opencv的学习之目标匹配

模板匹配：
模板匹配和卷积原理很像，模板在原图像上从原点开始滑动，计算模板与（图像被模板覆盖的地方）
的差别程度，这个差异程度的计算方法在opencv里有6中
然后每次计算的结果放入一个矩阵里，作为输出结果。
假如 原图像是A*B，而模板是a*b，则输出结果的矩阵是（A-a+1)*(B-b+1)

import cv2
import numpy as np


img = cv2.imread("D:photodatasmarties.png')#读取原图
template = cv2.imread("D:photoREd.jpg")#读取模板

methods = ['cv2.TM_CCOEFF','cv2.TM_CCOEFF_NoRMED','cv2.TM_CCORR'
           'cv2.TM_CCORR_NORMED','cv2.TM_SQDIFF','cv2.TM_SQDIFF_NORMED']
 
result = cv2.matchTemplate(img,template,cv2.TM_SQDIFF)#进行模板匹配
 
min_val,max_val,min_loc,max_loc = cv2.minMaxLoc(result)#查找最大最小值位置
h,w = template.shape[:2]#获取模板的高和宽
top_left = min_loc#查找最小值位置
bottom_right = (top_left[0]+w , top_left[1]+h)#查找右下角位置
 
cv2.rectangle(img,top_left,bottom_right,255,2)#绘制匹配到的矩形
 
cv2.imshow("img",img)
cv2.imshow("template",template)
cv2.waitKey(0)
cv2.destroyAllWindows()

 效果展示：

 多目标的皮配

import cv2
import numpy as np


img = cv2.imread("D:photodatasmarties.png')#读取原图
template = cv2.imread("D:photoREd.jpg")#读取模板

methods = ['cv2.TM_CCOEFF','cv2.TM_CCOEFF_NoRMED','cv2.TM_CCORR'
           'cv2.TM_CCORR_NORMED','cv2.TM_SQDIFF','cv2.TM_SQDIFF_NORMED']
 
result = cv2.matchTemplate(img,template,cv2.TM_SQDIFF)#进行模板匹配
 
min_val,max_val,min_loc,max_loc = cv2.minMaxLoc(result)#查找最大最小值位置
h,w = template.shape[:2]#获取模板的高和宽
top_left = min_loc#查找最小值位置
bottom_right = (top_left[0]+w , top_left[1]+h)#查找右下角位置
 
cv2.rectangle(img,top_left,bottom_right,255,2)#绘制匹配到的矩形
 
cv2.imshow("img",img)
cv2.imshow("template",template)
cv2.waitKey(0)
cv2.destroyAllWindows()