模板匹配与霍夫变换⑨

一、模板匹配

模板匹配是指在当前图像A内寻找图像B最相似的部分，一般将图像A称为输入图像，将图像B称为模板图像。模板匹配的操作方法是将模板图像B在图像A上滑动，遍历所有像素以完成匹配

1.案例解读。

1.1导入需要的库

import cv2
import numpy as np
#Matplotlib是RGB
import matplotlib.pyplot as plt
%matplotlib inline 

#定义显示图片的函数,避免重复代码
def cv_show(name, img):
    cv2.imshow(name, img)
    cv2.waitKey()
    cv2.destroyAllWindows()

1.2读取模板图片

#读取模板图片
template = cv2.imread("lena_eye.jpg")
cv_show("template",template)

1.3读取检测图片

img = cv2.imread("lena.jpg")
cv_show("img", img)

1.4获取模板的大小

#获取到我们模板的大小h,w
h, w = template.shape[:2]

1.5进行匹配

#开始模板匹配过程(采用计算归一化平方不同,计算值越接近0,越相关)
res = cv2.matchTemplate(img, template, cv2.TM_SQDIFF_NORMED)
min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)
top_left = min_loc
bottom_right = (top_left[0] + w, top_left[1] + h)
#画出检测到的部分
imgcpy = img.copy()
cv2.rectangle(imgcpy, top_left, bottom_right, 255, 2)
#因为matplotlib显示为RGB图像，做一次色彩空间空间转换
imgcpy = cv2.cvtColor(imgcpy, cv2.COLOR_BGR2RGB)
plt.imshow(imgcpy, cmap='gray')

二、霍夫变换

霍夫变换是一种在图像中寻找直线，圆形以及其他简单形状的方法。霍夫变换采用类似于投票的方式来获取当前图像内的形状集合，该变换由Paul Hough于1962年首次提出。

1.案例解读

1.1HoughLines

import cv2
import numpy

img = cv2.imread("shape.png")
# 1.轮廓检测算法检测出轮廓
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
edges = cv2.Canny(gray, 50, 150)

# 2.投射到Hough空间进行形状检测
# 任何一条线都可以用(ρ，θ)这两个术语表示。
# 1）先定义一个累加器，(ρ，θ)对应直线，ρ和θ都分别依次增大(根据精度)，计算每对(ρ，θ)的投票数。
#    其中，ρ以像素为单位，θ以弧度为单位。rho和theta是ρ和θ的精度。
# 2）然后，根据threshold(阈值，最低投票数)来判断是否归为一条直线
lines = cv2.HoughLines(edges, 1, np.pi / 180, 50)

# 画线
for line in lines:
    rho, theta = line[0]
    a = numpy.cos(theta)
    b = numpy.sin(theta)
    x0 = rho * a
    y0 = rho * b
    x1 = int(x0 + 1000 * (-b))                  #这里的1000是为了求延长线，其他数值也可以
    y1 = int(y0 + 1000 * a)
    x2 = int(x0 - 1000 * (-b))
    y2 = int(y0 - 1000 * a)
    # 画线
    cv2.line(img, (x1, y1), (x2, y2), (0, 255, 0), 2)

cv2.imshow("img", img)
cv2.imshow("gray", gray)
cv2.imshow("edges", edges)
cv2.waitKey(0)
cv2.destroyAllWindows()

1.2HoughLinesP

import cv2
import numpy

img = cv2.imread("shape.png")
# 1.轮廓检测算法检测出轮廓
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
edges = cv2.Canny(gray, 50, 150)

minLineLength = 10
maxLineGap = 30
lines = cv2.HoughLinesP(edges, 1, np.pi / 180, 10,minLineLength,maxLineGap)

# 画线
for line in lines:
    x1, y1, x2, y2 = line[0]
    cv2.line(img, (x1, y1), (x2, y2), (0, 255, 0), 2)

cv2.imshow("img", img)
cv2.imshow("gray", gray)
cv2.imshow("edges", edges)
cv2.waitKey(0)
cv2.destroyAllWindows()

1.3 HoughCircles

# 2.圆检测
import cv2
import numpy

img = cv2.imread("shape.png")

# 1.轮廓检测算法检测出轮廓
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
edges = cv2.Canny(gray, 50, 100)

# 2.投射到Hough空间进行形状检测
circles = cv2.HoughCircles(edges, cv2.HOUGH_GRADIENT, 1, 30,\
                           param1=40, param2=20, minRadius=5, maxRadius=100)

# 画圆
if not circles is None:
    # 转换为int
    circles = np.uint16(numpy.around(circles))
    for circle in circles:
        x, y, r = circle[0]
        # 画圆
        cv2.circle(img, (x, y), r, (0, 0, 255), 2)

cv2.imshow("gray", gray)
cv2.imshow("edges", edges)
cv2.imshow("img", img)
cv2.waitKey(0)
cv2.destroyAllWindows()

2.实战演练：车道检测

import cv2
import numpy as np

def canny(image):
    #gray = cv2.cvtColor(image,cv2.COLOR_RGB2GRAY)
    blur = cv2.GaussianBlur(image,(5,5),0)#降低噪点
    canny = cv2.Canny(blur,50,150)
    return canny

def region_of_interest(image):#应用遮罩
    height = image.shape[0]
    ploygons = np.array([[(200,height),(1100,height),(550,250)]])
    mask = np.zeros_like(image)
    cv2.fillPoly(mask,ploygons,255)
    
    return mask

# read image
image = cv2.imread('test_image.jpg',0)
imgcopy = image.copy()
# copy not reference to
lane_image = np.copy(image)
canny = canny(lane_image)
roi = region_of_interest(canny)#ROI 
masked_image = cv2.bitwise_and(image,roi)#通过 bitwise_and 对两个图像的每一个像素做与运算，来将遮罩应用图像

cv2.imshow('canny',canny)
cv2.imshow('region_of_interest(canny)',roi)
cv2.imshow('masked_image',masked_image)
cv2.waitKey(0)
cv2.destroyAllWindows()

# 应用图像阈值化
ret, thresh = cv2.threshold(masked_image, 130, 145, cv2.THRESH_BINARY)
cv2.imshow('thresh',thresh)
lines = cv2.HoughLinesP(thresh, 1, np.pi/180, 30, maxLineGap=200)
# 画图

for line in lines:
    x1, y1, x2, y2 = line[0]
    cv2.line(imgcopy, (x1, y1), (x2, y2), (255, 0, 0), 3)
cv2.imshow('image',imgcopy)
cv2.waitKey(0)
cv2.destroyAllWindows()

© Fu Xianjun. All Rights Reserved.