Image/Emotion_Recognition(Pre_processing)FER2013.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Emotion Recognition(Pre processing)FER2003.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/spazewalker/FER_Doggomaniacs/blob/master/Emotion_Recognition(Pre_processing)FER2013.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5K88Xb_PJlo2",
        "colab_type": "text"
      },
      "source": [
        "# Download Haar cascades."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "a1zawdCiOvIP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 228
        },
        "outputId": "ac8501d0-1abf-4fd3-e860-4d68cc33cd2c"
      },
      "source": [
        "!wget https://raw.githubusercontent.com/opencv/opencv/master/data/haarcascades/haarcascade_eye.xml"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-07-09 15:42:55--  https://raw.githubusercontent.com/opencv/opencv/master/data/haarcascades/haarcascade_eye.xml\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.0.133, 151.101.64.133, 151.101.128.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.0.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 341406 (333K) [text/plain]\n",
            "Saving to: ‘haarcascade_eye.xml.2’\n",
            "\n",
            "\rhaarcascade_eye.xml   0%[                    ]       0  --.-KB/s               \rhaarcascade_eye.xml 100%[===================>] 333.40K  --.-KB/s    in 0.06s   \n",
            "\n",
            "2020-07-09 15:42:55 (5.01 MB/s) - ‘haarcascade_eye.xml.2’ saved [341406/341406]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "c4A7fJEMOv_G",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 228
        },
        "outputId": "154315e2-cd82-4a44-f8ef-da9ccc44eecd"
      },
      "source": [
        "!wget https://raw.githubusercontent.com/opencv/opencv/master/data/haarcascades/haarcascade_frontalface_default.xml"
      ],
      "execution_count": 2,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-07-09 15:43:00--  https://raw.githubusercontent.com/opencv/opencv/master/data/haarcascades/haarcascade_frontalface_default.xml\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 151.101.0.133, 151.101.64.133, 151.101.128.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|151.101.0.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 930127 (908K) [text/plain]\n",
            "Saving to: ‘haarcascade_frontalface_default.xml.2’\n",
            "\n",
            "\r          haarcasca   0%[                    ]       0  --.-KB/s               \rhaarcascade_frontal 100%[===================>] 908.33K  --.-KB/s    in 0.09s   \n",
            "\n",
            "2020-07-09 15:43:00 (9.65 MB/s) - ‘haarcascade_frontalface_default.xml.2’ saved [930127/930127]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VvkPj0OXJJzd",
        "colab_type": "text"
      },
      "source": [
        "# Importing Packages."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PKh-PSMZyGPp",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 667
        },
        "outputId": "f1ff14dd-d189-473d-ba9a-1f15f95081dc"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "error",
          "ename": "KeyboardInterrupt",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/ipykernel/kernelbase.py\u001b[0m in \u001b[0;36m_input_request\u001b[0;34m(self, prompt, ident, parent, password)\u001b[0m\n\u001b[1;32m    728\u001b[0m             \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 729\u001b[0;31m                 \u001b[0mident\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreply\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msession\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstdin_socket\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    730\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/jupyter_client/session.py\u001b[0m in \u001b[0;36mrecv\u001b[0;34m(self, socket, mode, content, copy)\u001b[0m\n\u001b[1;32m    802\u001b[0m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 803\u001b[0;31m             \u001b[0mmsg_list\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msocket\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecv_multipart\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    804\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mzmq\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mZMQError\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/zmq/sugar/socket.py\u001b[0m in \u001b[0;36mrecv_multipart\u001b[0;34m(self, flags, copy, track)\u001b[0m\n\u001b[1;32m    474\u001b[0m         \"\"\"\n\u001b[0;32m--> 475\u001b[0;31m         \u001b[0mparts\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrecv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mflags\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrack\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mtrack\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    476\u001b[0m         \u001b[0;31m# have first part already, only loop while more to receive\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32mzmq/backend/cython/socket.pyx\u001b[0m in \u001b[0;36mzmq.backend.cython.socket.Socket.recv\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;32mzmq/backend/cython/socket.pyx\u001b[0m in \u001b[0;36mzmq.backend.cython.socket.Socket.recv\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;32mzmq/backend/cython/socket.pyx\u001b[0m in \u001b[0;36mzmq.backend.cython.socket._recv_copy\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/zmq/backend/cython/checkrc.pxd\u001b[0m in \u001b[0;36mzmq.backend.cython.checkrc._check_rc\u001b[0;34m()\u001b[0m\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: ",
