HALCON示例程序check_blister_mixed.hedv药品胶囊缺陷检测

示例程序源码（加注释）

• 读入图片与显示相关设置

  dev_close_window ()

  read_image (Image, ‘blister/blister_mixed_reference’)

  dev_open_window_fit_image (Image, 0, 0, -1, -1, WindowHandle)

  set_display_font (WindowHandle, 14, ‘mono’, ‘true’, ‘false’)

  dev_update_off ()

  dev_display (Image)

  dev_set_draw (‘margin’)

• 首先提取每个胶囊

  disp_message (WindowHandle, ‘Train gmm classifier on pill types’, ‘window’, -1, -1, ‘black’, ‘true’)

• 注意注意敲黑板，划重点。下边的这个函数是halcon函数的封装，里边代码有好多。内容是上一篇文章介绍过，大同小异，所以不理解的话看一下上一篇例子。下边的代码块是extract_pill_types里边的内容。

  extract_pill_types (Image, Chambers, ChambersUnion, Classes, PhiRef, RowRef, ColumnRef, PillTypeCount)

threshold (Image, Region, 90, 255)
connection (Region, ConnectedRegions)
select_shape (ConnectedRegions, SelectedRegions, 'area', 'and', 5000, 999999)
shape_trans (SelectedRegions, Blister, 'convex')
gen_empty_region (Chambers)
for I := 0 to 4 by 1
    Row := 107 + I * 70
    for J := 0 to 2 by 1
        Column := 177 + J * 150
        gen_rectangle2 (Rectangle, Row, Column, 0, 64, 30)
        concat_obj (Chambers, Rectangle, Chambers)
    endfor
endfor
difference (Blister, Chambers, Pattern)
union1 (Chambers, ChambersUnion)
orientation_region (Blister, PhiRef)
PhiRef := rad(180) + PhiRef
area_center (Blister, Area, RowRef, ColumnRef)

*  +++ Extract pattern for classification +++
select_shape (Chambers, PillType1, 'row', 'and', 1, 145)
union1 (PillType1, PillType1)
select_shape (Chambers, PillType2, 'row', 'and', 145, 270)
union1 (PillType2, PillType2)
select_shape (Chambers, PillType3, 'row', 'and', 270, 390)
union1 (PillType3, PillType3)

* Extract color space of yellow pills
reduce_domain (Image, PillType1, ImageReduced)
decompose3 (ImageReduced, ImageR, ImageG, ImageB)
threshold (ImageB, Region, 60, 95)

* Extract color space of red pills
reduce_domain (Image, PillType2, ImageReduced)
decompose3 (ImageReduced, ImageR, ImageG, ImageB)
* invert_image反转图像的灰度值
invert_image (ImageB, ImageInvert)
* hysteresis_threshold使用迟滞阈值操作
* 函数原型：hysteresis_threshold(Image : RegionHysteresis : Low, High, MaxLength : )
* Image ：输入图像；RegionHysteresis ：输出区域；Low：低于这个灰度不会被选中；High：高于这个灰度立刻被选中；
* MaxLength ：在Low与High之间的根据MaxLength 对比度进行选择，和carry边缘类似。
hysteresis_threshold (ImageInvert, RegionHysteresis2, 190, 200, 5)

* Extract color space of green pills
reduce_domain (Image, PillType3, ImageReduced)
decompose3 (ImageReduced, ImageR, ImageG, ImageB)
invert_image (ImageB, ImageInvert)
hysteresis_threshold (ImageInvert, RegionHysteresis3, 180, 200, 10)
 
 * 求两个区域的交集
intersection (Region, PillType1, PillType1)
intersection (RegionHysteresis2, PillType2, PillType2)
PillTypeCount := [3,6,6]
intersection (RegionHysteresis3, PillType3, PillType3)
concat_obj (PillType1, PillType2, Classes)
concat_obj (Classes, PillType3, Classes)
return ()
           

NumClasses := |PillTypeCount|

• 创建创建高斯混合模型
• 输入参数： NumDim：几个特征；NumClasses:样本分类个数；NumCenters:类中心的个数；CovarType:
• 协方差矩阵（‘Spherical’,‘diag’,‘full’）;Preprocessing:预处理是否使用，及特征向量转换类型；
• NumComponents:预处理参数，转换特征数；RandSend:迭代次数，初始化GMM；

• 输出参数：GMMHandle:高斯混合模型

  create_class_gmm (3, 3, [1,5], ‘spherical’, ‘normalization’, 10, 42, GMMHandle)

• 含义：增加训练样本（图片区），放入GMM 进行训练数据
• 输入参数：Image:训练的图像（多通道）;ClassRegions:图像中用于训练的类区域;

• GMMHandle:高斯混合模型;Randomize:高斯噪声标准差

  add_samples_image_class_gmm (Image, Classes, GMMHandle, 0)

• 含义：训练高斯混合模型
• 输入参数：GMMHandle：高斯混合模型;MixIter: 期望最大值算法中迭代最大数;
• Threshold:期望误差阈值，超过此值，计算取消;
• ClassPriors:类的先验概率类型;（‘training’,‘uniform’）'uniform’表示使用相同的权重，1/NumClasses;'training’表示类发生的概率是依据样本中的概率推断。
• Regularize:调整值，为了防止计算中出现奇异协方差矩阵。增加到协方差矩阵的对角线上的一个很小的值
• 输出参数：Centers:每个类的中心个数

