Safe and intelligent inspections in culvert scenarios are of great significance to ensure the normal operation of facilities. We have already developed actual projects in scenarios such as roads, highways, tunnels, bridges, etc., but data based on culvert scenarios have not yet been developed. There have been attempts, and the core purpose of this article is to build an intelligent target detection and recognition system based on culvert scenarios.
Different from the previous series, which mainly uses yolov5 series models as the benchmark for model selection, the models here are mainly selected based on yolov7. In order to compare and analyze the performance effectiveness of the models under different parameters, here we have selected three models: yolov7, yolov7x and yolov7e6e respectively. model to develop the system model in this article.
First, let’s take a look at the renderings:
Next, let’s take a look at the three model configurations chosen for this article.
【yolov7】
# parameters
nc: 80 # number of classes
depth_multiple: 1.0 # model depth multiple
width_multiple: 1.0 # layer channel multiple
# anchors
anchors:
- [12,16, 19,36, 40,28] # P3/8
- [36,75, 76,55, 72,146] # P4/16
- [142,110, 192,243, 459,401] # P5/32
# yolov7 backbone
backbone:
# [from, number, module, args]
[[-1, 1, Conv, [32, 3, 1]], # 0
[-1, 1, Conv, [64, 3, 2]], # 1-P1/2
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [128, 3, 2]], # 3-P2/4
[-1, 1, Conv, [64, 1, 1]],
[-2, 1, Conv, [64, 1, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[[-1, -3, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]], # 11
[-1, 1, MP, []],
[-1, 1, Conv, [128, 1, 1]],
[-3, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [128, 3, 2]],
[[-1, -3], 1, Concat, [1]], # 16-P3/8
[-1, 1, Conv, [128, 1, 1]],
[-2, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -3, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [512, 1, 1]], # 24
[-1, 1, MP, []],
[-1, 1, Conv, [256, 1, 1]],
[-3, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 2]],
[[-1, -3], 1, Concat, [1]], # 29-P4/16
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [1024, 1, 1]], # 37
[-1, 1, MP, []],
[-1, 1, Conv, [512, 1, 1]],
[-3, 1, Conv, [512, 1, 1]],
[-1, 1, Conv, [512, 3, 2]],
[[-1, -3], 1, Concat, [1]], # 42-P5/32
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [1024, 1, 1]], # 50
]
# yolov7 head
head:
[[-1, 1, SPPCSPC, [512]], # 51
[-1, 1, Conv, [256, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[37, 1, Conv, [256, 1, 1]], # route backbone P4
[[-1, -2], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -2, -3, -4, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]], # 63
[-1, 1, Conv, [128, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[24, 1, Conv, [128, 1, 1]], # route backbone P3
[[-1, -2], 1, Concat, [1]],
[-1, 1, Conv, [128, 1, 1]],
[-2, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[[-1, -2, -3, -4, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [128, 1, 1]], # 75
[-1, 1, MP, []],
[-1, 1, Conv, [128, 1, 1]],
[-3, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [128, 3, 2]],
[[-1, -3, 63], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -2, -3, -4, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]], # 88
[-1, 1, MP, []],
[-1, 1, Conv, [256, 1, 1]],
[-3, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 2]],
[[-1, -3, 51], 1, Concat, [1]],
[-1, 1, Conv, [512, 1, 1]],
[-2, 1, Conv, [512, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -2, -3, -4, -5, -6], 1, Concat, [1]],
[-1, 1, Conv, [512, 1, 1]], # 101
[75, 1, RepConv, [256, 3, 1]],
[88, 1, RepConv, [512, 3, 1]],
[101, 1, RepConv, [1024, 3, 1]],
[[102,103,104], 1, IDetect, [nc, anchors]], # Detect(P3, P4, P5)
]
【yolov7x】
# parameters
nc: 80 # number of classes
depth_multiple: 1.0 # model depth multiple
width_multiple: 1.0 # layer channel multiple
# anchors
anchors:
- [12,16, 19,36, 40,28] # P3/8
- [36,75, 76,55, 72,146] # P4/16
