def lip_segmentation2(inputRawImage, featureImage, inputLabel):
retLabel = inputLabel
width, height = retLabel.shape
for i in range(width):
for j in range(height):
retLabel[i][j] == 0
for i in range(width):
for j in range(height):
retLabel[i][j] = 255 * math.exp(
featureImage[i][j][0]) / (math.exp(featureImage[i][j][0]) +
math.exp(featureImage[i][j][1]))
print retLabel[i][j]
if (retLabel[i][j] > 128):
retLabel[i][j] = (retLabel[i][j] - 128) * 2
else:
retLabel[i][j] = 0
return 0, retLabel
features = extract_feature2(inputRawImage, featureImage, inputLabel)
center, clusterResult = kMeans.kMeans(mat(features), 2)
distance = sqrt(sum(power(center[0, 2:] - center[1, 2:], 2)))
for i in range(len(clusterResult)):
x = features[i][0]
y = features[i][1]
if distance > 0:
retLabel[x][y] = clusterResult[i, 0] + 1
else:
retLabel[x][y] = 2
return distance, retLabel
def lip_segmentation(inputRawImage, inputLabel):
retLabel = inputLabel
width, height = retLabel.shape
for i in range(width):
for j in range(height):
retLabel[i][j] == 0
features = extract_feature(inputRawImage, inputLabel)
center, clusterResult = kMeans.kMeans(mat(features), 2)
distance = sqrt(sum(power(center[0, 2:] - center[1, 2:], 2)))
for i in range(len(clusterResult)):
x = features[i][0]
y = features[i][1]
if distance > 0:
retLabel[x][y] = clusterResult[i, 0] + 1
else:
retLabel[x][y] = 2
return distance, retLabel
while loop:
print_menu()
try:
choice = int(input(' Enter your choice [1-5]: '))
except Exception:
print(' Enter valid choice!')
choice = 6
if choice == 1:
print(
'\n K-Means Clustering technique is used to group unrealated data.'
'\n For example : Unsupervised learning model (i.e. no additional information provided to work on) like where rich people lives'
)
input(' Press enter to proceed to plot')
kMeans()
elif choice == 2:
print(
'\n Naive Bayes is used for classifying data.'
'\n For example : Supervised learning model (i.e. making action according to previous deduction) like drug-test results'
)
input(' Press enter to proceed to result')
nBayes()
elif choice == 3:
print(
'\n Decision Tree is used for resolving to get conclusion from past decisions.'
'\n For example : Supervised learning model (i.e. making action according to previous deduction) like Should i play outside ?'
)
input(' Press enter to proceed to get decision tree')
if (not sum(dataSet[x])):
del dataSet[x]
t = list(userdata.keys())
for x in t:
if (not sum(userdata[x])):
del userdata[x]
centroids = []
for x in range(9):
centroids.append(dataSet[x])
print(len(dataSet))
dataSet = k_means.mat(dataSet)
centroids = k_means.kMeans(dataSet, 9, k_means.mat(centroids))
fl = []
for x in range(9):
fl.append(open('sort/' + str(x) + '.txt', 'w', encoding='utf-8'))
for x in userdata:
num = 0
distance = k_means.distEclud(centroids[0], userdata[x])
for i in range(1, 9):
if (k_means.distEclud(centroids[i], userdata[x]) < distance):
distance = k_means.distEclud(centroids[i], userdata[x])
num = i
fl[num].write(x + '\n')
for x in range(9):