一个关于OpenCV学习的问题

最近在为一个比赛做准备，打算搞个C++和OpenCV相关的。但是我只学了C++，没学过OpenCV。有人能提供点学习OpenCV（C++向的）的教程吗？最好是适合我这种只学过C++，对OpenCV一点都不懂的人的教程。
另外再问一下YoloV5是什么？谢谢大家🙏🙏

这个课讲得特别详细，看完你绝对懂了：

都不懂的话不建议看文字教程，直接看视频，B站直接搜索opencv入门，先多看几个，找一个适合你节奏和你觉得可以跟的上的看下去。看完入门差不多有个概念了，后面一些进阶的操作什么的看你自己有没有兴趣学了。
yolov5是一个yolo系列目标检测网络之一，目前从yoloV1到V8都有，各自有一些不同和创新，这里面比较好用的是V5，对新手是友好。

不知道你这个问题是否已经解决, 如果还没有解决的话:

• 你看下这篇博客吧, 应该有用👉 ：c++下通过onnx部署yolov5-6.0到OpenCV
• 除此之外, 这篇博客: [OpenCV实战]8 深度学习目标检测网络YOLOv3的训练中的 4 代码 部分也许能够解决你的问题, 你可以仔细阅读以下内容或跳转源博客中阅读:
  

  代码地址：

  https://github.com/luohenyueji/OpenCV-Practical-Exercise

  https://download.csdn.net/download/luohenyj/11025216

  如果没有积分（系统自动设定资源分数）看看参考链接。我搬运过来的，大修改没有。

  OpenImage获取Python3代码：

  linux:

  import csv
  import subprocess
  import os
  
  runMode = "train"
  classes = ["Snowman"]
  
  with open('class-descriptions-boxable.csv', mode='r') as infile:
      reader = csv.reader(infile)
      dict_list = {rows[1]:rows[0] for rows in reader}
  
  subprocess.run(['rm', '-rf', 'JPEGImages'])
  subprocess.run([ 'mkdir', 'JPEGImages'])
  
  subprocess.run(['rm', '-rf', 'labels'])
  subprocess.run([ 'mkdir', 'labels'],)
  
  for ind in range(0, len(classes)):
      
      className = classes[ind]
      print("Class " + str(ind) + " : " + className)
  
      commandStr = "grep " + dict_list[className] + " " + runMode + "-annotations-bbox.csv"
      print(commandStr)
      class_annotations = subprocess.run(commandStr.split(), stdout=subprocess.PIPE).stdout.decode('utf-8')
      class_annotations = class_annotations.splitlines()
  
      totalNumOfAnnotations = len(class_annotations)
      print("Total number of annotations : "+str(totalNumOfAnnotations))
  
      cnt = 0
      for line in class_annotations[0:totalNumOfAnnotations]:
          cnt = cnt + 1
          print("annotation count : " + str(cnt))
          lineParts = line.split(',')
          subprocess.run([ 'aws', 's3', '--no-sign-request', '--only-show-errors', 'cp', 's3://open-images-dataset/'+runMode+'/'+lineParts[0]+".jpg", 'JPEGImages/'+lineParts[0]+".jpg"])
          with open('labels/%s.txt'%(lineParts[0]),'a') as f:
              f.write(' '.join([str(ind),str((float(lineParts[5]) + float(lineParts[4]))/2), str((float(lineParts[7]) + float(lineParts[6]))/2), str(float(lineParts[5])-float(lineParts[4])),str(float(lineParts[7])-float(lineParts[6]))])+'\n')

  windows:

  import csv
  import subprocess
  import os
  
  #要下的数据集rain,test,valid
  runMode = "train"
  #类别
  classes = ["Snowman"]
  
  with open('class-descriptions-boxable.csv', mode='r') as infile:
      reader = csv.reader(infile)
      dict_list = {rows[1]:rows[0] for rows in reader}
  
  #删除以前下载的
  subprocess.run(['rd', '/s/q', 'JPEGImages'],shell=True)
  subprocess.run(['mkdir', 'JPEGImages'],shell=True)
  
  subprocess.run(['rd', '/s/q', 'labels'],shell=True)
  subprocess.run(['mkdir', 'labels'],shell=True)
  
  for ind in range(0, len(classes)):
      
      className = classes[ind]
      print("Class " + str(ind) + " : " + className)
      
      strs = dict_list[className]
      commandStr = "findstr /r "+ '"\<' + strs + '\>"' + " " + runMode + "-annotations-bbox.csv"
  
      class_annotations = subprocess.run(commandStr, stdout=subprocess.PIPE,shell=True).stdout.decode('utf-8')
      class_annotations = class_annotations.splitlines()
      print(commandStr.split(','))
      #多少张图像被下载
      totalNumOfAnnotations = len(class_annotations)
      print("Total number of annotations : "+str(totalNumOfAnnotations))
  
      cnt = 0
      for line in class_annotations[0:totalNumOfAnnotations]:
          cnt = cnt + 1
          print("annotation count : " + str(cnt))
          lineParts = line.split(',')
          subprocess.run([ 'aws', 's3', '--no-sign-request', '--only-show-errors', 'cp', 's3://open-images-dataset/'+runMode+'/'+lineParts[0]+".jpg", 'JPEGImages/'+lineParts[0]+".jpg"],shell=True)
          with open('labels/%s.txt'%(lineParts[0]),'a') as f:
              f.write(' '.join([str(ind),str((float(lineParts[5]) + float(lineParts[4]))/2), str((float(lineParts[7]) + float(lineParts[6]))/2), str(float(lineParts[5])-float(lineParts[4])),str(float(lineParts[7])-float(lineParts[6]))])+'\n')

  数据集分割：

  import random
  import os
  import subprocess
  import sys
  
  image_dir='/home/your/darknet/OpenImage/JPEGImages'
  def split_data_set():
  
      f_val = open("snowman_test.txt", 'w')
      f_train = open("snowman_train.txt", 'w')
      
      path, dirs, files = next(os.walk(image_dir))
      data_size = len(files)
  
      ind = 0
      data_test_size = int(0.1 * data_size)
      test_array = random.sample(range(data_size), k=data_test_size)
      
      for f in os.listdir(image_dir):
          if(f.split(".")[1] == "jpg"):
              ind += 1
              
              if ind in test_array:
                  f_val.write(image_dir+'/'+f+'\n')
              else:
                  f_train.write(image_dir+'/'+f+'\n')
  
  
  split_data_set()
  

  draw loss:

  import matplotlib.pyplot as plt
   
  log='train.log'
  lines = []
  for line in open(log):
      if "avg" in line:
          lines.append(line)
   
  iterations = []
  avg_loss = []
   
  print('Retrieving data and plotting training loss graph...')
  for i in range(len(lines)):
      try:
          lineParts = lines[i].split(',')
          iterations.append(int(lineParts[0].split(':')[0]))
          avg_loss.append(float(lineParts[1].split()[0]))
      except:
          continue
   
  fig = plt.figure()
  for i in range(0, len(lines)):
      plt.plot(iterations[i:i+2], avg_loss[i:i+2], 'r.-')
   
  plt.xlabel('Batch Number')
  plt.ylabel('Avg Loss')
  fig.savefig('training_loss_plot.png', dpi=300)
   
  print('Done! Plot saved as training_loss_plot.png')

如果你已经解决了该问题, 非常希望你能够分享一下解决方案, 写成博客, 将相关链接放在评论区, 以帮助更多的人 ^-^