opencv kmeans聚類 實現圖像色彩量化

kmeans聚類實現灰度圖像色彩量化（使用更少灰度值表示原灰度圖像）

# coding: utf-8
import cv2
import numpy as np
import matplotlib.pyplot as plt

#讀取原始圖像灰度顏色
img = cv2.imread('d:/paojie_g.jpg', 0) 
#print(img.shape)

#獲取圖像高度、寬度
rows, cols = img.shape[:]

#圖像二維像素轉換爲一維
data = img.reshape((rows * cols))
data = np.float32(data)

#定義中心 (type,max_iter,epsilon)
criteria = (cv2.TERM_CRITERIA_EPS +
            cv2.TERM_CRITERIA_MAX_ITER, 10, 1.0)

#設置標籤
flags = cv2.KMEANS_RANDOM_CENTERS

#K-Means聚類 聚集成4類
compactness, labels, centers = cv2.kmeans(data, 4, None, criteria, 10, flags)

#生成最終圖像
res = centers[labels.flatten()]
dst = res.reshape((img.shape[0],img.shape[1]))

#用來正常顯示中文標籤
plt.rcParams['font.sans-serif']=['SimHei']

#顯示圖像
titles = [u'原始圖像', u'聚類圖像']  
images = [img, dst]  
for i in range(2):  
   plt.subplot(1,2,i+1), plt.imshow(images[i], 'gray'), 
   plt.title(titles[i])  
   plt.xticks([]),plt.yticks([])  
plt.show()

程序輸出結果

kmeans聚類實現彩色圖像色彩量化（使用更少色彩值表示原彩色圖像）

# coding: utf-8
import cv2
import numpy as np
import matplotlib.pyplot as plt

#讀取原始圖像
img = cv2.imread('d:paojie.png') 
print(img.shape)

#圖像二維像素轉換爲一維
data = img.reshape((-1,3))
data = np.float32(data)

#定義中心 (type,max_iter,epsilon)
criteria = (cv2.TERM_CRITERIA_EPS +
            cv2.TERM_CRITERIA_MAX_ITER, 10, 1.0)

#設置標籤
flags = cv2.KMEANS_RANDOM_CENTERS

#K-Means聚類 聚集成2類
compactness, labels2, centers2 = cv2.kmeans(data, 2, None, criteria, 10, flags)
print('compactness：', compactness)
print('labels2.shape：', labels2.shape)
print('centers2.shape：', centers2.shape)
print('labels2：\n', labels2)
print('centers2：\n', centers2)

#K-Means聚類 聚集成4類
compactness, labels4, centers4 = cv2.kmeans(data, 4, None, criteria, 10, flags)

#K-Means聚類 聚集成8類
compactness, labels8, centers8 = cv2.kmeans(data, 8, None, criteria, 10, flags)

#K-Means聚類 聚集成16類
compactness, labels16, centers16 = cv2.kmeans(data, 16, None, criteria, 10, flags)

#K-Means聚類 聚集成64類
compactness, labels64, centers64 = cv2.kmeans(data, 64, None, criteria, 10, flags)

#圖像轉換回uint8二維類型
centers2 = np.uint8(centers2)
res = centers2[labels2.flatten()]
print('res：\n', res)
dst2 = res.reshape((img.shape))

centers4 = np.uint8(centers4)
res = centers4[labels4.flatten()]
dst4 = res.reshape((img.shape))

centers8 = np.uint8(centers8)
res = centers8[labels8.flatten()]
dst8 = res.reshape((img.shape))

centers16 = np.uint8(centers16)
res = centers16[labels16.flatten()]
dst16 = res.reshape((img.shape))

centers64 = np.uint8(centers64)
res = centers64[labels64.flatten()]
dst64 = res.reshape((img.shape))

#圖像轉換爲RGB顯示
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
dst2 = cv2.cvtColor(dst2, cv2.COLOR_BGR2RGB)
dst4 = cv2.cvtColor(dst4, cv2.COLOR_BGR2RGB)
dst8 = cv2.cvtColor(dst8, cv2.COLOR_BGR2RGB)
dst16 = cv2.cvtColor(dst16, cv2.COLOR_BGR2RGB)
dst64 = cv2.cvtColor(dst64, cv2.COLOR_BGR2RGB)

#用來正常顯示中文標籤
plt.rcParams['font.sans-serif']=['SimHei']

#顯示圖像
titles = [u'原始圖像', u'聚類圖像 K=2', u'聚類圖像 K=4',
          u'聚類圖像 K=8', u'聚類圖像 K=16',  u'聚類圖像 K=64']  
images = [img, dst2, dst4, dst8, dst16, dst64]  
for i in range(6):  
   plt.subplot(2,3,i+1), plt.imshow(images[i], 'gray'), 
   plt.title(titles[i])  
   plt.xticks([]),plt.yticks([])  
plt.show()

控制檯輸出

量化結果輸出