13 Star 117 Fork 56

livingbody / AutoCutout

加入 Gitee
与超过 1200万 开发者一起发现、参与优秀开源项目,私有仓库也完全免费 :)
免费加入
该仓库未声明开源许可证文件(LICENSE),使用请关注具体项目描述及其代码上游依赖。
克隆/下载
meiyan.py 11.90 KB
一键复制 编辑 原始数据 按行查看 历史
livingbody 提交于 2020-05-29 00:43 . 增加ocr服务
# coding:utf-8
# author: Livingbody
# date: 2020.05.06
from flask import Flask, render_template, request, jsonify
import os
import requests
import base64
import time
from flask import Blueprint, render_template
import numpy as np
import cv2
import json
from PIL import Image, ImageDraw, ImageFont
import math
index_meiyan = Blueprint("meiyan", __name__)
# 设置允许的文件格式
ALLOWED_EXTENSIONS = set(['png', 'jpg', 'bmp', 'jpeg'])
# 当前文件所在路径
basepath = os.path.dirname(__file__)
def allowed_file(filename):
filename = filename.lower()
return '.' in filename and filename.rsplit('.', 1)[1] in ALLOWED_EXTENSIONS
# 上传并抠图
@index_meiyan.route('/meiyan', methods=['POST', 'GET']) # 添加路由
def meiyan():
if request.method == 'POST':
try:
f = request.files['file']
if not (f and allowed_file(f.filename)):
return render_template('404.html')
t = time.time()
dst_filename = str(t) + '.' + f.filename.split('.')[-1]
new_img_filename = 'meiyan' + dst_filename
sourcefile = os.path.join('static/images/source', new_img_filename)
# sourcefilepath = os.path.join('static/images/source', new_img_filename)
sourcefilepath = os.path.join(basepath, sourcefile)
f.save(sourcefilepath)
selected_meiyan = request.form.get('selected_meiyan')
meiyan_imgfile = meiyan_fun(sourcefilepath, selected_meiyan)
filename = os.path.join('static/images/target', meiyan_imgfile)
return render_template('meiyan_ok.html', val1=time.time(), sourcefile=sourcefile, filename=filename)
except Exception:
return render_template('404.html')
return render_template('meiyan.html')
def cv2_to_base64(image):
data = cv2.imencode('.jpg', image)[1]
return base64.b64encode(data.tostring()).decode('utf8')
# 美颜
def meiyan_fun(filename, selected_meiyan):
source_img_path = filename
t = time.time()
dst_filename = str(t) + '.' + source_img_path.split('.')[-1]
new_img_filename = 'meiyan' + dst_filename
new_img_path = os.path.join(basepath, 'static/images/target', new_img_filename)
src_img = cv2.imread(source_img_path)
url = "http://127.0.0.1:8866/predict/face_landmark_localization"
data = {'images': [cv2_to_base64(cv2.imread(source_img_path))]}
headers = {"Content-type": "application/json"}
r = requests.post(url=url, headers=headers, data=json.dumps(data))
# 打印预测结果
result = r.json()["results"]
data = result[0]['data'][0]
face_landmark = np.array(data, dtype='int')
if isinstance(selected_meiyan, str):
# 瘦脸
if selected_meiyan == '4':
src_img = thin_face(src_img, face_landmark)
cv2.imwrite(new_img_path, src_img)
# 美白
elif selected_meiyan == '2':
src_img = whitening(src_img, face_landmark)
cv2.imwrite(new_img_path, src_img)
# 在瘦脸的基础上,继续放大双眼
elif selected_meiyan == '3':
enlarge_eyes(src_img, face_landmark, radius=13, strength=13)
cv2.imwrite(new_img_path, src_img)
# 全套
elif selected_meiyan == '1':
src_img = whitening(src_img, face_landmark)
# cv2.imwrite(new_img_path, src_img)
enlarge_eyes(src_img, face_landmark, radius=13, strength=13)
cv2.imwrite(new_img_path, src_img)
else:
raise Exception('选择设置有误')
else:
raise Exception('设置有误')
print('美颜照已生成,已保存到' + new_img_path)
