Python基于百度AI的人脸识别系统--颜值检测

基于百度AI的人脸识别系统–颜值检测

刚开始学，觉得好玩就写了这个
主要是分为人脸识别系统的对接，UI的设计

人脸识别系统：
用的百度的AI，其中的AK,SK可以换成自己的，在百度开放平台上注册就能获得，都是免费的

import base64
import json
import requests

def face_baidu():
    class BaiduPicIndentify:
        def __init__(self,img):
            self.AK = "juqVLsljMBigcM4soXoVmMGr"
            self.SK = "g5EgLoGOxEs3jogREqGVWUYl1e5tLkUL"
            self.img_src = img
            self.headers = {
                "Content-Type": "application/json; charset=UTF-8"
            }
 
        def get_accessToken(self):
            host = 'https://aip.baidubce.com/oauth/2.0/token?grant_type=client_credentials&client_id=' + self.AK + '&client_secret=' + self.SK
            response = requests.get(host, headers=self.headers)
            json_result = json.loads(response.text)
            return json_result['access_token']
 
        def img_to_BASE64(slef,path):
            with open(path,'rb') as f:
                base64_data = base64.b64encode(f.read())
                return base64_data
 
        def detect_face(self):
            # 人脸检测与属性分析
            img_BASE64 = self.img_to_BASE64(self.img_src)
            request_url = "https://aip.baidubce.com/rest/2.0/face/v3/detect"
            post_data = {
                "image": img_BASE64,
                "image_type": "BASE64",
                "face_field": "gender,age,beauty,gender,race,expression",
                "face_type": "LIVE"
            }
            access_token = self.get_accessToken()
            request_url = request_url + "?access_token=" + access_token
            response = requests.post(url=request_url, data=post_data, headers=self.headers)
            json_result = json.loads(response.text)
            if json_result['error_msg']!='pic not has face':
               
                t1 = tk.Label(win, text=json_result['result']['face_num']).grid(row=4,column=1)
                t2 = tk.Label(win, text=json_result['result']['face_list'][0]['age']).grid(row=5,column=1)
                t3 = tk.Label(win, text=json_result['result']['face_list'][0]['beauty']).grid(row=6,column=1)
                t4 = tk.Label(win, text=json_result['result']['face_list'][0]['gender']['type']).grid(row=7,column=1)
                t5 = tk.Label(win, text=json_result['result']['face_list'][0]['race']['type']).grid(row=8,column=1)
                t6 = tk.Label(win, text=json_result['result']['face_list'][0]['expression']['type']).grid(row=9,column=1)
 
    if __name__=='__main__':
        img_src=fpath.get()
        baiduDetect = BaiduPicIndentify(img_src)
        baiduDetect.detect_face()

UI界面：

import tkinter as tk
from tkinter import filedialog,ttk

win = tk.Tk()
win.title("颜值检测")
win.geometry("400x200")
fpath=tk.StringVar()

l = tk.Label(win, text='颜值检测系统-由百度AI提供', bg='brown', font='黑体,20,bold',fg='white')
l.grid(row=1,column=0)

ttk.Button(win,text='打开图片',command=getfile).grid(row=2,column=0)
ttk.Entry(win,textvariable=fpath).grid(row=2,column=1)

l1 = tk.Label(win, text='人脸数：')
l1.grid(row=4,column=0)
l2 = tk.Label(win, text='人物年龄：')
l2.grid(row=5,column=0)
l3 = tk.Label(win, text='人物颜值评分：')
l3.grid(row=6,column=0)
l4 = tk.Label(win, text='人物性别：')
l4.grid(row=7,column=0)
l5 = tk.Label(win, text='人物种族：')
l5.grid(row=8,column=0)
l6 = tk.Label(win, text='人物表情：')
l6.grid(row=9,column=0)

b=tk.Button(win,text="点我检测",width=15,height=2,command=face_baidu)
b.grid(row=10,column=0)



win.mainloop()

