"""

# 用于存储椭圆肤色模型

"""

import cv2

import math

import numpy as np

import tkinter as tk

from tkinter import filedialog

import pandas as pd

from pandas import DataFrame

import time

root = tk.Tk()

root.withdraw()

file_path = filedialog.askopenfilename() # 读取RGB空间的图像

# Cr_csv_path = filedialog.askopenfilename()

# Cb_csv_path = filedialog.askopenfilename()

# img_csv_path = filedialog.askopenfilename()

# face_out_path = filedialog.askopenfilename()

dir_path = filedialog.askdirectory()

oval_name = dir_path + r'/oval_03.png'

img_read = cv2.imread(file_path) # 读取图像内容

img_YCrCb = cv2.cvtColor(img_read, cv2.COLOR_RGB2YCrCb) # 转换色彩空间

# 构建椭圆肤色模型

oval_matrix = np.zeros((256, 256), dtype=np.uint8) # 建立一个用于存储椭圆模型的矩阵

center = (round(113), round(152))

size = (25, 12)

angle = -20

color = 255

thickness = np.random.randint(1, 9)

img = cv2.ellipse(oval_matrix, center, size, angle, 0, 360, color, thickness=-1) # 取得椭圆肤色模型

# img_csv = DataFrame(img)

# img_csv.to_csv(img_csv_path, index=False, header=False, mode='w', decimal=',')

# print(img)

x, y, d = img_read.shape

x_matrix = np.zeros((x, y), dtype='uint8') # 建立一个匹配的二值矩阵图像

Y, Cr, Cb = cv2.split(img_YCrCb) # 取得三种矩阵

# Cb_csv = DataFrame(Cb)

# Cb_csv.to_csv(Cb_csv_path, index=False, header=False, mode='w', decimal=',')

# 遍历图片的每一个像素点

for row in range(x):

for col in range(y):

Cr_v = Cr[row, col]

Cb_v = Cb[row, col]

x_matrix[row, col] = img[Cb_v, Cr_v]

# x_csv = DataFrame(x_matrix)

# x_csv.to_csv(face_out_path, index=False, header=False, decimal=',', mode='w')

# 将图片和矩阵进行与运算

and_matrix = np.zeros((x, y, 3), dtype='uint8')

and_matrix[:, :, 0] = x_matrix

and_matrix[:, :, 1] = x_matrix

and_matrix[:, :, 2] = x_matrix

img_and = cv2.bitwise_and(img_read, and_matrix) # 得到去除干扰的完整脸部

print(img_and.shape)

img_gray = cv2.cvtColor(img_and, cv2.COLOR_BGR2GRAY) # 脸部转换成灰度图

r, img_binary = cv2.threshold(img_gray, 127, 255, cv2.THRESH_BINARY) # 脸部转换成二值图，准备进行腐蚀操作

img_read_gray = cv2.cvtColor(img_read,cv2.COLOR_BGR2GRAY) # 完整图片转换

# 图像腐蚀操作

kernel = np.ones((2, 2), np.uint8)

erosion = cv2.erode(img_binary, kernel)

# print(Y, Cr, Cb)

'''

# 圆检测

circles = cv2.HoughCircles(img_read_gray, cv2.HOUGH_GRADIENT, 1, 10, param1=50, param2=100, minRadius=10, maxRadius=200)

circles = np.uint16(np.around(circles))

for i in circles[0, :]:

# 画圆

cv2.circle(img_read, (i[0], i[1]), i[2], (0, 255, 0), 2)

# 画圆心点

cv2.circle(img_read, (i[0], i[1]), 2, (0, 0, 255), 3)

cv2.imshow('detected circles', img_read)

# cv2.imshow('name', erosion)

# cv2.imwrite(oval_name, img_binary)

cv2.waitKey(0)

'''