 第二段
[-1, 3, C3, [256, False]],
[-1, 1, Conv, [256, 3, 2]],
[[-1, 14], 1, Concat, [1]], # 第三段
[-1, 3, C3, [512, False]],
[-1, 1, Conv, [512, 3, 2]],
[[-1, 10], 1, Concat, [1]], # 第四段
[-1, 3, C3, [1024, False]],
[[17, 20, 23], 1, Detect, [nc, anchors]], # 第五段
]
改变后yaml：
head:
[[-1, 1, Conv, [512, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 6], 1, Concat, [1]], # 第一段
[-1, 3, C3, [512, False]],
[-1, 1, Conv, [512, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 4], 1, Concat, [1]], # 在第一段和第二段之间加入一段
[-1, 3, C3, [512, False]],
[-1, 1, Conv, [256, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[[-1, 2], 1, Concat, [1]], # 将第二段这个地方的4改成2
[-1, 3, C3, [256, False]],
[-1, 1, Conv, [256, 3, 2]],
[[-1, 18], 1, Concat, [1]], # 在第二段和第三段之间加入一段
[-1, 3, C3, [256, False]],
[-1, 1, Conv, [256, 3, 2]],
[[-1, 14], 1, Concat, [1]], # 第三段
[-1, 3, C3, [512, False]],
[-1, 1, Conv, [512, 3, 2]],
[[-1, 10], 1, Concat, [1]], # 第四段
[-1, 3, C3, [1024, False]],
[[21, 24, 27, 30], 1, Detect, [nc, anchors]], #将第五段[17, 20, 23]改成[21, 24, 27, 30]
]
这样就改好了。
注释：在yolov5的6.0版本作者将CSP换为C3，YOLOv5 2020年5月出来后不断更新，不断实践，设计出C3模块用来替换BottleneckCSP模块。当然这是作者在COCO等特定数据集上进行实验得出的，如果大家要进行迁移，也可以考虑不替换。二者的网络结构如下：
可以看出C3相对于BottleneckCSP模块，少了一个1x1 conv，同时撤掉了一个BN层和激活函数。
参考文献：
YOLOv5改进—增加小目标检测层_加勒比海带66的博客-CSDN博客_yolov5增加小目标检测层
yolov5增加小目标检测层
YOLOV5网络结构设计的思考_奔跑的阿诺的博客-CSDN博客_网络架构设计