Halcon实例转OpenCV之焊点检测

Posted 2022-12-01

    Halcon例程中有一个焊点检测例程ball.hdev，大家可以在例程中搜索查看，如下是实例使用的测试图和运行结果图：

    测试图像：

    运行结果图：

    实现的大致步骤：

    ① 二值化后获取Board区域，设置为分析的ROI

    ② 二值化 + 开运算获取和分割大致为焊点的区域

    ③ 通过Blob的圆度阈值删选，剔除非圆的Blob，保留圆形焊点

    ④ 计算圆形焊点的最小外接矩形直径并标示

    Halcon实现代码：

* ball.hdev: Inspection of Ball Bonding
* 
dev_update_window (off)
dev_close_window ()
dev_open_window (0, 0, 728, 512, black, WindowID)
read_image (Bond, die/die_03)
dev_display (Bond)
set_display_font (WindowID, 14, mono, true, false)
disp_continue_message (WindowID, black, true)
stop ()
threshold (Bond, Bright, 100, 255)
shape_trans (Bright, Die, rectangle2)
dev_set_color (green)
dev_set_line_width (3)
dev_set_draw (margin)
dev_display (Die)
disp_continue_message (WindowID, black, true)
stop ()
reduce_domain (Bond, Die, DieGrey)
threshold (DieGrey, Wires, 0, 50)
fill_up_shape (Wires, WiresFilled, area, 1, 100)
dev_display (Bond)
dev_set_draw (fill)
dev_set_color (red)
dev_display (WiresFilled)
disp_continue_message (WindowID, black, true)
stop ()
opening_circle (WiresFilled, Balls, 15.5)
dev_set_color (green)
dev_display (Balls)
disp_continue_message (WindowID, black, true)
stop ()
connection (Balls, SingleBalls)
select_shape (SingleBalls, IntermediateBalls, circularity, and, 0.85, 1.0)
sort_region (IntermediateBalls, FinalBalls, first_point, true, column)
dev_display (Bond)
dev_set_colored (12)
dev_display (FinalBalls)
disp_continue_message (WindowID, black, true)
stop ()
smallest_circle (FinalBalls, Row, Column, Radius)
NumBalls := |Radius|
Diameter := 2 * Radius
meanDiameter := mean(Diameter)
minDiameter := min(Diameter)
dev_display (Bond)
disp_circle (WindowID, Row, Column, Radius)
dev_set_color (white)
disp_message (WindowID, D:  + Diameter$.4, image, Row - 2 * Radius, Column, white, false)
dev_update_window (on)

    Halcon转OpenCV思路步骤：

    ① 获取Board区域并设置ROI，本文直接给定ROI，当然大家也可以先二值化，然后通过获取轮廓外接矩形的方式获取ROI，这里直接指定：

ROI = img[50:img.shape[0],0:img.shape[1]]

    ② 二值化 + 开运算获取大致焊点区域

gray = cv2.cvtColor(ROI,cv2.COLOR_BGR2GRAY)
gray = cv2.medianBlur(gray, 3)#中值滤波
ret,thresh = cv2.threshold(gray, 30, 255, cv2.THRESH_BINARY_INV)
cv2.imshow(thresh,thresh)
#矩形结构元素
k1=np.ones((19,19), np.uint8)
#圆形结构元素
k2=np.zeros((19,19), np.uint8)
cv2.circle(k2,(9,9),10,(1,1,1),-1, cv2.LINE_AA)
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, k2)

    这里特别需要注意的是halcon做开运算使用的opening_circle()算子，使用的圆形结构元素，对于提取圆形轮廓来说，圆形结构元素比矩形结构元素更好，下面是比对效果：

    矩形结构元素做开运算效果：

结构元素做开运算效果(更接近圆形轮廓本身)：

    ③ 通过圆形度来剔除非圆形轮廓，圆形度计算方式这里使用轮廓面积与轮廓最小外接矩形的比值，代码如下(圆形度大于0.65则保留)：

index = 0
for cnt in contours:
  area = cv2.contourArea(cnt)
  center,radius = cv2.minEnclosingCircle(cnt)
  circularity = area / (PI * radius * radius)
  if circularity > 0.65:
    random_color = colors[index % len(colors)]
    cv2.circle(img,(int(center[0]),int(center[1]+50)),int(radius),random_color,-1, cv2.LINE_AA)
    cv2.putText(img,("D=%0.2f"%(radius*2)), (int(center[0]-5),int(center[1]+50-25)), font, 0.45, (255,255,255), 2)
    index = index + 1

    ④ 计算标示相对简单，注意标注颜色如果我们使用0~255的随机数，可能生成的颜色会比较暗，这里使用一个包含6种颜色的数组，通过取余的方式来给对应的圆形轮廓赋值，然后用最小外接圆的参数来绘制。

color1 = (255, 0, 0)
color2 = (0, 255, 0)
color3 = (0, 0, 255)
color4 = (255, 255, 0)
color5 = (255, 0, 255)
color6 = (0, 255, 255)
colors = [color1,color2,color3,color4,color5,color6]
random_color = colors[index % len(colors)]

最终效果如下：

喜欢请关注本公众号，谢谢！