基于OpenCV实现两种方法测量圆弧长度（步骤 + 代码）

Color Space

发布于 2024-06-05 15:02:51

1040

发布于 2024-06-05 15:02:51

文章被收录于专栏：OpenCV与AI深度学习OpenCV与AI深度学习

导读

本文主要介绍基于OpenCV实现两种方法测量圆弧长度（步骤 + 代码）。

背景介绍

要求：如上所示，分别用OpenCV计算出图1和图2中圆弧的长度。因为OpenCV中没有提供现成计算圆弧的方法，所以需要自己编写，本文将提供2种不同的方法来实现，仅供参考。

实现步骤

首先以图1为例，如上图所示，方法一具体实现步骤如下：

【1】二值化 + 查找轮廓

img = cv2.imread('11.png')
cv2.imshow('src',img)

gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
ret,thresh = cv2.threshold(gray, 70, 255, cv2.THRESH_BINARY)

contours,hierarchy = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)

【2】查找轮廓凸包缺陷，确定圆弧起点和终点坐标

hull = cv2.convexHull(contours[0],returnPoints=False)#默认returnPoints=True

defects = cv2.convexityDefects(contours[0],hull)
start = end = (0,0)
for i in range(0,defects.shape[0]):
  s,e,f,d = defects[i,0]
  start = tuple(contours[0][s][0])
  end = tuple(contours[0][e][0])
  far = tuple(contours[0][f][0])
  if d > 5000:
    cv2.line(img,start,end,[0,255,0],2)#凸包
    #cv2.circle(img,far,5,[0,0,255],-1)#凸包缺陷点
    cv2.circle(img,end,5,[0,0,255],-1)#凸包缺陷点
    cv2.circle(img,start,5,[0,0,255],-1)#凸包缺陷点
    break
cv2.imshow('defects', img)
cv2.waitKey(0)

【3】获取轮廓最小外接圆，获取圆心，计算圆弧角度

center,radius = cv2.minEnclosingCircle(contours[0])
cv2.circle(img,(int(center[0]),int(center[1])),8,(255,0,255),-1)
cv2.circle(img,end,8,[0,255,255],-1)#凸包缺陷点
cv2.circle(img,start,8,[0,255,255],-1)#凸包缺陷点
cv2.line(img,start,(int(center[0]),int(center[1])),[0,255,0],2)#凸包
cv2.line(img,end,(int(center[0]),int(center[1])),[0,255,0],2)#凸包

angle = cal_ang(start,center,end)
print('angle = %0.2f' % angle)

【4】通过外接圆周长角度比例来计算弧长

length = (1 - angle / 360.0) * math.pi * radius * 2
print((angle / 360.0))
print('radius = %0.2f' % radius)
strL = 'length=%0.2f' % length

cv2.putText(img,strL,(int(center[0]-40),int(center[1]+40)),0,0.8,(255,0,0),2) 
cv2.imshow('result', img)

angle_1 = cal_ang(start, center, ((center[0]+100),(center[1])))
angle_2 = cal_ang(end, center, ((center[0]+100),(center[1])))
cv2.ellipse(img,(int(center[0]),int(center[1])),(int(radius),int(radius)),0,-angle_1,0,(255,0,255),2, cv2.LINE_AA)
cv2.ellipse(img,(int(center[0]),int(center[1])),(int(radius),int(radius)),0,0,angle_2,(255,0,255),2, cv2.LINE_AA)

cv2.imshow('result', img)
cv2.imwrite('result.bmp',img)

cv2.waitKey(0)
cv2.destroyAllWindows()

以图2做测试，验证结果：

方法二将步骤【2】替换为通过角点检测来获取弧线端点坐标：

feature_params = dict(maxCorners = 2,
                      qualityLevel = 0.3,
                      minDistance = 50,
                      blockSize = 5 )

corners = cv2.goodFeaturesToTrack(thresh,mask=None,**feature_params)

print ('cornerNum = ' + str(len(corners)))

for i in range(0,len(corners)):
    cv2.circle(img,(int(corners[i][0][0]),int(corners[i][0][1])),8,(0,255,0),-1, cv2.LINE_AA)
cv2.imshow('cornerNum', img)