            "\nDuring handling of the above exception, another exception occurred:\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-3-d5df0069828e>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mgoogle\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcolab\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mdrive\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mdrive\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmount\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'/content/drive'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/google/colab/drive.py\u001b[0m in \u001b[0;36mmount\u001b[0;34m(mountpoint, force_remount, timeout_ms, use_metadata_server)\u001b[0m\n\u001b[1;32m    240\u001b[0m       \u001b[0mauth_prompt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0md\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmatch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgroup\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m'\\nEnter your authorization code:\\n'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    241\u001b[0m       \u001b[0;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfifo\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'w'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mfifo_file\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 242\u001b[0;31m         \u001b[0mfifo_file\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwrite\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0m_getpass\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgetpass\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mauth_prompt\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0;34m'\\n'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    243\u001b[0m       \u001b[0mwrote_to_fifo\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    244\u001b[0m   \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mwrote_to_fifo\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/ipykernel/kernelbase.py\u001b[0m in \u001b[0;36mgetpass\u001b[0;34m(self, prompt, stream)\u001b[0m\n\u001b[1;32m    685\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_parent_ident\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    686\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_parent_header\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 687\u001b[0;31m             \u001b[0mpassword\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    688\u001b[0m         )\n\u001b[1;32m    689\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.6/dist-packages/ipykernel/kernelbase.py\u001b[0m in \u001b[0;36m_input_request\u001b[0;34m(self, prompt, ident, parent, password)\u001b[0m\n\u001b[1;32m    732\u001b[0m             \u001b[0;32mexcept\u001b[0m \u001b[0mKeyboardInterrupt\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    733\u001b[0m                 \u001b[0;31m# re-raise KeyboardInterrupt, to truncate traceback\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 734\u001b[0;31m                 \u001b[0;32mraise\u001b[0m \u001b[0mKeyboardInterrupt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    735\u001b[0m             \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    736\u001b[0m                 \u001b[0;32mbreak\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "PTbVuGbhYyHC",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import cv2\n",
        "import csv\n",
        "from math import sqrt\n",
        "import scipy.misc\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import tensorflow as tf\n",
        "import matplotlib.pyplot as plt\n",
        "eye_cascade = cv2.CascadeClassifier('/content/haarcascade_eye.xml')\n",
        "face_cascade = cv2.CascadeClassifier('/content/haarcascade_frontalface_default.xml')\n",
        "emotions = [\"Angry\", \"Disgust\", \"Fear\", \"Happy\", \"Sad\", \"Surprise\", \"Neutral\"]"
      ],
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "T9V_NngdJOxY",
        "colab_type": "text"
      },
      "source": [
        "# Loading Data.\n",
        "Change the location of zip file."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JTrhZJpS_jLL",
        "colab_type": "text"
      },
