def regiongrowing(mat,epsilon,meancount,start_point): print mat.shape #use stack store candidates ST = Stack() #extracted result is stored here s = [] x = start_point[0] y = start_point[1] #feed enter stack ST.ensta((x,y)) while not ST.isEmpty(): #get mean temp = ST.getLastN(meancount) sum = 0 for i in temp: xt = i[0] yt = i[1] sum += mat[xt][yt] mean = float(sum)/len(temp) t = ST.desta() x = t[0] y = t[1] if t not in s: s.append( t ) #8-connected neighborhood if x < len(mat)-1 and abs(mat[x+1][y] - mean ) <= epsilon : if not ST.isInside( (x + 1 , y) ) and not (x + 1 , y) in s: ST.ensta( (x + 1 , y) ) if x < len(mat)-1 and y < len(mat[0])-1 and abs(mat[x+1][y+1] - mean ) <= epsilon : if not ST.isInside( (x + 1 , y+1) ) and not (x + 1 , y+1) in s: ST.ensta( (x + 1 , y+1) ) if y < len(mat[0])-1 and abs( mat[x][y+1]-mean ) <= epsilon: if not ST.isInside( (x, y + 1) ) and not (x , y + 1) in s: ST.ensta( (x , y + 1) ) if x >0 and y < len(mat[0])-1 and abs(mat[x-1][y+1] - mean ) <= epsilon : if not ST.isInside( (x - 1 , y+1) ) and not (x - 1 , y+1) in s: ST.ensta( (x - 1 , y+1) ) if x > 0 and abs( mat[x - 1] [y] - mean ) <= epsilon: if not ST.isInside( (x - 1 , y) ) and not (x - 1 , y) in s: ST.ensta( (x - 1 , y) ) if x > 0 and y >0 and abs(mat[x-1][y-1] - mean ) <= epsilon : if not ST.isInside( (x - 1 , y - 1) ) and not (x - 1 , y -1) in s: ST.ensta( (x - 1 , y-1) ) if y > 0 and abs( mat[x][y-1]-mean ) <= epsilon: if not ST.isInside( (x ,y - 1) ) and not (x , y - 1) in s: ST.ensta( (x , y - 1) ) if x < len(mat)-1 and y >0 and abs(mat[x+1][y-1] - mean ) <= epsilon : if not ST.isInside( (x +1 , y - 1) ) and not (x + 1 , y -1) in s: ST.ensta( (x + 1 , y - 1 ) ) print len(s) return s
def regiongrowing(mat, epsilon, meancount, start_point):
print mat.shape
#use stack store candidates
ST = Stack()
#extracted result is stored here
s = []
x = start_point[0]
y = start_point[1]
#feed enter stack
ST.ensta((x, y))
while not ST.isEmpty():
#get mean
temp = ST.getLastN(meancount)
sum = 0
for i in temp:
xt = i[0]
yt = i[1]
sum += mat[xt][yt]
mean = float(sum) / len(temp)
t = ST.desta()
x = t[0]
y = t[1]
if t not in s:
s.append(t)
#8-connected neighborhood
if x < len(mat) - 1 and abs(mat[x + 1][y] - mean) <= epsilon:
if not ST.isInside((x + 1, y)) and not (x + 1, y) in s:
ST.ensta((x + 1, y))
if x < len(mat) - 1 and y < len(mat[0]) - 1 and abs(mat[x + 1][y + 1] -
mean) <= epsilon:
if not ST.isInside((x + 1, y + 1)) and not (x + 1, y + 1) in s:
ST.ensta((x + 1, y + 1))
if y < len(mat[0]) - 1 and abs(mat[x][y + 1] - mean) <= epsilon:
if not ST.isInside((x, y + 1)) and not (x, y + 1) in s:
ST.ensta((x, y + 1))
if x > 0 and y < len(mat[0]) - 1 and abs(mat[x - 1][y + 1] -
mean) <= epsilon:
if not ST.isInside((x - 1, y + 1)) and not (x - 1, y + 1) in s:
ST.ensta((x - 1, y + 1))
if x > 0 and abs(mat[x - 1][y] - mean) <= epsilon:
if not ST.isInside((x - 1, y)) and not (x - 1, y) in s:
ST.ensta((x - 1, y))
if x > 0 and y > 0 and abs(mat[x - 1][y - 1] - mean) <= epsilon:
if not ST.isInside((x - 1, y - 1)) and not (x - 1, y - 1) in s:
ST.ensta((x - 1, y - 1))
if y > 0 and abs(mat[x][y - 1] - mean) <= epsilon:
if not ST.isInside((x, y - 1)) and not (x, y - 1) in s:
ST.ensta((x, y - 1))
if x < len(mat) - 1 and y > 0 and abs(mat[x + 1][y - 1] -
mean) <= epsilon:
if not ST.isInside((x + 1, y - 1)) and not (x + 1, y - 1) in s:
ST.ensta((x + 1, y - 1))
print len(s)
return s
