去除痘痘/软化图像/用周围像素替换图像像素

Question

http://beautifulapps.mobi/acne/

还有另一个例子：

我也需要在我自己的应用程序中实现这样的痤疮去除。我尝试在 Stack Overflow 上搜索，发现了几十个问题，但它们都与简单的图像过滤器有关。我想过应用模糊，但后来我想了想，我认为模糊不是这里的解决方案。

Brad GPUImage 库也在那里。它有很多影响，我已经经历过它们对我的情况没有帮助。我可能错过了一个重要的事情。我怎样才能实现这个目标？

Answer 1

GPUImage 框架将为您提供帮助。看看这里描述的过滤器：JH Labs。你需要做的是（这并不容易；很多工作）

1. 找出脸部肤色。使用人脸检测算法定位人脸，然后从前额和脸颊区域采样肤色。
2. 对于每种肤色类型，您需要有一系列您认为适合肤色的颜色。
3. 在您的过滤器中，如果相关像素落在肤色范围内，则应用最小或中值过滤器（本质上是噪声消除过滤器；而不是模糊过滤器 - 这会让您无处可去）。您可能需要应用多次过滤器。

Answer 2

也许这对概念有帮助。我用 python Opencv 项目做了粉刺去除器，下面是代码（我已经在需要的地方对代码进行了解释）：

from fileinput import filename
import cv2 as cv
from cv2 import waitKey
from cv2 import destroyAllWindows
from cv2 import VideoWriter
from cv2 import minMaxLoc
from more_itertools import padded
import numpy as py
import matplotlib.pyplot as plt
import matplotlib as mat
import math
from sympy import fps


# OpenCV Utility Class for Mouse Handling
# With functions for to mouse callback to work on
class Sketcher:
    def __init__(self, windowname, dests, colors_func):
        self.prev_pt = None
        self.windowname = windowname
        self.dests = dests
        self.colors_func = colors_func
        self.dirty = False
        self.show()
        cv.setMouseCallback(self.windowname, self.on_mouse)

    # Displaying windows
    def show(self):
        cv.imshow(self.windowname, self.dests[0])
        cv.imshow(self.windowname + ": mask", self.dests[1])

    # onMouse function for Mouse Handling
    def on_mouse(self, event, x, y, flags, param):
        pt = (x, y)
        if event == cv.EVENT_LBUTTONDOWN:
            self.prev_pt = pt
        elif event == cv.EVENT_LBUTTONUP:
            self.prev_pt = None

        # Draw the line between mouse clicks
        if self.prev_pt and flags & cv.EVENT_FLAG_LBUTTON:
            for dst, color in zip(self.dests, self.colors_func()):
                cv.line(dst, self.prev_pt, pt, color, 5)
            self.dirty = True
            self.prev_pt = pt
            self.show()


src = cv.imread("C:/Python/Python_OpenCV/pics/blemish.png", -1)

plt.figure(figsize=(10,5))
plt.imshow(src, cmap="gray")
plt.title("Original Image")

if src is None:
    print("Failed to load the image: {}".format(filename))

# Create the copy mage

src_mask = src.copy()

# Create a black copy of the original image
# Acts as a mask
inpaintMask = py.zeros(src.shape[:2], py.uint8)
# Create sketch using Opencv Utility Class:Sketcher
sketch = Sketcher("image", [src_mask, inpaintMask], lambda:((255, 255, 25), 255))

while True:
    ch = cv.waitKey()
    if ch == 27:
        break
    if ch == ord("t"):
        # Use Fast Marching Algorithm method
        res = cv.inpaint(src= src_mask, inpaintMask= inpaintMask, inpaintRadius= 3, flags= cv.INPAINT_TELEA)
        cv.imshow("Inpaint output using FMM", res)
    
    if ch == ord("n"):
        # Use Navier-Stokes Algorithm method
        res = cv.inpaint(src= src_mask, inpaintMask= inpaintMask, inpaintRadius= 3, flags= cv.INPAINT_NS)
        cv.imshow("Inpaint using NS technique", res)
    
    if ch == ord("r"):
        # ord = Return the Unicode code point for a one-character string.
        # original image
        src_mask[:] = src
        inpaintMask[:] = 0
        sketch.show()


plt.show()
cv.waitKey(0)
cv.destroyAllWindows()