基于face_recognition与Django人脸识别服务案例

基础

人脸识别face_recognition编程案例-CSDN博客

案例

需求一、编写代码实现“人脸库注册”服务

post http://127.0.0.1:8000/faceregister 请求参数：imgFile：图片base64编码cardid：身份证ID

步骤一：创建sourceai/controller/face_view.py文件，内容如下：

from django.http import HttpResponse

import json
from sourceai.model.face.custom import base64util

def custom_facec_regiter(request):

    try:
        base64 =request.POST.get('imgFile')
        cardid = request.POST.get('cardid')
        print(base64)
        print(cardid)
        base64 = base64.replace(' ','+')

        # base64 = 'iVBORw0KGgoAAAANSUhEUgAAAZMAAAF...'
        # cardid = 'z_test_6'
        path = 'static/facedb/' + cardid + '.jpg'
        res =base64util.base2picture(base64,path)
    except Exception as e:
        print('出现问题',e)
        res = ''

    return HttpResponse(json.dumps({"res": res}))

步骤二：soft863ai/urls.py中新增如下内容：

from sourceai.controller import face_view
    path('faceregister', face_view.custom_facec_regiter),

全部内容如下：

"""soft863ai URL Configuration

The `urlpatterns` list routes URLs to views. For more information please see:
    https://docs.djangoproject.com/en/3.2/topics/http/urls/
Examples:
Function views
    1. Add an import:  from my_app import views
    2. Add a URL to urlpatterns:  path('', views.home, name='home')
Class-based views
    1. Add an import:  from other_app.views import Home
    2. Add a URL to urlpatterns:  path('', Home.as_view(), name='home')
Including another URLconf
    1. Import the include() function: from django.urls import include, path
    2. Add a URL to urlpatterns:  path('blog/', include('blog.urls'))
"""
from django.contrib import admin
from django.urls import path
from sourceai import views
from sourceai.controller import imgc_view
from sourceai.controller import nlp_view
from sourceai.controller import chat_view
from sourceai.controller import draw_view
from sourceai.controller import ppt_view
from sourceai.controller import face_view

urlpatterns = [
    path('admin/', admin.site.urls),
    path('first', views.first_index),
    path('firstvalue', views.first_value),
    path('urltotext', views.url_to_text),
    path('ocrurl', views.ocr_index_url),
    path('imgtotext', views.img_to_text),
    path('ocrindex', views.ocr_index),
    path('imgc', imgc_view.imgc_index),
    path('imgcto', imgc_view.img_to_text),
    path('poem', nlp_view.poem_index),
    path('poemto', nlp_view.poem_to_create),
    path('chat', chat_view.chat_index),
    path('chatto', chat_view.chat_to_create),
    path('sparkchat', chat_view.spark_index),
    path('sparkchatto', chat_view.spark_to_create),
    path('baichuan2chat', chat_view.baichuan2_index),
    path('baichuan2chatto', chat_view.baichuan2_to_create),
    path('sparkdrawindex', draw_view.sparkdraw_index),
    path('sparkdrawto', draw_view.sparkdraw_to_create),
    path('sparkppt', ppt_view.ppt_index),
    path('sparkpptto', ppt_view.ppt_to_create),
    path('faceregister', face_view.custom_facec_regiter),
]

步骤三：soft863ai/settings.py中注释MIDDLEWARE的'django.middleware.csrf.CsrfViewMiddleware'

步骤四：创建static/facedb文件夹，后续将人脸库中已有图片放置其中

步骤五：启动项目

步骤六：打开Postman进行测试

浏览器输入https://www.bchrt.com/tools/image-to-base64/

选择一张图片加载后获取base64编码

打开postman，输入http://127.0.0.1:8000/faceregister

设置imgFile与cardid参数化

需求二：编写代码实现“人脸库检索”服务

post http://127.0.0.1:8000/facedbsearch 请求参数：imgFile：图片base64编码

步骤一：在sourceai/controller/face_view.py文件中添加内容如下：

from django.shortcuts import render
from django.http import HttpResponse

import json
from sourceai.model.face.custom import facesearch
from sourceai.model.face.custom import base64util

