Yolov5 adds a small target detection layer

Small target detection has always been CV One of the difficulties in the field , that ,YOLOv5 How to add a small target detection layer ？

---

YOLOv5 Code changes ———— For small target detection

---

1.YOLOv5 Introduction to the algorithm

YOLOv5 Mainly by input terminal 、Backone、Neck as well as Prediction Four-part composition . among ：

(1) Backbone： The convolution neural network of image features is formed by combining different fine-grained images .

(2) Neck： A network layer that mixes and combines image features , And transfer the image features to the prediction layer .

(3) Head： Prediction of image features , Generate bounding boxes and predict categories .

Detection framework ：

2. original YOLOv5 Model

# 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)
  ]

If you enter the image size =640X640,

# P3/8 The size of the corresponding detection feature map is 80X80, Used to detect that the size is in 8X8 The above goals .

# P4/16 The size of the corresponding detection feature map is 40X40, Used to detect that the size is in 16X16 The above goals .

# P5/32 The size of the corresponding detection feature map is 20X20, Used to detect that the size is in 32X32 The above goals .

3. Add a small target detection layer

# parameters
nc: 1  # 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]  #4
  - [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, BottleneckCSP, [128]],   #160*160
   [-1, 1, Conv, [256, 3, 2]],  # 3-P3/8
   [-1, 9, BottleneckCSP, [256]],  #80*80
   [-1, 1, Conv, [512, 3, 2]],  # 5-P4/16
   [-1, 9, BottleneckCSP, [512]], #40*40
   [-1, 1, Conv, [1024, 3, 2]],  # 7-P5/32
   [-1, 1, SPP, [1024, [5, 9, 13]]],
   [-1, 3, BottleneckCSP, [1024, False]],  # 9   20*20
  ]

# YOLOv5 head
head:
  [[-1, 1, Conv, [512, 1, 1]],  #20*20
   [-1, 1, nn.Upsample, [None, 2, 'nearest']], #40*40
   [[-1, 6], 1, Concat, [1]],  # cat backbone P4  40*40
   [-1, 3, BottleneckCSP, [512, False]],  # 13     40*40

   [-1, 1, Conv, [512, 1, 1]], #40*40
   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [[-1, 4], 1, Concat, [1]],  # cat backbone P3   80*80
   [-1, 3, BottleneckCSP, [512, False]],  # 17 (P3/8-small)  80*80

   [-1, 1, Conv, [256, 1, 1]], #18  80*80
   [-1, 1, nn.Upsample, [None, 2, 'nearest']], #19  160*160
   [[-1, 2], 1, Concat, [1]], #20 cat backbone p2  160*160
   [-1, 3, BottleneckCSP, [256, False]], #21 160*160

   [-1, 1, Conv, [256, 3, 2]],  #22   80*80
   [[-1, 18], 1, Concat, [1]], #23 80*80
   [-1, 3, BottleneckCSP, [256, False]], #24 80*80

   [-1, 1, Conv, [256, 3, 2]], #25  40*40
   [[-1, 14], 1, Concat, [1]],  # 26  cat head P4  40*40
   [-1, 3, BottleneckCSP, [512, False]],  # 27 (P4/16-medium) 40*40

   [-1, 1, Conv, [512, 3, 2]],  #28  20*20
   [[-1, 10], 1, Concat, [1]],  #29 cat head P5  #20*20
   [-1, 3, BottleneckCSP, [1024, False]],  # 30 (P5/32-large)  20*20

   [[21, 24, 27, 30], 1, Detect, [nc, anchors]],  # Detect(p2, P3, P4, P5)
  ]

# New addition 160X160 Detection characteristic diagram of , Used to detect 4X4 The above goals .

After improvement , Although the amount of computation and the speed of detection have increased , However, the detection accuracy of small targets has been significantly improved .