python如何将图片转换素描画

代码如下

# -*- coding:utf-8 -*-

import cv2
import numpy as np
from tkinter import filedialog, Tk
from os import getcwd
from re import findall

def open_path():
 # 图片路径
 root = Tk()
 root.withdraw()
 file_path = (filedialog.askopenfilename(title='选择图片文件', filetypes=[('All Files', '*')]))
 return file_path

def dodgeNaive(image, mask):
 # determine the shape of the input image
 width, height = image.shape[:2]

# prepare output argument with same size as image
 blend = np.zeros((width, height), np.uint8)

for col in range(width):
   for row in range(height):
     # do for every pixel
     if mask[col, row] == 255:
       # avoid division by zero
       blend[col, row] = 255
     else:
       # shift image pixel value by 8 bits
       # divide by the inverse of the mask
       tmp = (image[col, row] << 8) / (255 - mask)
       # print('tmp={}'.format(tmp.shape))
       # make sure resulting value stays within bounds
       if tmp.any() > 255:
         tmp = 255
         blend[col, row] = tmp

return blend

def dodgeV2(image, mask):
 return cv2.divide(image, 255 - mask, scale=256)

def burnV2(image, mask):
 return 255 - cv2.divide(255 - image, 255 - mask, scale=256)

def rgb_to_sketch(src_image_name):
 print('转换中......')
 img_rgb = cv2.imread(src_image_name)
 img_gray = cv2.cvtColor(img_rgb, cv2.COLOR_BGR2GRAY)
 # 读取图片时直接转换操作
 # img_gray = cv2.imread('example.jpg', cv2.IMREAD_GRAYSCALE)

img_gray_inv = 255 - img_gray
 img_blur = cv2.GaussianBlur(img_gray_inv, ksize=(21, 21),
               sigmaX=0, sigmaY=0)
 img_blend = dodgeV2(img_gray, img_blur)

# cv2.imshow('original', img_rgb)
 # cv2.imshow('gray', img_gray)
 # cv2.imshow('gray_inv', img_gray_inv)
 # cv2.imshow('gray_blur', img_blur)
 cv2.imwrite(dst_image_name, img_blend)
 save_path = getcwd() + "\\" + dst_image_name # 保存路径
 print('转换完成!!!\n')
 print('保存路径:' + save_path)
 cv2.imshow(save_path, img_blend)
 cv2.waitKey(0)
 cv2.destroyAllWindows()

if __name__ == '__main__':
 print('请选择图片(路径不要含中文)：')
 src_image_name = open_path() # 文件路径
 print(src_image_name + '\n')
 image_name = ''.join(findall(r'[^\\/:*?"<>|\r\n]+$', src_image_name)) # 获取文件名
 dst_image_name = 'Sketch_' + image_name
 rgb_to_sketch(src_image_name)

效果如下

来源：https://cloud.tencent.com/developer/article/1640047