preface

With the rise of live broadcasting, the career of anchor has gradually come into people's vision. Now each platform has the title of "master Huadan", "first brother", "first sister", etc. In fact, popularity is one aspect, but beauty is the hard power.

Next, I'll take you to the beauty test score of the anchor to see who is the most beautiful baby

General content of this article:

1. Crawling the live face image of the host

2. Call Baidu face detection open interface to score the face value

Environment introduction:

python 3.6

pycharm

requests

parsel(xapth)

1. Crawling the picture of the host

1.1 import module

import requests
import parsel

1.2 analyze the target web page to determine the url path to be crawled and the headers parameter

 

 

The code is as follows:

base_url = 'https://www.huya.com/g/4079'
headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/79.0.3945.117 Safari/537.36'

}

1.3 send request -- requests simulate browser to send request and get response data

 

The code is as follows:

response = requests.get(url=base_url, headers=headers)
html_data = response.text

1.4 parse data -- parsel is converted to Selector object, which has the method of xpath and can process the converted data

 

The code is as follows:

parse = parsel.Selector(html_data)
data_list = parse.xpath('//li[@class="game-live-item"]')
# print(data_list)

for data in data_list:
    img_url = data.xpath('./a/img/@data-original').get()  # Host face picture url address
    img_title = data.xpath('.//span/i/@title').get()  # Name of the anchor
    print(img_url, img_title)

    # Request picture data
    img_data = requests.get(url=img_url, headers=headers).content

1.5 save data

 

The code is as follows:

    # Prepare file name
    file_name = img_title + '.jpg'
    with open('img\\' + file_name, mode='wb') as f:
        print('Saving:', file_name)
        f.write(img_data)

2. Call Baidu face detection open interface

Register account in Baidu AI open platform

 

Click to enter face recognition

 

Create an app

 

After creation, enter the management application, open the application, and click download SDK

 

Do not download, click the instructions

 

 

Install SDK

 

pip install baidu-aip

Write code according to Baidu interface example

 

 

The code is as follows:

from aip import AipFace
import base64


def face_rg(file_Path):
    """ Your api_id AK SK """
    api_id = '20107883'
    api_key = 'Xela0yPoFtERUBSTFtNlEKbO'
    secret_key = 'E5TmneyfAzxfzowgwRErLT8RYe7MfkfG'

    client = AipFace(api_id, api_key, secret_key)  # Call the interface of color detection

    with open(file_Path, 'rb') as file:
         data = base64.b64encode(file.read())  # Picture type BASE64:Picturesque base64 Value, base64 Encoded picture data

    image = data.decode()

    imageType = "BASE64"
    options = {}
    options["face_field"] = 'beauty'

    """ Call face detection """
    result = client.detect(image, imageType, options)
    print(result)
    return result['result']['face_list'][0]['beauty']


if __name__ == '__main__':
    face_rg(r'..\Host color detection\img\Blue cloud-Summer flowers still.jpg')

3. Test score

3.1 import module, interface for cycle detection

The code is as follows:

import os

from Host color detection.Beauty detection_Interface import face_rg

path = './img'
image_list = os.listdir(path)
# print(image_list)

score_dict = {}

for image in image_list:
    try:
        name = image.split('.')[0]
        # print(name)
        image_path = path + '\\' + image  # Path to picture
        face_score = face_rg(image_path)
        # print(face_score)
        score_dict[name] = face_score
        # print(score_dict)
    except Exception as e:
        print('Detecting:{}|Detection failed!!!'.format(str(name)))
    else:
        print('Detecting:{}|Beauty score:{}'.format(str(name), str(face_score)))

print('\n===========================================Test complete===========================================')

print(score_dict.items())

# Dictionary in descending order of values
change_score = sorted(score_dict.items(), key=lambda x: x[1], reverse=True)  # lambda 1 in is the index of the tuple
print(change_score)

3.2 data output

for a, b in enumerate(change_score):
    print('The little sister's name is:{}Beauty value rank: No{}Her beauty score is:{}'.format(change_score[a][0], a+1, change_score[a][1]))

After running the code, get the final test result of beauty rating

 

Add Penguin Group 695185429 and you can get it for free. All the information is in the group file. Materials can be obtained, including but not limited to Python practice, PDF electronic documents, interview brochures, learning materials, etc