用的tkinter的库，不太熟悉，其实界面没花多少时间，主要是打开图片，然后把图片路径赋值给百度AI函数那里怎么弄都不对，网上找了好久

附上完整代码：

import tkinter as tk
from tkinter import filedialog,ttk
import base64
import json
import requests

#打开文件对话框
def getfile():
  file_path=filedialog.askopenfilename()
  fpath.set(file_path)
  
def face_baidu():
    class BaiduPicIndentify:
        def __init__(self,img):
            self.AK = "juqVLsljMBigcM4soXoVmMGr"
            self.SK = "g5EgLoGOxEs3jogREqGVWUYl1e5tLkUL"
            self.img_src = img
            self.headers = {
                "Content-Type": "application/json; charset=UTF-8"
            }
 
        def get_accessToken(self):
            host = 'https://aip.baidubce.com/oauth/2.0/token?grant_type=client_credentials&client_id=' + self.AK + '&client_secret=' + self.SK
            response = requests.get(host, headers=self.headers)
            json_result = json.loads(response.text)
            return json_result['access_token']
 
        def img_to_BASE64(slef,path):
            with open(path,'rb') as f:
                base64_data = base64.b64encode(f.read())
                return base64_data
 
        def detect_face(self):
            # 人脸检测与属性分析
            img_BASE64 = self.img_to_BASE64(self.img_src)
            request_url = "https://aip.baidubce.com/rest/2.0/face/v3/detect"
            post_data = {
                "image": img_BASE64,
                "image_type": "BASE64",
                "face_field": "gender,age,beauty,gender,race,expression",
                "face_type": "LIVE"
            }
            access_token = self.get_accessToken()
            request_url = request_url + "?access_token=" + access_token
            response = requests.post(url=request_url, data=post_data, headers=self.headers)
            json_result = json.loads(response.text)
            if json_result['error_msg']!='pic not has face':
                t1 = tk.Label(win, text=json_result['result']['face_num']).grid(row=4,column=1)
                t2 = tk.Label(win, text=json_result['result']['face_list'][0]['age']).grid(row=5,column=1)
                t3 = tk.Label(win, text=json_result['result']['face_list'][0]['beauty']).grid(row=6,column=1)
                t4 = tk.Label(win, text=json_result['result']['face_list'][0]['gender']['type']).grid(row=7,column=1)
                t5 = tk.Label(win, text=json_result['result']['face_list'][0]['race']['type']).grid(row=8,column=1)
                t6 = tk.Label(win, text=json_result['result']['face_list'][0]['expression']['type']).grid(row=9,column=1)
 
    if __name__=='__main__':
        img_src=fpath.get()
        baiduDetect = BaiduPicIndentify(img_src)
        baiduDetect.detect_face()

win = tk.Tk()
win.title("颜值检测")
win.geometry("400x200")
fpath=tk.StringVar()

l = tk.Label(win, text='颜值检测系统-由百度AI提供', bg='brown', font='黑体,20,bold',fg='white')
l.grid(row=1,column=0)

ttk.Button(win,text='打开图片',command=getfile).grid(row=2,column=0)
ttk.Entry(win,textvariable=fpath).grid(row=2,column=1)

l1 = tk.Label(win, text='人脸数：')
l1.grid(row=4,column=0)
l2 = tk.Label(win, text='人物年龄：')
l2.grid(row=5,column=0)
l3 = tk.Label(win, text='人物颜值评分：')
l3.grid(row=6,column=0)
l4 = tk.Label(win, text='人物性别：')
l4.grid(row=7,column=0)
l5 = tk.Label(win, text='人物种族：')
l5.grid(row=8,column=0)
l6 = tk.Label(win, text='人物表情：')
l6.grid(row=9,column=0)

b=tk.Button(win,text="点我检测",width=15,height=2,command=face_baidu)
b.grid(row=10,column=0)

win.mainloop()

效果图：

PS:
百度AI系统那里主要是借鉴的其他博主的，这里附上原链接：https://blog.csdn.net/cskywit/article/details/81540500