91国在线产,免费一级在线视频,一级在线免费视频

算法中，初始種子可自動選擇（通過不同的劃分可以得到不同的種子，可按照自己需要改進算法），圖分別為原圖（自己畫了兩筆為了分割成不同區域）、灰度圖直方圖、初始種子圖、區域生長結果圖。

另外，不管時初始種子選擇還是區域生長，閾值選擇很重要。

				?

									import cv2

									import numpy as np

									import matplotlib.pyplot as plt

									#初始種子選擇

									def originalSeed(gray, th):

									 ret, thresh = cv2.cv2.threshold(gray, th, 255, cv2.THRESH_BINARY)#二值圖，種子區域(不同劃分可獲得不同種子)

									 kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3,3))#3×3結構元

									 thresh_copy = thresh.copy() #復制thresh_A到thresh_copy

									 thresh_B = np.zeros(gray.shape, np.uint8) #thresh_B大小與A相同，像素值為0

									 seeds = [ ] #為了記錄種子坐標

									 #循環，直到thresh_copy中的像素值全部為0

									 while thresh_copy.any():

									  Xa_copy, Ya_copy = np.where(thresh_copy > 0) #thresh_A_copy中值為255的像素的坐標

									  thresh_B[Xa_copy[0], Ya_copy[0]] = 255 #選取第一個點，并將thresh_B中對應像素值改為255

									  #連通分量算法，先對thresh_B進行膨脹，再和thresh執行and操作（取交集）

									  for i in range(200):

									   dilation_B = cv2.dilate(thresh_B, kernel, iterations=1)

									   thresh_B = cv2.bitwise_and(thresh, dilation_B)

									  #取thresh_B值為255的像素坐標，并將thresh_copy中對應坐標像素值變為0

									  Xb, Yb = np.where(thresh_B > 0)

									  thresh_copy[Xb, Yb] = 0

									  #循環，在thresh_B中只有一個像素點時停止

									  while str(thresh_B.tolist()).count("255") > 1:

									   thresh_B = cv2.erode(thresh_B, kernel, iterations=1) #腐蝕操作

									  X_seed, Y_seed = np.where(thresh_B > 0) #取處種子坐標

									  if X_seed.size > 0 and Y_seed.size > 0:

									   seeds.append((X_seed[0], Y_seed[0]))#將種子坐標寫入seeds

									  thresh_B[Xb, Yb] = 0 #將thresh_B像素值置零

									 return seeds

									#區域生長

									def regionGrow(gray, seeds, thresh, p):

									 seedMark = np.zeros(gray.shape)

									 #八鄰域

									 if p == 8:

									  connection = [(-1, -1), (-1, 0), (-1, 1), (0, 1), (1, 1), (1, 0), (1, -1), (0, -1)]

									 elif p == 4:

									  connection = [(-1, 0), (0, 1), (1, 0), (0, -1)]

									 #seeds內無元素時候生長停止

									 while len(seeds) != 0:

									  #棧頂元素出棧

									  pt = seeds.pop(0)

									  for i in range(p):

									   tmpX = pt[0] + connection[i][0]

									   tmpY = pt[1] + connection[i][1]

									   #檢測邊界點

									   if tmpX < 0 or tmpY < 0 or tmpX >= gray.shape[0] or tmpY >= gray.shape[1]:

									    continue

									   if abs(int(gray[tmpX, tmpY]) - int(gray[pt])) < thresh and seedMark[tmpX, tmpY] == 0:

									    seedMark[tmpX, tmpY] = 255

									    seeds.append((tmpX, tmpY))

									 return seedMark

									path = "_rg.jpg"

									img = cv2.imread(path)

									gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

									#hist = cv2.calcHist([gray], [0], None, [256], [0,256])#直方圖

									seeds = originalSeed(gray, th=253)

									seedMark = regionGrow(gray, seeds, thresh=3, p=8)

									#plt.plot(hist)

									#plt.xlim([0, 256])

									#plt.show()

									cv2.imshow("seedMark", seedMark)

									cv2.waitKey(0)