return new_img_filename
# 去除背景色
def convert(upload_path):
file_list = [upload_path]
files = [("image", (open(item, "rb"))) for item in file_list]
# 指定图片分割方法为deeplabv3p_xception65_humanseg并发送post请求
url = "http://127.0.0.1:8866/predict/image/deeplabv3p_xception65_humanseg"
r = requests.post(url=url, files=files)
t = time.time()
filename = str(t) + '.jpg'
results = eval(r.json()["results"])
for item in results:
mypath = os.path.join(basepath, 'static/images/target', filename)
with open(mypath, "wb") as fp:
fp.write(base64.b64decode(item["base64"].split(',')[-1]))
item.pop("base64")
return filename
# 更换背景
def change_back_groundcolor(filename, background_color):
if isinstance(background_color, str):
if background_color == '1':
color = [255, 0, 0, 1]
elif background_color == '2':
color = [67, 142, 219, 1]
elif background_color == '3':
color = [255, 255, 255, 1]
else:
raise Exception('背景色设置有误')
elif isinstance(background_color, list) or isinstance(background_color, tuple):
color = [background_color[0], background_color[1], background_color[2], 1]
else:
raise Exception('背景色设置有误')
base_img_filename = os.path.join(basepath, 'static/images/target', filename)
new_img_filename = 'color' + filename
new_img_path = os.path.join(basepath, 'static/images/target', new_img_filename)
base_img = Image.open(base_img_filename)
img = np.array(base_img)
for i in range(0, img.shape[0]):
for j in range(0, img.shape[1]):
if img[i][j][3] < 1:
img[i][j] = color
im = Image.fromarray(img)
im = im.convert('RGB')
im.save(new_img_path)
print('证件照已生成,已保存到' + new_img_path)
return new_img_filename
def bilinear_insert(image, new_x, new_y):
"""
双线性插值法
"""
w, h, c = image.shape
if c == 3:
x1 = int(new_x)
x2 = x1 + 1
y1 = int(new_y)
y2 = y1 + 1
part1 = image[y1, x1].astype(np.float) * (float(x2) - new_x) * (float(y2) - new_y)
part2 = image[y1, x2].astype(np.float) * (new_x - float(x1)) * (float(y2) - new_y)
part3 = image[y2, x1].astype(np.float) * (float(x2) - new_x) * (new_y - float(y1))
part4 = image[y2, x2].astype(np.float) * (new_x - float(x1)) * (new_y - float(y1))
insertValue = part1 + part2 + part3 + part4
return insertValue.astype(np.int8)
def local_traslation_warp(image, start_point, end_point, radius):
"""
局部平移算法
"""
radius_square = math.pow(radius, 2)
image_cp = image.copy()
dist_se = math.pow(np.linalg.norm(end_point - start_point), 2)
height, width, channel = image.shape
for i in range(width):
for j in range(height):
# 计算该点是否在形变圆的范围之内
# 优化,第一步,直接判断是会在(start_point[0], start_point[1])的矩阵框中
if math.fabs(i - start_point[0]) > radius and math.fabs(j - start_point[1]) > radius:
continue
distance = (i - start_point[0]) * (i - start_point[0]) + (j - start_point[1]) * (j - start_point[1])
if (distance < radius_square):
# 计算出(i,j)坐标的原坐标
# 计算公式中右边平方号里的部分
ratio = (radius_square - distance) / (radius_square - distance + dist_se)
ratio = ratio * ratio
# 映射原位置
new_x = i - ratio * (end_point[0] - start_point[0])
new_y = j - ratio * (end_point[1] - start_point[1])
new_x = new_x if new_x >= 0 else 0
new_x = new_x if new_x < height - 1 else height - 2
new_y = new_y if new_y >= 0 else 0
new_y = new_y if new_y < width - 1 else width - 2
# 根据双线性插值法得到new_x, new_y的值
image_cp[j, i] = bilinear_insert(image, new_x, new_y)