      "source": [
        "Loading csv file."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iJhOBhTzE3Rd",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "data = pd.read_csv(\"/content/drive/My Drive/FER/fer2013.zip\",compression = 'zip',sep = ',')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tiTAtYjV_pU8",
        "colab_type": "text"
      },
      "source": [
        "Loading Y part"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "P3SXR6U1F-_P",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "6f4202d6-d053-4827-e585-c10e7b819486"
      },
      "source": [
        "y = np.array(data['emotion'])\n",
        "print(y)\n",
        "y = [y]\n",
        "y = np.array(y)\n",
        "y= y.T"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[0 0 2 ... 0 3 2]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3IIhR3veys2o",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "1feefbe5-e782-419c-e705-0e366b5c92ed"
      },
      "source": [
        "print(y.shape)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(35887, 1)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Od3SsVK-I_WG",
        "colab_type": "text"
      },
      "source": [
        "#Preprocessing Starts here."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cG2vU-Y_B4ra",
        "colab_type": "text"
      },
      "source": [
        "Converting the array of pixels to image.\n",
        "1. Converted to a numpy array\n",
        "2. Then resized to higher dimension for processing."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jwe6nAg6B45a",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def modification(image):\n",
        "  image = np.mat(image).reshape(48, 48,1) \n",
        "  image = np.uint8(image)\n",
        "  image = cv2.resize(image,(480,480), interpolation = cv2.INTER_AREA)\n",
        "  return image"
      ],
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JohBikW-DlXx",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "33a13f55-2910-4913-c4ee-e4dd2d107149"
      },
      "source": [
        "image  = modification(data['pixels'][5])\n",
        "plt.imshow(image)\n",
        "plt.show()"
      ],
      "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": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nDSRtpjSAZ04",
        "colab_type": "text"
      },
      "source": [
        "Rotation:\n",
        "1. Finding the face and then eye using Haar cascades.\n",
        "2. Upon detection, finding the angle made by line joining the eyes and horizontal.\n",
        "3. Then transformation matrix if found.\n",
        "4. Finally cropping is done."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FFFVewoNAGvM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def rotate(image):\n",
        "  l1=[]\n",
        "  faces = face_cascade.detectMultiScale(image)\n",
        "  if(len(faces)==1):\n",
        "    for (x,y,w,h) in faces:\n",
        "      face_detected = image[y:y+h,x:x+h]\n",
        "      eyes = eye_cascade.detectMultiScale(face_detected)\n",
        "      for (ex,ey,ew,eh) in eyes:\n",
        "          l1.append((ex+ew/2,ey+eh/2))\n",
        "      \n",
        "      if(len(l1)==2):\n",
        "        dist_x = l1[1][0]-l1[0][0]\n",
        "        dist_y = l1[1][1] - l1[0][1]\n",
        "        if dist_x<0:\n",
        "          dist_y = -dist_y\n",
        "        dist_x = np.abs(dist_x)     \n",
        "        angle = np.arctan(dist_y/dist_x) * 180/3.14\n",
        "        M = cv2.getRotationMatrix2D((240, 240), angle, 1.0)\n",
        "        image = cv2.warpAffine(image, M,(480,480))\n",
        "  return image"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xWo-OWivD0xP",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "149d4d56-f510-46be-cedc-750da788d490"
      },
      "source": [
        "image = rotate(image)\n",
        "plt.imshow(image)\n",
        "plt.show()"
      ],
      "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": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MS3X95oxA9YA",
        "colab_type": "text"
      },
      "source": [
        "Cropping:\n",
        "1. Face is detected and then image is cropped to the face detected.\n",
        "2. Now, eyes are detected and the distance between them is calculated.\n",
        "3. Using x_factor and y_factor with distance, crop the image."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "42MJckxsA9ob",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def crop(image,x_factor=2.1,y_factor=3.2):\n",
        "  l1=[]\n",
        "  faces = face_cascade.detectMultiScale(image)\n",
        "  if(len(faces)==1):\n",
        "    for (x,y,w,h) in faces:\n",
        "      image = image[y:y+h,x:x+h]\n",
        "\n",
        "  eyes = eye_cascade.detectMultiScale(image)\n",
        "  for (ex,ey,ew,eh) in eyes:\n",
        "      l1.append((ex+ew/2,ey+eh/2))\n",
        "  if(len(l1)==2):\n",