# 返回图像识别结果
def custom_facec_search(request):

    dict_data = {}
    try:
        base64 =request.POST.get('imgFile')
        base64 = base64.replace(' ', '+')
        import time
        path = 'static/test/' +str(time.time())  + '.jpg'

        res =base64util.base2picture(base64,path)
        if res:
            train = 'static/facedb/'
            result = facesearch.start(train, path)
            dict_data['flag'] = 1
            dict_data['msg'] = '成功'
            dict_data['cardid'] = result
        else:
            result=''
            dict_data['flag'] = 0
            dict_data['msg'] = '未匹配到人脸数据'
            dict_data['cardid'] = result

    except Exception as e:
        print('出现问题',e)
        dict_data['flag'] = 0
        dict_data['msg'] = '异常'
        dict_data['cardid'] = ''

    return HttpResponse(json.dumps({"res": dict_data}))

步骤二：soft863ai/urls.py中新增如下内容：

from sourceai.controller import face_view
    path('facedbsearch', face_view.custom_facec_search),

步骤三：soft863ai/settings.py中注释MIDDLEWARE的'django.middleware.csrf.CsrfViewMiddleware'

步骤四：确保有static/facedb文件夹，将人脸库中已有图片放置其中

步骤五：创建static/test文件夹，放置人脸搜索原始图片

步骤六：启动项目

步骤七：打开Postman进行测试

人脸库搜索http://127.0.0.1:8000/facedbsearch

需求三：编写代码实现“人脸库图片列表”服务

get http://127.0.0.1:8000/facelist

步骤一：在sourceai/controller/face_view.py文件中添加内容如下：

def custom_facec_list(request):
    # if request.POST:
    list = []
    try:
        import os
        path = 'static/facedb/'
        datanames = os.listdir(path)
        list = []
        for i in datanames:
            list.append(i)
        print(list)
    except Exception as e:
        print('出现问题',e)

    return HttpResponse(json.dumps({"res": list}))

步骤二：soft863ai/urls.py中新增如下内容：

from sourceai.controller import face_view
    path('facelist', face_view.custom_facec_list),

步骤三：启动项目

步骤四：打开Postman进行测试

人脸库图片列表：http://127.0.0.1:8000/facelist

需求四：编写代码实现“人脸库某个图片获取”服务

get http://127.0.0.1:8000/faceget 请求参数：cardid：身份证ID

步骤一：在sourceai/controller/face_view.py文件中添加内容如下：

def custom_facec_get(request):
    # if request.POST:
    dict_data = {}
    try:
        name = request.GET.get('cardid')
        path = 'facedb/'+name+'.jpg'
        strbase64 = base64util.pic2base64(path)
        dict_data = {}
        dict_data['path'] = path
        dict_data['img'] = strbase64

    except Exception as e:
        print('出现问题',e)

    return HttpResponse(json.dumps({"res": dict_data}))

步骤二：soft863ai/urls.py中新增如下内容：

path('faceget', face_view.custom_facec_get),

步骤三：启动项目

步骤四：打开Postman进行测试

人脸库某张图片获取：

获取到的img内容可以复制到https://www.bchrt.com/tools/base64-to-image/，进行图片还原

FAQ手册

1、执行报错AttributeError: module 'cv2' has no attribute 'gapi_wip_gst_GStreamerPipeline'

解决办法

pip install --user --upgrade opencv-python -i https://pypi.tuna.tsinghua.edu.cn/simple

2、incompatible function arguments.

TypeError: call(): incompatible function arguments. The following argument types are supported: 1. (self: dlib.fhog_object_detector, image: array, upsample_num_times: int=0) -> dlib.rectangles

Invoked with: <dlib.fhog_object_detector object at 0x00000242BB795FB8>, None, 1

解决方案：查看输入路径是否存在

3、Unable to onen C:NUsers veny lib site-dackaces face recognition modelsimodels shape-predictor-68-face-landmarks.dat

解决方案：Python路径中不要有中文