- [142,110, 192,243, 459,401] # P5/32
# yolov7 backbone
backbone:
# [from, number, module, args]
[[-1, 1, Conv, [40, 3, 1]], # 0
[-1, 1, Conv, [80, 3, 2]], # 1-P1/2
[-1, 1, Conv, [80, 3, 1]],
[-1, 1, Conv, [160, 3, 2]], # 3-P2/4
[-1, 1, Conv, [64, 1, 1]],
[-2, 1, Conv, [64, 1, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 13
[-1, 1, MP, []],
[-1, 1, Conv, [160, 1, 1]],
[-3, 1, Conv, [160, 1, 1]],
[-1, 1, Conv, [160, 3, 2]],
[[-1, -3], 1, Concat, [1]], # 18-P3/8
[-1, 1, Conv, [128, 1, 1]],
[-2, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [640, 1, 1]], # 28
[-1, 1, MP, []],
[-1, 1, Conv, [320, 1, 1]],
[-3, 1, Conv, [320, 1, 1]],
[-1, 1, Conv, [320, 3, 2]],
[[-1, -3], 1, Concat, [1]], # 33-P4/16
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [1280, 1, 1]], # 43
[-1, 1, MP, []],
[-1, 1, Conv, [640, 1, 1]],
[-3, 1, Conv, [640, 1, 1]],
[-1, 1, Conv, [640, 3, 2]],
[[-1, -3], 1, Concat, [1]], # 48-P5/32
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [1280, 1, 1]], # 58
]
# yolov7 head
head:
[[-1, 1, SPPCSPC, [640]], # 59
[-1, 1, Conv, [320, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[43, 1, Conv, [320, 1, 1]], # route backbone P4
[[-1, -2], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 73
[-1, 1, Conv, [160, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[28, 1, Conv, [160, 1, 1]], # route backbone P3
[[-1, -2], 1, Concat, [1]],
[-1, 1, Conv, [128, 1, 1]],
[-2, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [160, 1, 1]], # 87
[-1, 1, MP, []],
[-1, 1, Conv, [160, 1, 1]],
[-3, 1, Conv, [160, 1, 1]],
[-1, 1, Conv, [160, 3, 2]],
[[-1, -3, 73], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 102
[-1, 1, MP, []],
[-1, 1, Conv, [320, 1, 1]],
[-3, 1, Conv, [320, 1, 1]],
[-1, 1, Conv, [320, 3, 2]],
[[-1, -3, 59], 1, Concat, [1]],
[-1, 1, Conv, [512, 1, 1]],
[-2, 1, Conv, [512, 1, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [640, 1, 1]], # 117
[87, 1, Conv, [320, 3, 1]],
[102, 1, Conv, [640, 3, 1]],
[117, 1, Conv, [1280, 3, 1]],
[[118,119,120], 1, IDetect, [nc, anchors]], # Detect(P3, P4, P5)
]
【yolov7e6e】
# parameters
nc: 80 # number of classes
depth_multiple: 1.0 # model depth multiple
width_multiple: 1.0 # layer channel multiple
# anchors
anchors:
- [ 19,27, 44,40, 38,94 ] # P3/8
- [ 96,68, 86,152, 180,137 ] # P4/16
- [ 140,301, 303,264, 238,542 ] # P5/32
- [ 436,615, 739,380, 925,792 ] # P6/64
# yolov7 backbone
backbone:
# [from, number, module, args],
[[-1, 1, ReOrg, []], # 0
[-1, 1, Conv, [80, 3, 1]], # 1-P1/2
[-1, 1, DownC, [160]], # 2-P2/4
[-1, 1, Conv, [64, 1, 1]],
[-2, 1, Conv, [64, 1, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [160, 1, 1]], # 12
[-11, 1, Conv, [64, 1, 1]],
[-12, 1, Conv, [64, 1, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [160, 1, 1]], # 22
[[-1, -11], 1, Shortcut, [1]], # 23
[-1, 1, DownC, [320]], # 24-P3/8
[-1, 1, Conv, [128, 1, 1]],
[-2, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 34
[-11, 1, Conv, [128, 1, 1]],
[-12, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 44
[[-1, -11], 1, Shortcut, [1]], # 45
[-1, 1, DownC, [640]], # 46-P4/16
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [640, 1, 1]], # 56
[-11, 1, Conv, [256, 1, 1]],
[-12, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [640, 1, 1]], # 66
[[-1, -11], 1, Shortcut, [1]], # 67
[-1, 1, DownC, [960]], # 68-P5/32
[-1, 1, Conv, [384, 1, 1]],
[-2, 1, Conv, [384, 1, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [960, 1, 1]], # 78
[-11, 1, Conv, [384, 1, 1]],
[-12, 1, Conv, [384, 1, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[-1, 1, Conv, [384, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [960, 1, 1]], # 88
[[-1, -11], 1, Shortcut, [1]], # 89
[-1, 1, DownC, [1280]], # 90-P6/64
[-1, 1, Conv, [512, 1, 1]],
[-2, 1, Conv, [512, 1, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [1280, 1, 1]], # 100
[-11, 1, Conv, [512, 1, 1]],
[-12, 1, Conv, [512, 1, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[-1, 1, Conv, [512, 3, 1]],
