旅程的起點

發表於 : **2022-03-07, 14:59**

參考資料：
Colab File: https://colab.research.google.com/drive ... g(以火災煙霧為例)
Datasets : https://public.roboflow.com/

流程
1.pull YoloV5專案 + 安裝套件
2.註冊Roboflow + 下載訓練用公開資料
3.建立自定義模組並設定參數
4.訓練
5.訓練完畢，測試
－－－－－－－－－－－－－－－－－－－－－－－－－－－
1.建立工作區(資料夾)

2.pull專案 + 安裝套件

代碼: 選擇全部

git clone https://github.com/ultralytics/yolov5
pip install -qr yolov5/requirements.txt
cd yolov5

3.註冊Roboflow
Roboflow官網: https://roboflow.com/

4.取得公開資料集下載連結(Public Dataset Download Link)並下載
Roboflow Public Dataset: https://public.roboflow.com/

代碼: 選擇全部

curl -L "PASTE YOUR LINK HERE" > roboflow.zip; unzip roboflow.zip; rm roboflow.zip

5.建立自定義模組Use a custom model configuration(以火災煙霧為例)
# yolov5/models/custom_yolov5s.yaml

代碼: 選擇全部

# parameters
nc: {num_classes}  # 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, BottleneckCSP, [128]],
   [-1, 1, Conv, [256, 3, 2]],  # 3-P3/8
   [-1, 9, BottleneckCSP, [256]],
   [-1, 1, Conv, [512, 3, 2]],  # 5-P4/16
   [-1, 9, BottleneckCSP, [512]],
   [-1, 1, Conv, [1024, 3, 2]],  # 7-P5/32
   [-1, 1, SPP, [1024, [5, 9, 13]]],
   [-1, 3, BottleneckCSP, [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, BottleneckCSP, [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, BottleneckCSP, [256, False]],  # 17 (P3/8-small)

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

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

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

6.訓練

代碼: 選擇全部

python train.py --img 416 --batch 16 --epochs 100 --data =..\data.yaml --cfg =.\models\custom_yolov5s.yaml --weights '' --name yolov5s_results  --cache

7.執行
照片

代碼: 選擇全部

python detect.py --weights runs/exp0_yolov5s_results/weights/best.pt --img 416 --conf 0.4 --source ../test/images

輸出結果

代碼: 選擇全部

import glob
from IPython.display import Image, display

for imageName in glob.glob('/content/yolov5/inference/output/*.jpg'):
    display(Image(filename=imageName))
    print("\n")

影像

代碼: 選擇全部

python detect.py --weights runs/exp0_yolov5s_results/weights/best.pt --source ../video4.mp4 --conf 0.4

旅程的起點

[python] YoloV5 自訂物件偵測 w/ OpenCV on Windows

[python] YoloV5 自訂物件偵測 w/ OpenCV on Windows