HALCON示例程序train_characters_ocr.hdev使用SVM分类器训练字体

小哥哥小姐姐觉得有用点个赞呗！

示例程序源码（加注释）

蓝色字体均为算子解释链接，可以前往查看解答

• 关于显示类函数解释

  read_image (Image, ‘ocr/chars_training_01.png’)

• 获取单通道图像的指针

  get_image_pointer1 (Image, Pointer, Type, Width, Height)

  dev_close_window ()

  dev_open_window (0, 0, Width, Height, ‘white’, WindowHandle)

  set_display_font (WindowHandle, 12, ‘mono’, ‘true’, ‘false’)

  dev_set_draw (‘margin’)

  dev_set_line_width (2)

  dev_display (Image)

  dev_update_window (‘off’)

  dev_update_pc (‘off’)

  dev_update_var (‘off’)

• 创建一个 SVM 分类器内容，从A-G

  ClassNames := [‘A’,‘B’,‘C’,‘D’,‘E’,‘F’,‘G’]

• 创建OCR分类器

  create_ocr_class_svm (8, 10, ‘constant’, [‘convexity’,‘num_holes’,‘projection_horizontal’,‘projection_vertical’], ClassNames, ‘rbf’, 0.02, 0.05, ‘one-versus-one’, ‘normalization’, 10, OCRHandle)

for i := 1 to 7 by 1

read_image (Image, ‘ocr/chars_training_’ + i$’.2d’)

dev_display (Image)

* halcon自定义函数大括号内为其代码

get_regions (Image, SortedRegions)

{

dev_set_color (‘white’)

• 二值化

  threshold (Image, Region, 0, 125)

• 分割连通域

  connection (Region, ConnectedRegions)

• 通过面积筛选区域

  select_shape (ConnectedRegions, SelectedRegions, ‘area’, ‘and’, 50, 999999)

• 对区域进行排序

  sort_region (SelectedRegions, SortedRegions, ‘character’, ‘true’, ‘row’)

  return ()

  }

  • 对元素进行计数

    count_obj (SortedRegions, NumberObjects)

    for j := 1 to NumberObjects by 1

  • 选取数组内指定元素

    select_obj (SortedRegions, ObjectSelected, j)

    if (i == 1 and j == 1)

    • 将训练字符储存到指定文件

      write_ocr_trainf (ObjectSelected, Image, ClassNames[j - 1], ‘train_characters_ocr.trf’)

      else

    • 向训练文件中添加对象

      append_ocr_trainf (ObjectSelected, Image, ClassNames[j - 1], ‘train_characters_ocr.trf’)

      endif

      dev_set_color (‘gray’)

      dev_display (ObjectSelected)

      disp_message (WindowHandle, ClassNames[j - 1], ‘window’, 10, 10 + (j * 20), ‘black’, ‘true’)

      endfor

      disp_continue_message (WindowHandle, ‘black’, ‘true’)

      stop ()

      endfor

      读训练文件并转化为图像

      read_ocr_trainf (Characters, ‘train_characters_ocr.trf’, CharacterNames)

      count_obj (Characters, NumberCharacters)

for i := 1 to NumberCharacters by 1

select_obj (Characters, CharacterSelected, i)

dev_clear_window ()

dev_display (CharacterSelected)

disp_message (WindowHandle, CharacterNames[i - 1], ‘window’, 10, 10, ‘black’, ‘true’)

endfor

• 训练OCR分类器

  trainf_ocr_class_svm (OCRHandle, ‘train_characters_ocr.trf’, 0.001, ‘default’)

• 缩减SVM分类器，以减少分类所用时间。

  reduce_ocr_class_svm (OCRHandle, ‘bottom_up’, 2, 0.001, OCRHandleReduced)

• 将训练出的分类器写入文件

  write_ocr_class_svm (OCRHandleReduced, ‘font_characters_ocr’)

• 清除OCR分类器

  clear_ocr_class_svm (OCRHandle)

  clear_ocr_class_svm (OCRHandleReduced)

  stop ()

  dev_clear_window ()

  disp_message (WindowHandle, ‘No more lines to execute’, ‘window’, 10, 10, ‘black’, ‘true’)

处理思路

这个例子是主要讲解了如何使用，SVM分类器进行OCR字符训练。

后记

大家有什么问题可以向我提问哈，我看到了第一时间回复，希望在学习的路上多多结交良师益友。