[[-1, -3, -5, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [1280, 1, 1]], # 110
[[-1, -11], 1, Shortcut, [1]], # 111
]
# yolov7 head
head:
[[-1, 1, SPPCSPC, [640]], # 112
[-1, 1, Conv, [480, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[89, 1, Conv, [480, 1, 1]], # route backbone P5
[[-1, -2], 1, Concat, [1]],
[-1, 1, Conv, [384, 1, 1]],
[-2, 1, Conv, [384, 1, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [480, 1, 1]], # 126
[-11, 1, Conv, [384, 1, 1]],
[-12, 1, Conv, [384, 1, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [480, 1, 1]], # 136
[[-1, -11], 1, Shortcut, [1]], # 137
[-1, 1, Conv, [320, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[67, 1, Conv, [320, 1, 1]], # route backbone P4
[[-1, -2], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 151
[-11, 1, Conv, [256, 1, 1]],
[-12, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 161
[[-1, -11], 1, Shortcut, [1]], # 162
[-1, 1, Conv, [160, 1, 1]],
[-1, 1, nn.Upsample, [None, 2, 'nearest']],
[45, 1, Conv, [160, 1, 1]], # route backbone P3
[[-1, -2], 1, Concat, [1]],
[-1, 1, Conv, [128, 1, 1]],
[-2, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [160, 1, 1]], # 176
[-11, 1, Conv, [128, 1, 1]],
[-12, 1, Conv, [128, 1, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[-1, 1, Conv, [64, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [160, 1, 1]], # 186
[[-1, -11], 1, Shortcut, [1]], # 187
[-1, 1, DownC, [320]],
[[-1, 162], 1, Concat, [1]],
[-1, 1, Conv, [256, 1, 1]],
[-2, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 199
[-11, 1, Conv, [256, 1, 1]],
[-12, 1, Conv, [256, 1, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[-1, 1, Conv, [128, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [320, 1, 1]], # 209
[[-1, -11], 1, Shortcut, [1]], # 210
[-1, 1, DownC, [480]],
[[-1, 137], 1, Concat, [1]],
[-1, 1, Conv, [384, 1, 1]],
[-2, 1, Conv, [384, 1, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [480, 1, 1]], # 222
[-11, 1, Conv, [384, 1, 1]],
[-12, 1, Conv, [384, 1, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[-1, 1, Conv, [192, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [480, 1, 1]], # 232
[[-1, -11], 1, Shortcut, [1]], # 233
[-1, 1, DownC, [640]],
[[-1, 112], 1, Concat, [1]],
[-1, 1, Conv, [512, 1, 1]],
[-2, 1, Conv, [512, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [640, 1, 1]], # 245
[-11, 1, Conv, [512, 1, 1]],
[-12, 1, Conv, [512, 1, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[-1, 1, Conv, [256, 3, 1]],
[[-1, -2, -3, -4, -5, -6, -7, -8], 1, Concat, [1]],
[-1, 1, Conv, [640, 1, 1]], # 255
[[-1, -11], 1, Shortcut, [1]], # 256
[187, 1, Conv, [320, 3, 1]],
[210, 1, Conv, [640, 3, 1]],
[233, 1, Conv, [960, 3, 1]],
[256, 1, Conv, [1280, 3, 1]],
[186, 1, Conv, [320, 3, 1]],
[161, 1, Conv, [640, 3, 1]],
[136, 1, Conv, [960, 3, 1]],
[112, 1, Conv, [1280, 3, 1]],
[[257,258,259,260,261,262,263,264], 1, IAuxDetect, [nc, anchors]], # Detect(P3, P4, P5, P6)
]
By default the exact same parameters are used for training, as shown below:
#yolov7
python train.py --cfg cfg/training/yolov7.yaml --weights weights/yolov7_training.pt --name yolov7 --epochs 100 --batch-size 32 --img 640 640 --device 0
#yolov7x
python3 train.py --cfg cfg/training/yolov7x.yaml --weights weights/yolov7x_training.pt --name yolov7x --epochs 100 --batch-size 32 --img 640 640 --device 0
#yolov7-e6e
python train_aux.py --cfg cfg/training/yolov7-e6e.yaml --weights weights/yolov7-e6e_training.pt --name yolov7e6e --epochs 100 --batch-size 32 --img 640 640 --device 0
Next, let’s visually compare and analyze the results:
[Precision Curve]
The Precision-Recall Curve is a visual tool used to evaluate the precision performance of a binary classification model under different thresholds. It helps us understand how the model performs at different thresholds by plotting the relationship between precision and recall at different thresholds.