        "    dist = np.sqrt((l1[0][0]-l1[1][0])**2+(l1[0][1]-l1[1][1])**2)\n",
        "    center_x = image.shape[1]//2\n",
        "    center_y = image.shape[0]//2\n",
        "    shift_x = int(dist*x_factor)//2\n",
        "    shift_y = int(dist*y_factor)//2\n",
        "    start_x = center_x - shift_x\n",
        "    start_x = max(start_x,0)\n",
        "    end_x = center_x+shift_x\n",
        "    end_x = min(end_x,image.shape[1])\n",
        "    start_y = center_y - shift_y\n",
        "    start_y = max(start_y,0)\n",
        "    end_y = center_y + shift_y\n",
        "    end_y = min(end_y,image.shape[0])\n",
        "    image = image[start_y:end_y,start_x:end_x]\n",
        "  return image"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KheXi0JZEAiN",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "ec203521-32de-453b-a563-6af0e980b7d7"
      },
      "source": [
        "image = crop(image)\n",
        "plt.imshow(image)\n",
        "plt.show()"
      ],
      "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": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_nVWvAFag-YM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def preProcessing(images,x_factor=2.1,y_factor=3.2):\n",
        "  x_images= []\n",
        "  for i in images:\n",
        "    image = i\n",
        "    image = modification(image)\n",
        "    image = rotate(image)\n",
        "    image = crop(image,x_factor,y_factor)\n",
        "    #Smoothing\n",
        "    image = cv2.bilateralFilter(image, 4, 15, 15)\n",
        "    #Resizing\n",
        "    image = cv2.resize(image,(48,48), interpolation = cv2.INTER_AREA)\n",
        "    x_images.append(image)\n",
        "  x_images= np.array(x_images)\n",
        "  return x_images\n"
      ],
      "execution_count": 6,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1JWZyE-e92PO",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x = preProcessing(data['pixels'])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QnoqrKN-5-SC",
        "colab_type": "text"
      },
      "source": [
        "To show the progress."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wjgp0oVC4sZP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def progress(percent=0, width=30):\n",
        "  left = int(width * percent) // 100\n",
        "  right = width - left\n",
        "  print('\\r[', '#' * left, ' ' * right, ']',\n",
        "        f' {percent:.2f}%',\n",
        "        sep='', end='', flush=True)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jC_KCSCo2Ao9",
        "colab_type": "text"
      },
      "source": [
        "Since preprocessing takes too much time. It is beneficial to save the preprocessed data in for of CSV and then use it when required. This function do the same."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Xigaz_X91_vP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def savePreprocessed(data,x_factor=2.1,y_factor=3.2):\n",
        "  images = data['pixels']\n",
        "  y = np.array(data['emotion'])\n",
        "  x_images= []\n",
        "  y_emotion = []\n",
        "  total = images.shape[0]\n",
        "  for i in range(total):\n",
        "    progress(i*100/total)\n",
        "    image = images[i]\n",
        "    image = modification(image)\n",
        "    # plt.imshow(image)\n",
        "    # plt.show()\n",
        "    image = rotate(image)\n",
        "    image = crop(image,x_factor,y_factor)\n",
        "    #Smoothing\n",
        "    image = cv2.bilateralFilter(image, 4, 15, 15)\n",
        "    #Resizing\n",
        "    image = cv2.resize(image,(48,48), interpolation = cv2.INTER_AREA)\n",
        "    flattened = image.flatten()\n",
        "    flattened = list(flattened)\n",
        "    flattened = ' '.join(map(str, flattened))\n",
        "    x_images.append(flattened)\n",
        "    y_emotion.append(y[i])\n",
        "  data1 = {'emotion':y_emotion,'pixels':x_images}\n",
        "  df = pd.DataFrame(data1)\n",
        "  df.to_csv('/content/sample_data/preprocessed.csv',index = False)\n",
        "  return"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AKpLTDkr4itx",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "c4adde45-5b9c-48c2-fad0-8b977ca2eac2"
      },
      "source": [
        "savePreprocessed(data[0:100])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[############################# ] 99.00%"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vDfeH8uR_5o8",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def get_faces(image,x_factor=2.1,y_factor=3.2):\n",
        "  height,width,_ = image.shape\n",
        "  factor = 720/height\n",
        "  image = cv2.resize(image,(int(width*factor),int(height*factor)),interpolation = cv2.INTER_AREA)\n",
        "  faces = face_cascade.detectMultiScale(image)\n",
        "  face_list = []\n",
        "  labels = []\n",
        "  for (x,y,w,h) in faces:\n",