• Iter:每个类的迭代次数

  train_class_gmm (GMMHandle, 100, 0.001, ‘training’, 0.0001, Centers, Iter)

• 检测胶囊

  Count := 12

  for FileIndex := 1 to Count by 1

  • 读入图片

    read_image (Image, ‘blister/blister_mixed_’ + FileIndex$‘02’)

  • 阈值分割

    threshold (Image, Region, 90, 255)

  • 分割连通域

    connection (Region, ConnectedRegions)

  • 面积对区域筛选

    select_shape (ConnectedRegions, SelectedRegions, ‘area’, ‘and’, 5000, 9999999)

  • 求取最大区域边缘

    shape_trans (SelectedRegions, RegionTrans, ‘convex’)

  • 利用椭圆变换，得出区域角度

    orientation_region (RegionTrans, Phi)

    if (abs(Phi) > rad(90))

    Phi := rad(180) + Phi

    endif

  • 求取区域面积与中心坐标

    area_center (RegionTrans, Area1, Row, Column)

  • 创建平移旋转变换矩阵

    vector_angle_to_rigid (Row, Column, Phi, RowRef, ColumnRef, PhiRef, HomMat2D)

  • 使用仿射变换矩阵进行任意仿射变换

    affine_trans_image (Image, ImageAffinTrans, HomMat2D, ‘constant’, ‘false’)

  • 减少图像定义域

    reduce_domain (ImageAffinTrans, ChambersUnion, ImageReduced)

  • 将图片分为rgb三通道图像

    decompose3 (ImageAffinTrans, ImageR, ImageG, ImageB)

  • 对每个胶囊进行识别检测
  • 含义：使用GMM分类一个图像
  • 输入参数：Image:训练的图像（多通道）；GMMHandle:GMM模型；
  • RejectionThreshold:分类拒绝阈值，如果图像所有的像素概率低于此值将不会分类任何类。

  • 输出参数：ClassRegions:分类的各个区域

    classify_image_class_gmm (ImageReduced, ClassRegions, GMMHandle, 0.005)

  • 对ClassRegions进行计数

    count_obj (ClassRegions, Number)

  • 生成空区域

    gen_empty_obj (FinalClasses)

  • 分割连通域

    connection (Chambers, ChambersRemaining)

    for Index := Number to 1 by -1

    dev_clear_window ()

    • 选择ClassRegions区域中的第Index个对象

      select_obj (ClassRegions, Region, Index)

    • 求交集

      intersection (ChambersRemaining, Region, Region)

    • 使用面积与宽度进行区域筛选

      select_shape (Region, PillsOfOneType, [‘area’,‘width’], ‘and’, [200,40], [3000,68])

    • 求取两个区域的差集

      difference (ChambersUnion, PillsOfOneType, RegionDifference)

    • 分割连通域

      connection (RegionDifference, ConnectedRegions)

    • 使用面积进行区域筛选

      select_shape (ConnectedRegions, SelectedRegions, ‘area’, ‘and’, 0, 7868)

    • 求取区域的最大边缘

      shape_trans (SelectedRegions, SelectedRegions, ‘convex’)

    • 联合区域

      union1 (SelectedRegions, SelectedRegions)

    • 求取区域差集

      difference (ChambersRemaining, SelectedRegions, ChambersRemaining)

    • 联合两个区域

      concat_obj (SelectedRegions, FinalClasses, FinalClasses)

      endfor

  • 得出结果进行处理得出结论

    gen_empty_obj (MissingPills)

    gen_empty_obj (WrongPills)

    gen_empty_obj (WrongNumberOfPills)

    difference (ChambersUnion, FinalClasses, LeftOvers)

    area_center (LeftOvers, Area, Row1, Column1)

    if (Area > 0)

    connection (LeftOvers, LeftOvers)

    count_obj (LeftOvers, Number)

    for Index := 1 to Number by 1

    select_obj (LeftOvers, ObjectSelected, Index)

    intensity (ObjectSelected, ImageB, Mean, Deviation)

    if (Deviation > 40)

    concat_obj (WrongPills, ObjectSelected, WrongPills)

    else

    concat_obj (MissingPills, ObjectSelected, MissingPills)

    endif

    endfor

    endif

  • Compute histogram

    CountFinalClass := []

    for Index := 1 to NumClasses by 1

    select_obj (FinalClasses, ObjectSelected, Index)

    connection (ObjectSelected, ObjectSelected)

    count_obj (ObjectSelected, Size)

    CountFinalClass := [CountFinalClass,Size]

    endfor

  • Display classification results and output allover statistic

    display_results (ImageAffinTrans, LeftOvers, FinalClasses, WrongPills, CountFinalClass, PillTypeCount, WindowHandle)

    if (FileIndex < Count)

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

    endif

    stop ()

    endfor

• Clear classifier handle

  clear_class_gmm (GMMHandle)

处理思路

此例子使用GMM高斯分类器进行胶囊识别识别，因为一版药上有多个种类的药片。

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。