Precision refers to the ratio of the number of samples that are correctly predicted as positive examples to the number of samples that are predicted to be positive examples. Recall refers to the ratio of the number of samples that are correctly predicted as positive examples to the number of samples that are actually positive examples.
The steps for plotting a precision curve are as follows:
Convert predicted probabilities into binary class labels using different thresholds. Usually, when the predicted probability is greater than the threshold, the sample is classified as a positive example, otherwise it is classified as a negative example.
For each threshold, the corresponding precision and recall are calculated.
Plot precision and recall at each threshold on the same graph to form a precision curve.
Based on the shape and changing trend of the accuracy curve, an appropriate threshold can be selected to achieve the required performance requirements.
By observing the precision curve, we can determine the best threshold according to our needs to balance precision and recall. Higher precision means fewer false positives, while higher recall means fewer false negatives. Depending on specific business needs and cost trade-offs, appropriate operating points or thresholds can be selected on the curve.
Precision curves are often used together with recall curves to provide a more comprehensive analysis of classifier performance and help evaluate and compare the performance of different models. 
[Recall Curve]
Recall Curve is a visualization tool used to evaluate the recall performance of a binary classification model under different thresholds. It helps us understand the performance of the model under different thresholds by plotting the relationship between the recall rate at different thresholds and the corresponding precision rate.
Recall refers to the ratio of the number of samples that are correctly predicted as positive examples to the number of samples that are actually positive examples. Recall rate is also called sensitivity (Sensitivity) or true positive rate (True Positive Rate).
The steps for plotting a recall curve are as follows:
Convert predicted probabilities into binary class labels using different thresholds. Usually, when the predicted probability is greater than the threshold, the sample is classified as a positive example, otherwise it is classified as a negative example.
For each threshold, the corresponding recall rate and the corresponding precision rate are calculated.
Plot recall and precision at each threshold on the same graph to form a recall curve.
Based on the shape and changing trend of the recall curve, an appropriate threshold can be selected to achieve the required performance requirements.
By observing the recall curve, we can determine the best threshold according to our needs to balance recall and precision. Higher recall means fewer false negatives, while higher precision means fewer false positives. Depending on specific business needs and cost trade-offs, appropriate operating points or thresholds can be selected on the curve.
Recall curves are often used together with precision curves to provide a more comprehensive analysis of classifier performance and help evaluate and compare the performance of different models. .
[F1 value curve]
The F1 value curve is a visualization tool used to evaluate the performance of a binary classification model under different thresholds. It helps us understand the overall performance of the model by plotting the relationship between Precision, Recall and F1 score at different thresholds.
The F1 score is the harmonic average of precision and recall, which takes into account both performance indicators. The F1 value curve can help us determine a balance point between different precision and recall rates to choose the best threshold.
The steps for plotting an F1 value curve are as follows:
Convert the predicted probabilities into binary class labels using different thresholds. Usually, when the predicted probability is greater than the threshold, the sample is classified as a positive example, otherwise it is classified as a negative example.
For each threshold, the corresponding precision, recall and F1 score are calculated.
Plot the precision, recall and F1 score at each threshold on the same graph to form an F1 value curve.
According to the shape and changing trend of the F1 value curve, an appropriate threshold can be selected to achieve the required performance requirements.
F1 value curves are often used together with receiver operating characteristic curves (ROC curves) to help evaluate and compare the performance of different models. They provide a more comprehensive analysis of classifier performance, allowing the selection of appropriate models and threshold settings based on specific application scenarios.
It may also be because there is no parameter adjustment. The overall experimental results show that the effect of the yolov7 model is optimal. In terms of model training time consumption, the training time consumption of yolov7 is also relatively the least. In the actual project development process, it is still necessary to choose a truly suitable model according to your own data scenario, which is the best way.