return image_cp
def thin_face(image, face_landmark):
"""
实现自动人像瘦脸
image: 人像图片
face_landmark: 人脸关键点
"""
end_point = face_landmark[30]
# 瘦左脸,3号点到5号点的距离作为瘦脸距离
dist_left = np.linalg.norm(face_landmark[3] - face_landmark[5])
local_traslation_warp(image, face_landmark[3], end_point, dist_left)
# 瘦右脸,13号点到15号点的距离作为瘦脸距离
dist_right = np.linalg.norm(face_landmark[13] - face_landmark[15])
image = local_traslation_warp(image, face_landmark[13], end_point, dist_right)
return image
def enlarge_eyes(image, face_landmark, radius=15, strength=10):
"""
放大眼睛
image: 人像图片
face_landmark: 人脸关键点
radius: 眼睛放大范围半径
strength:眼睛放大程度
"""
# 以左眼最低点和最高点之间的中点为圆心
left_eye_top = face_landmark[37]
left_eye_bottom = face_landmark[41]
left_eye_center = (left_eye_top + left_eye_bottom) / 2
# 以右眼最低点和最高点之间的中点为圆心
right_eye_top = face_landmark[43]
right_eye_bottom = face_landmark[47]
right_eye_center = (right_eye_top + right_eye_bottom) / 2
# 放大双眼
local_zoom_warp(image, left_eye_center, radius=radius, strength=strength)
local_zoom_warp(image, right_eye_center, radius=radius, strength=strength)
def local_zoom_warp(image, point, radius, strength):
"""
图像局部缩放算法
"""
height = image.shape[0]
width = image.shape[1]
left = int(point[0] - radius) if point[0] - radius >= 0 else 0
top = int(point[1] - radius) if point[1] - radius >= 0 else 0
right = int(point[0] + radius) if point[0] + radius < width else width - 1
bottom = int(point[1] + radius) if point[1] + radius < height else height - 1
radius_square = math.pow(radius, 2)
for y in range(top, bottom):
offset_y = y - point[1]
for x in range(left, right):
offset_x = x - point[0]
dist_xy = offset_x * offset_x + offset_y * offset_y
if dist_xy <= radius_square:
scale = 1 - dist_xy / radius_square
scale = 1 - strength / 100 * scale
new_x = offset_x * scale + point[0]
new_y = offset_y * scale + point[1]
new_x = new_x if new_x >= 0 else 0
new_x = new_x if new_x < height - 1 else height - 2
new_y = new_y if new_y >= 0 else 0
new_y = new_y if new_y < width - 1 else width - 2
image[y, x] = bilinear_insert(image, new_x, new_y)
def whitening(img, face_landmark):
"""
美白
"""
# 简单估计额头所在区域
# 根据0号、16号点画出额头(以0号、16号点所在线段为直径的半圆)
radius = (np.linalg.norm(face_landmark[0] - face_landmark[16]) / 2).astype('int32')
center_abs = tuple(((face_landmark[0] + face_landmark[16]) / 2).astype('int32'))
angle = np.degrees(np.arctan((lambda l: l[1] / l[0])(face_landmark[16] - face_landmark[0]))).astype('int32')
face = np.zeros_like(img)
cv2.ellipse(face, center_abs, (radius, radius), angle, 180, 360, (255, 255, 255), 2)
points = face_landmark[0:17]
hull = cv2.convexHull(points)
cv2.polylines(face, [hull], True, (255, 255, 255), 2)
index = face > 0
face[index] = img[index]
dst = np.zeros_like(face)
# v1:磨皮程度
v1 = 9
# v2: 细节程度
v2 = 2
tmp1 = cv2.bilateralFilter(face, v1 * 5, v1 * 12.5, v1 * 12.5)
tmp1 = cv2.subtract(tmp1, face)
tmp1 = cv2.add(tmp1, (10, 10, 10, 128))
tmp1 = cv2.GaussianBlur(tmp1, (2 * v2 - 1, 2 * v2 - 1), 0)
tmp1 = cv2.add(img, tmp1)
dst = cv2.addWeighted(img, 0.1, tmp1, 0.9, 0.0)
dst = cv2.add(dst, (10, 10, 10, 255))
index = dst > 0
img[index] = dst[index]
return img
Python
1
https://gitee.com/livingbody/AutoCutout.git
git@gitee.com:livingbody/AutoCutout.git
livingbody
AutoCutout
AutoCutout
master

搜索帮助