import cv2
import numpy as np


def face_detect_function(img):
    if len(img.shape) == 3:
        img = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)
    face_detect = cv2.CascadeClassifier(
        'haarcascade_frontalface_default.xml')  # 人脸识别分类器
    face = face_detect.detectMultiScale(img)
    for x, y, w, h in face:
        cv2.rectangle(img, (x, y, w, h), color=(0, 0, 255), thickness=2)
    return img


def removeBG(img, back):
    imgt = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    back = cv2.resize(back, (imgt.shape[1], imgt.shape[0]))
    if len(back.shape) == 3:
        back = cv2.cvtColor(back, cv2.COLOR_RGB2GRAY)
    bgModel = cv2.createBackgroundSubtractorMOG2(0, 50)
    img = cv2.bilateralFilter(img, 5, 50, 100)  # smoothing filter
    fgmask = bgModel.apply(img, learningRate=0.01)
    kernel = np.ones((3, 3), np.uint8)
    fgmask = cv2.erode(fgmask, kernel, iterations=1)
    img = cv2.bitwise_and(img, img, mask=fgmask)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    img = cv2.GaussianBlur(gray, (41, 41), 0)
    _, img = cv2.threshold(img, 0, 255, cv2.THRESH_OTSU)
    kernel = cv2.getStructuringElement(cv2.MORPH_CROSS, (17, 17), (-1, -1))
    img = cv2.dilate(img, kernel, iterations=5)
    result = imgt & img
    img = ~img
    result = (back & img) + result
    result = cv2.GaussianBlur(result, (5, 5), 0)
    return result