kmeans聚类图像的像素并可视化

kmeans聚类图像的像素并可视化

1. 题目要求
给定的图像，对其像素进行聚类并可视化。

2. 代码实现

from sklearn.cluster import KMeans
import matplotlib.pyplot as plt
import numpy as np
# from sklearn.cluster.KMeans import evaluate_clustering
import PIL.Image as image
import os
def loadData(filePath):
    with open(filePath, 'rb') as f:
    # f = open(filePath, 'rb')  # deal with binary
        data = []
        img = image.open(f)  # return to pixel(像素值)
        m, n = img.size  # the size of image
        for i in range(m):
            for j in range(n):
                x, y, z = img.getpixel((i, j))
                # deal with pixel to the range 0-1 and save to data
                data.append([x / 256.0, y / 256.0, z / 256.0])
    # f.close()
    return np.mat(data), m, n

def main():
    imgData, row, col = loadData(r"D:\work\python\Horse100\0075.jpg")
    # setting clusers(聚类中心)
    #print("imgData.shape=",imgData.shape)
    img = image.open((r'D:\work\python\Horse100\0075.jpg'))
    plt.figure("Image")  # 图像窗口名称
    plt.subplot(231)
    plt.imshow(img)
    plt.axis('off')  # 关掉坐标轴为 off
    plt.xticks([])
    plt.yticks([])
    plt.title("Picture")
    for k in range(2, 7):
        index = '23'+str(k)
        plt.subplot(int(index))
        label = KMeans(n_clusters=k).fit_predict(imgData)
        #print("label.shape=",label.shape)
        # get the label of each pixel
        label = label.reshape([row, col])#
        # create a new image to save the result of K-Means
        pic_new = image.new("RGB", (row, col))
        for i in range(row):
            for j in range(col):
                if label[i][j] == 0:
                    pic_new.putpixel((i, j), (0, 0, 255))
                elif label[i][j] == 1:
                    pic_new.putpixel((i, j), (255, 0, 0))
                elif label[i][j] == 2:
                    pic_new.putpixel((i, j), (0, 255, 0))
                elif label[i][j] == 3:
                    pic_new.putpixel((i, j), (60, 0, 220))
                elif label[i][j] == 4:
                    pic_new.putpixel((i, j), (249, 219, 87))
                elif label[i][j] == 5:
                    pic_new.putpixel((i, j), (167, 255, 167))
                elif label[i][j] == 6:
                    pic_new.putpixel((i, j), (216, 109, 216))

        plt.title("K=%i"% k)
        plt.imshow(pic_new)
        plt.axis('off')  # 关掉坐标轴为 off
        plt.xticks([])
        plt.yticks([])
    # pic_new.save("./face_images/k6.jpg", "JPEG")
    plt.axis('off')  # 关掉坐标轴为 off
    plt.xticks([])
    plt.yticks([])
    plt.show()

if __name__ == "__main__":
    main()

ps:该代码来自某大佬，以下是我更改后的版本（迂回了一下），由于PIL的库函数函数前后配套，我只用了其中一个，要进行相应的转换。

import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
import PIL.Image as image
#读取原始图像
paths ="D:\\work\\python\\Horse100\\0075.jpg"
X = plt.imread(paths)
X = np.array(X)
#print(X.shape)
shape = row ,col ,dim =X.shape
X_ = X.reshape(-1,3)#(将矩阵化为2维，才能使用聚类)
#print(X_.shape)
def kmeans(X, n):
    kmeans = KMeans(n_clusters=n)
    kmeans.fit(X)
    Y = kmeans.predict(X)
    return Y

plt.figure(1)  # 图像窗口名称
plt.subplot(2,3,1)
plt.imshow(X)
plt.axis('off')  # 关掉坐标轴为 off
plt.xticks([])
plt.yticks([])
plt.title("Picture")
for t in range(2, 7):
    index = '23' + str(t)
    plt.subplot(int(index))
    label = kmeans(X_,t)
    print("label.shape=",label.shape)
    # get the label of each pixel
    label = label.reshape(row,col)
    # create a new image to save the result of K-Means
    pic_new = image.new("RGB", (col, row))#定义的是图像大小为y*x*3的图像，这里列在前面行在后面
    for i in range(col):
        for j in range(row):
                if label[j][i] == 0:
                    pic_new.putpixel((i, j), (0, 0, 255))#填写的是位置为（j,i）位置的像素，列和行也是反的
                elif label[j][i] == 1:
                    pic_new.putpixel((i, j), (255, 0, 0))
                elif label[j][i] == 2:
                    pic_new.putpixel((i, j), (0, 255, 0))
                elif label[j][i] == 3:
                    pic_new.putpixel((i, j), (60, 0, 220))
                elif label[j][i] == 4:
                    pic_new.putpixel((i, j), (249, 219, 87))
                elif label[j][i] == 5:
                    pic_new.putpixel((i, j), (167, 255, 167))
                elif label[j][i] == 6:
                    pic_new.putpixel((i, j), (216, 109, 216))
    title = "k="+str(t)
    plt.title(title)
    plt.imshow(pic_new)
    plt.axis('off')  # 关掉坐标轴为 off
    plt.xticks([])
    plt.yticks([])

plt.show()

3. 输出图像