将RGB图像转换为黑白PIL手部识别

Question

我试图用 python 编写与此处所写相同的内容，但我的代码没有产生好的结果。我的目标是获取 RGB 图像，调整大小并转换为 YCbCr，然后将背景像素值设置为 0，将手部像素值设置为 1。有人可以帮我使用 PIL 在 python 中编写此代码吗？（我试图复制的代码，我在步骤 3-6 时遇到了一些问题）function image_out = processSkinImage(filename)    Step 1...    % Read the image    original = imread(filename);    ...    Step 2...    % Resize the image to 50x50    image_resized = imresize(original, scale);    [M N Z] = size(image_resized);    % Initialize the output image    image_out = zeros(height,width);    image_out = zeros(M,N);    ...    Step 3...    % Convert the image from RGB to YCbCr    img_ycbcr = rgb2ycbcr(image_resized);    Cb = img_ycbcr(:,:,2);    Cr = img_ycbcr(:,:,3);    ...    Step 4...    % Get the central color of the image    % Expected the hand to be in the central of the image    central_color = img_ycbcr(int32(M/2),int32(N/2),:);    Cb_Color = central_color(:,:,2);    Cr_Color = central_color(:,:,3);    % Set the range    Cb_Difference = 15;    Cr_Difference = 10;    ...    Step 5...    % Detect skin pixels    [r,c,v] = find(Cb>=Cb_Color-Cr_Difference & Cb<=Cb_Color+Cb_Difference & Cr>=Cr_Color-Cr_Difference & Cr<=Cr_Color+Cr_Difference);    ...    Step 6...    % Mark detected pixels    for i=1:match_count        image_out(r(i),c(i)) = 1;    endend这就是我写的代码：from PIL import Image as imimage = im.open('/Users/eitan/Desktop/eell.jpg')image = image.resize((50,50), im.NEAREST)grayScale = image.convert(mode='L')width, height = grayScale.sizemid_pixel=grayScale.getpixel((width/2,height/2))print (mid_pixel)pixels = grayScale.load()for i in range(grayScale.size[0]):    # for every col:    for j in range(grayScale.size[1]):    # For every row        if grayScale.getpixel((i,j)) < mid_pixel+40 and grayScale.getpixel((i,j)) > mid_pixel-15:            pixels[i,j] = 255        else:            pixels[i, j] = 0grayScale.show()如果有人可以帮助我用python编写这段代码，那就太好了！

Answer 1

你可以像这样处理它，我使用的是 HSV 颜色空间而不是 YCbCr 颜色空间：

#!/usr/bin/env python3

import numpy as np

from PIL import Image

# Open image and convert to HSV colourspace

im = Image.open('hand.png').convert('HSV')

# Convert to Numpy array

ni = np.array(im)

# Get H, S and V of central pixel - consider taking a median of a larger area here

h,s,v = ni[int(ni.shape[0]/2), int(ni.shape[1]/2)]

# Separate each channel to own array

H = ni[:,:,0]

S = ni[:,:,1]

V = ni[:,:,2]

# Permissible +/- tolerances on each channel

deltah = 20

deltas = 80

deltav = 50

# Make masks of pixels with acceptable H, S and V

hmask = np.where((H > h-deltah) & (H < h+deltah), 255, 0).astype(np.uint8)

smask = np.where((S > s-deltas) & (S < s+deltas), 255, 0).astype(np.uint8)

vmask = np.where((V > v-deltav) & (V < v+deltav), 255, 0).astype(np.uint8)

# Save as images for inspection

Image.fromarray(hmask).save('hmask.png')

Image.fromarray(smask).save('smask.png')

Image.fromarray(vmask).save('vmask.png')

生成的色调蒙版：

产生的饱和度遮罩：

结果值掩码：

然后，您可以将掩码与或或组合在一起以获得更复杂的掩码组合。