计算机视觉经典案例---从入门到深入学习（基于python---OpenCV）案例三：物体计数

# '''在一个对象的质点绘制文字说明'''
# import cv2
# import numpy as np
# o = cv2.imread("D:\\img\\cat3.jpg",1)
# cv2.imshow("original",o)
# gray = cv2.cvtColor(o,cv2.COLOR_BGR2GRAY)   #首先转化为灰度图像
# ret,binary = cv2.threshold(gray,127,255,cv2.THRESH_BINARY)    #转化为二值图像，绘制轮廓只能用二值图
# contours,heirarchy = cv2.findContours(binary,cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)
# x = cv2.drawContours(o,contours,0,(0,0,255),3)
# m00 = cv2.moments(contours[0])['m00']
# m01 = cv2.moments(contours[0])['m01']
# m10 = cv2.moments(contours[0])['m10']
# cx = int(m10/m00)
# cy = int(m01/m00)
# cv2.putText(o,'cat',(cx,cy),cv2.FONT_HERSHEY_SIMPLEX,1,(0,0,255),3)
# cv2.imshow("result",o)
# cv2.waitKey()
# cv2.destroyAllWindows()

# '''使用cv.contoursArea()计算轮廓面积'''
# import cv2
# o = cv2.imread("D:\\img\\opencv.png",1)
# cv2.imshow("original",o)
# gray = cv2.cvtColor(o,cv2.COLOR_BGR2GRAY)
# ret,binary = cv2.threshold(gray,127,255,cv2.THRESH_BINARY)
# contours,heirarchy = cv2.findContours(binary,cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)
# n = len(contours)
# for i in range(n) :
#     print("contours["+str(i)+"]面积 = ",cv2.contourArea(contours[i]))
#     cv2.drawContours(o,contours,i,(0,0,255),3)
# cv2.imshow("result",o)
# cv2.waitKey()
# cv2.destroyAllWindows()

# '''将面积大于1000的前景图片筛选出来'''
# import cv2
# o = cv2.imread("D:\\img\\opencv.png",1)
# gray = cv2.cvtColor(o,cv2.COLOR_BGR2GRAY)
# ret,binary = cv2.threshold(gray,127,255,cv2.THRESH_BINARY)
# contours ,hierarchy= cv2.findContours(binary,cv2.RETR_LIST,cv2.CHAIN_APPROX_SIMPLE)
#
# contoursok=[]
# for i in contours:
#     if cv2.contourArea(i) > 1000:
#         contoursok.append(i)
#         cv2.drawContours(o,contoursok,-1,(0,0,255),3)
# cv2.imshow("result",o)
# cv2.waitKey()
# cv2.destroyAllWindows()

# '''使用cv2.getStructuringElement函数生成不同的核'''
# import cv2
# kernel1 = cv2.getStructuringElement(cv2.MORPH_RECT,(5,5))
# kernel2 = cv2.getStructuringElement(cv2.MORPH_CROSS,(5,5))
# kernel3 = cv2.getStructuringElement(cv2.MORPH_ELLIPSE,(5,5))
# print("kernel1 = \n",kernel1)
# print("kernel2 = \n",kernel2)
# print("kernel3 = \n",kernel3)

# '''观察不同结构的核对形态学的影响'''
# import cv2
# o = cv2.imread("D:\\img\\kernel.bmp",cv2.IMREAD_UNCHANGED)
# kernel1 = cv2.getStructuringElement(cv2.MORPH_RECT,(59,59))
# kernel2 = cv2.getStructuringElement(cv2.MORPH_CROSS,(59,59))
# kernel3 = cv2.getStructuringElement(cv2.MORPH_ELLIPSE,(59,59))
# dst1 = cv2.dilate(o,kernel1)
# dst2 = cv2.dilate(o,kernel2)
# dst3 = cv2.dilate(o,kernel3)
# cv2.imshow("dst1",dst1)
# cv2.imshow("dst2",dst2)
# cv2.imshow("dst3",dst3)
# cv2.waitKey()
# cv2.destroyAllWindows()

# '''使用zip函数遍历多个数组'''
# a = '张三','李四','王五','赵六','小七'
# b = '机器视觉','python','opencv','getStructuringElement','dilate'
# for n,i,j in zip(range(len(a)),a,b):
#     print(n,i,j)
import cv2
img = cv2.imread("D:\\img\\count.jpg",1)
        # 图像与处理
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY) #色彩空间转换:彩色-->灰度图片
ret, binary = cv2.threshold(gray, 150, 255, cv2.THRESH_BINARY_INV) # 阈值处理二值反色
kernel=cv2.getStructuringElement(cv2.MORPH_ELLIPSE,(5,5))#核
erotion=cv2.erode(binary,kernel,iterations=4) #腐蚀操作
dilation=cv2.dilate(erotion,kernel,iterations=3)   #膨胀操作
gaussian = cv2.GaussianBlur(dilation,(3,3),0)# 高斯滤波
        # 查找所有轮廓
contours,hirearchy=cv2.findContours(gaussian, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
contoursok = []
for i in contours:
    if cv2.contourArea(i) > 30:
        contoursok.append(i)
        # 标注出轮廓
draw = cv2.drawContours(img,contoursok,-1,(0,0,255),1)
        # 计算每一个细胞的质心，并标记数字
for i,j in zip(contoursok,range(len(contoursok))):
    M = cv2.moments(i)
    x = int(M["m10"] / M["m00"])
    y = int(M["m01"] / M["m00"])
    cv2.putText(draw,str(j),(x,y),cv2.FONT_HERSHEY_PLAIN,1.5,(0,0,255),2)
        # 显示图片
cv2.imshow("gaussian",gaussian)
cv2.imshow("draw",draw)
cv2.waitKey()
cv2.destroyAllWindows()