Face_Photo.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Face_Photo.ipynb",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "m5E9lr8W02zA"
      },
      "source": [
        "!pip install pytesseract\n",
        "!sudo apt-get install tesseract-ocr-ind"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mgKK43G208p_"
      },
      "source": [
        "!pip install PassportEye\n",
        "# tessaract should also be connected in command line"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NG1nKanlylfH"
      },
      "source": [
        "%matplotlib inline\n",
        "from passporteye.mrz.image import MRZPipeline\n",
        "from matplotlib import pyplot as plt\n",
        "from random import randint\n",
        "import cv2\n",
        "import sys, os, re\n",
        "import traceback\n",
        "import pytesseract\n",
        "import datetime\n",
        "import pandas as pd\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "00CdhRfoy4My"
      },
      "source": [
        "##############    roots     ##################\n",
        "\n",
        "pytesseract.pytesseract.tesseract_cmd = r'/usr/bin/tesseract'\n",
        "cv2_base_dir = os.path.dirname(os.path.abspath(cv2.__file__))\n",
        "CASCADE = os.path.join(cv2_base_dir, 'data/haarcascade_frontalface_default.xml')\n",
        "FACE_CASCADE=cv2.CascadeClassifier(CASCADE)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u17l4W2Q9S-N"
      },
      "source": [
        "#######################  Create   ########################\n",
        "\n",
        "if not os.path.isdir('Face'):\n",
        "  !mkdir '/content/Face'\n",
        "\n",
        "\n",
        "dataset = pd.DataFrame(columns = [\"firstname\",\"surname\", 'nationality', 'passport_type', 'date_of_birth', 'ID_number', 'gender', 'expiration_date', 'age'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t-BIbNyuL86S"
      },
      "source": [
        "######################    Upload Image   ######################\n",
        "from google.colab import files\n",
        "\n",
        "files.upload()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3b_5BT0HdB7N"
      },
      "source": [
        "image = '/content/passport2.jpeg'"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Kdmx0fy-zhAq"
      },
      "source": [
        "#################################   Text Extraction   ########################################\n",
        "\n",
        "p = MRZPipeline(image)\n",
        "mrz_data = p.result.to_dict()\n",
        "\n",
        "if mrz_data['names']:\n",
        "  firstname =     mrz_data['names'].split()[0]\n",
        "else:\n",
        "  firstname = ''\n",
        "surname   =     mrz_data['surname']\n",
        "nationality =   mrz_data['nationality']\n",
        "passport_type = re.sub(\"[^A-Z^.]\", \"\", mrz_data['type'])\n",
        "date_of_birth = pd.to_datetime(re.sub(\"[^0-9^.]\", \"\", mrz_data['date_of_birth']), format='%y%m%d', errors='coerce')\n",
        "ID_number =     mrz_data['number'].strip().replace('<', '')\n",
        "gender  =       mrz_data['sex']\n",
        "expiration_date = mrz_data['expiration_date']\n",
        "\n",
        "birth = int(re.sub(\"[^0-9^.]\", \"\", mrz_data['date_of_birth'])[:2])\n",
        "current = int(str(datetime.datetime.now().year)[2:])\n",
        "\n",
        "if birth >  current:\n",
        "  age = current - birth + 100\n",
        "else:\n",
        "  age = current - birth"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aRgelTGm_oZQ"
      },
      "source": [
        "####################     Save     ######################\n",
        "\n",
        "dataset.loc[dataset.shape[0]] = [firstname, surname, nationality, passport_type, \n",
        "                                 date_of_birth, ID_number, gender, expiration_date, age]\n",
        "\n",
        "dataset.to_csv('dataset.csv', index=False)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 142
        },
        "id": "3NrQ4uDtK8wI",
        "outputId": "0aad5427-04de-47d3-d234-75625ad02910"
      },
      "source": [
        "dataset"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>firstname</th>\n",
              "      <th>surname</th>\n",
              "      <th>nationality</th>\n",
              "      <th>passport_type</th>\n",
              "      <th>date_of_birth</th>\n",
              "      <th>ID_number</th>\n",
              "      <th>gender</th>\n",
              "      <th>expiration_date</th>\n",
              "      <th>age</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>ANGELA</td>\n",
              "      <td>ST HELENA SPECIMEN</td>\n",
              "      <td>GBR</td>\n",
              "      <td>P</td>\n",
              "      <td>1988-09-11</td>\n",
              "      <td>760641560</td>\n",
              "      <td>F</td>\n",
              "      <td>250310</td>\n",
              "      <td>33</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>SUNDAL</td>\n",
              "      <td>SHAMIM</td>\n",
              "      <td>PAK</td>\n",
              "      <td>P</td>\n",
              "      <td>1996-09-05</td>\n",
              "      <td>RZ5153771</td>\n",
              "      <td>F</td>\n",
              "      <td>240228</td>\n",
              "      <td>25</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>SEYMOOR</td>\n",
              "      <td>PIPP</td>\n",
              "      <td>OGB</td>\n",
              "      <td>P</td>\n",
              "      <td>NaT</td>\n",
              "      <td>T6TO7TEOP</td>\n",
              "      <td>A</td>\n",
              "      <td>N15051</td>\n",
              "      <td>71</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "  firstname             surname nationality  ... gender expiration_date age\n",
              "1    ANGELA  ST HELENA SPECIMEN         GBR  ...      F          250310  33\n",
              "2    SUNDAL              SHAMIM         PAK  ...      F          240228  25\n",
              "3   SEYMOOR                PIPP         OGB  ...      A          N15051  71\n",
              "\n",
              "[3 rows x 9 columns]"
            ]
          },
          "metadata": {},
          "execution_count": 51
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7qPHIaJFb2vs",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 268
        },
        "outputId": "04f37ce7-6cdd-4ae3-d666-8403547ba8e3"
      },
      "source": [
        "################    Photo     ########################\n",
        "\n",
        "def detect_faces(image_path, name='', display=True):\n",
        "  image=cv2.imread(image_path)\n",
        "  image_grey=cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)\n",
        "  faces = list(FACE_CASCADE.detectMultiScale(image_grey, \n",
        "                                             scaleFactor=1.16, minNeighbors=5, minSize=(25,25), flags=0))\n",
        "\n",
        "  if faces is 0:\n",
        "    print(\"No faces found\")\n",
        "    exit()\n",
        "\n",
        "\n",
        "  x, y, w, h = faces[0]\n",
        "  face = image[y-10:y+h+10,x-10:x+w+10]\n",
        "  os.chdir('/content/Face/')\n",
        "  \n",
        "  if name:\n",
        "    cv2.imwrite(name+\".jpg\",face)\n",
        "  else:\n",
        "    cv2.imwrite(str(randint(0,10000))+\".jpg\",face)\n",
        "  os.chdir(\"../\")\n",
        "  cv2.rectangle(image,(x,y),(x+w,y+h),(255, 255,0),2)\n",
        "  \n",
        "  if display:\n",
        "    plt.imshow(face)\n",
        "    plt.show()\n",
        "\n",
        "\n",
        "detect_faces(image, firstname)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}