cv2.matchTemplate(img,template,cv2.TM_SQDIFF)
參數(shù)一：原圖圖像對(duì)象名稱
參數(shù)二：模板圖像對(duì)象名稱
參數(shù)三：差別程度的計(jì)算方法(六選一推薦使用帶歸一化的)

模板匹配和卷積原理很像，模板從原圖像上從原點(diǎn)開始滑動(dòng)，計(jì)算模板與(圖像被模板覆蓋的地方)的差別程度，這個(gè)差別程度的計(jì)算方法在opencv中有六種，然后將每次計(jì)算的結(jié)果放入一個(gè)矩陣?yán)?#xff0c;作為結(jié)果輸出。
假如原圖像是A×B的大小，而模板是a×b的大小，則輸出結(jié)果的矩陣是(A-a+1)×(B-b+1)

import cv2 import numpy as np from matplotlib import pyplot as pltdef show_photo(name,picture):#圖像顯示函數(shù)cv2.imshow(name,picture)cv2.waitKey(0)cv2.destroyAllWindows()img = cv2.imread('E:\Jupyter_workspace\study\data/A.jpg',0) template = cv2.imread('E:\Jupyter_workspace\study\data/Aa.jpg',0) h, w =template.shape[:2]img.shape#結(jié)果為：(300,300) （h,w）template.shape#結(jié)果為：(100,100) （h,w）res = cv2.matchTemplate(img,template,cv2.TM_SQDIFF) res.shape#結(jié)果為：(201, 201)min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res) min_val#結(jié)果為：-16.0 max_val#結(jié)果為：88739848.0 min_loc#結(jié)果為：(99, 60) max_loc#結(jié)果為：(61, 0) 差別程度方法名稱效果

TM_SQDIFF	計(jì)算平方不同，計(jì)算出來的值越小，越相關(guān)
TM_CCORR	計(jì)算相關(guān)性，計(jì)算出來的值越大，越相關(guān)
TM_CCOEFF	計(jì)算相關(guān)系數(shù)，計(jì)算出來的值越大，越相關(guān)
TM_SQDIFF_NORMED	計(jì)算歸一化平方不同，計(jì)算出來的值越接近0，越相關(guān)
TM_CCORR_NORMED	計(jì)算歸一化相關(guān)性，計(jì)算出來的值越接近1，越相關(guān)
TM_CCOEFF_NORMED	計(jì)算歸一化相關(guān)系數(shù)，計(jì)算出來的值越接近1，越相關(guān)

盡量用后三個(gè)歸一化操作效果更好
六種具體差別程度公式參考手冊(cè)

import cv2 import numpy as np from matplotlib import pyplot as pltdef show_photo(name,picture):#圖像顯示函數(shù)cv2.imshow(name,picture)cv2.waitKey(0)cv2.destroyAllWindows()img = cv2.imread('E:\Jupyter_workspace\study\data/A.jpg',0) template = cv2.imread('E:\Jupyter_workspace\study\data/Aa.jpg',0) show_photo('img',img)#原圖 show_photo('template',template)#匹配模板，需要從原圖中找到與之相對(duì)應(yīng) h, w =template.shape[:2]methods = ['cv2.TM_CCOEFF','cv2.TM_CCOEFF_NORMED','cv2.TM_CCORR','cv2.TM_CCORR_NORMED','cv2.TM_SQDIFF','cv2.TM_SQDIFF_NORMED'] res = cv2.matchTemplate(img,template,cv2.TM_SQDIFF) min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)for meth in methods:img2 = img.copy()#匹配方法的真假值method = eval(meth)print(method)res = cv2.matchTemplate(img, template, method)min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(res)#如果是平方差匹配TM_SQDIFF或歸一化平方差匹配TM_SQDIFF_NORMED，取最小值if method in [cv2.TM_SQDIFF, cv2.TM_SQDIFF_NORMED]:top_left = min_locelse:top_left = max_locbottom_right = (top_left[0] + w, top_left[1]+ h)#畫矩陣cv2.rectangle(img2, top_left, bottom_right, 255, 2)plt.subplot(121), plt.imshow(res,cmap='gray')plt.xticks([]), plt.yticks([])#隱藏坐標(biāo)軸plt.subplot(122), plt.imshow(img2,cmap='gray')plt.xticks([]), plt.yticks([])plt.suptitle(meth)plt.show()

原圖：

模板：

匹配多個(gè)對(duì)象

上面的模板匹配原圖中只有一個(gè)模板的存在
下面進(jìn)行匹配原圖中存在多個(gè)模板，進(jìn)行多個(gè)對(duì)象的匹配

import cv2 import numpy as np from matplotlib import pyplot as pltdef show_photo(name,picture):#圖像顯示函數(shù)cv2.imshow(name,picture)cv2.waitKey(0)cv2.destroyAllWindows()img_rgb = cv2.imread('E:\Jupyter_workspace\study\data/mario.jpg')#原圖 show_photo('img_rgb',img_rgb) img_gray = cv2.cvtColor(img_rgb, cv2.COLOR_BGR2GRAY) template = cv2.imread('E:\Jupyter_workspace\study\data/mario_1.jpg',0)#模板，0為灰度圖，1為原色 show_photo('template',template) h, w = template.shape[:2]#template.shape為[h,w,c]這里只用h和w即可，c為顏色通道，彩色為3(BGR)，灰度圖不顯示res = cv2.matchTemplate(img_gray, template, cv2.TM_CCOEFF_NORMED) threshold = 0.8#取匹配程度大于80%的坐標(biāo) loc = np.where(res >= threshold) for pt in zip(*loc[::-1]):#*表示可選參數(shù)bottom_right = (pt[0] + w, pt[1] + h)cv2.rectangle(img_rgb, pt, bottom_right, (0,0,255), 2)show_photo('img_rgb',img_rgb)

原圖：

模板：

匹配多個(gè)對(duì)象：

總結(jié)

以上是生活随笔為你收集整理的16-模板匹配的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。