        "    x  = x - int(0.2*w)\n",
        "    y = y - int(0.2*h)\n",
        "    w = int(1.2*w)\n",
        "    h = int(1.2*w)\n",
        "    face = image[y:y+h,x:x+h]\n",
        "    face = rotate(face)\n",
        "    face = crop(face,x_factor,y_factor)\n",
        "    #Smoothing\n",
        "    face = cv2.bilateralFilter(face, 4, 15, 15)\n",
        "    #Resizing\n",
        "    face = cv2.resize(face,(48,48), interpolation = cv2.INTER_AREA)\n",
        "    face_list.append(face)\n",
        "    labels.append((int(x/factor),int(y/factor)))\n",
        "  return face_list,labels "
      ],
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hQxTk7IIDlfV",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "image = plt.imread(\"/content/sample_data/person2.jpeg\")\n",
        "faces,labels = get_faces(image)\n",
        "#labels is if you want to label each face with emotion.\n",
        "for face,label in zip(faces,labels):\n",
        "  plt.imshow(face)\n",
        "  plt.show()\n",
        "  print(label)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MXo8mc3XJblY",
        "colab_type": "text"
      },
      "source": [
        "# Dividing Data between train,dev,test."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "81M1y8nQUVr9",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def partition(x,y):\n",
        "  size = y.shape[0]\n",
        "  sizetrain = (8*size)//10\n",
        "  size2 = (10*size)//100\n",
        "  order=np.random.permutation(size)\n",
        "  x=x[order]\n",
        "  y=y[order]\n",
        "  train_x = x[0:sizetrain]\n",
        "  train_y = y[0:sizetrain]\n",
        "  dev_x = x[sizetrain:sizetrain+size2]\n",
        "  dev_y = y[sizetrain:sizetrain+size2]\n",
        "  test_x = x[sizetrain+size2:size]\n",
        "  test_y = y[sizetrain+size2:size]\n",
        "  return train_x,train_y,dev_x,dev_y,test_x,test_y"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "dSSqkATwFJRl"
      },
      "source": [
        "# Class Implementation\n",
        "Currently this is not working."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "WTuC8dt-FJR5",
        "colab": {}
      },
      "source": [
        "class PreProcessing:\n",
        "  def __init__(self,images,x_factor = 2.1,rotate = True, crop = True, smoothing = True, resize = True, y_factor = 3.2,smoothing_params=(4, 15, 15),resize_shape=(48,48)):\n",
        "    self.images = images\n",
        "    self.total = images.shape[0]\n",
        "    self.face = 0\n",
        "    self.eye = 0\n",
        "    self.face_count = 0\n",
        "    self.eye_count = 0\n",
        "    self.image = 0\n",
        "    self.face_detected = 0\n",
        "    self.eye_detected = 0\n",
        "    self.rotate_state = True\n",
        "    self.crop_state =True\n",
        "    self.smoothing_state = True\n",
        "    self.resize_state = True\n",
        "    self.x_factor = x_factor\n",
        "    self.y_factor = y_factor\n",
        "    self.smoothing_params = smoothing_params\n",
        "    self.resize_shape = resize_shape\n",
        "\n",
        "  def modification(self,image):\n",
        "    image = np.mat(image).reshape(48,48,1)\n",
        "    image = np.uint8(image)\n",
        "    image = cv2.resize(image,(480,480), interpolation = cv2.INTER_AREA)\n",
        "    self.image = image\n",
        "  def detection(self):\n",
        "    faces = face_cascade.detectMultiScale(self.image)\n",
        "    if(len(faces)==1):\n",
        "      self.face = 1\n",
        "      self.face_count+=1\n",
        "      for (x,y,w,h) in faces:\n",
        "        self.face_detected = self.image[y:y+h,x:x+h]\n",
        "        eyes = eye_cascade.detectMultiScale(self.face_detected)\n",
        "        for (ex,ey,ew,eh) in eyes:\n",
        "          l1.append((ex+ew/2,ey+eh/2))\n",
        "        if(len(l1)==2):\n",
        "          self.eye = 1\n",
        "          self.eye_count+=1\n",
        "          self.eye_detected = l1\n",
        "        else:\n",
        "          self.eye = 0\n",
        "    else:\n",
        "      self.face = 0\n",
        "  \n",
        "  def rotation(self):\n",
        "    self.detection()\n",
        "    if self.eye == 1:\n",
        "      dist_x = self.eye_detected[1][0]-self.eye_detected[0][0]\n",
        "      dist_y = self.eye_detected[1][1] - self.eye_detected[0][1]\n",
        "      if dist_x<0:\n",
        "        dist_y = -dist_y\n",
        "      dist_x = np.abs(dist_x)\n",
        "      angle = np.arctan(dist_y/dist_x) * 180/3.14\n",
        "      M = cv2.getRotationMatrix2D((240, 240), angle, 1.0)\n",
        "      self.image = cv2.warpAffine(self.image, M,(480,480))\n",
        "  def crop(self):\n",
        "    self.detection()\n",
        "    if self.face == 1:\n",
        "      self.image = self.face_detected\n",
        "      if self.eye ==1:\n",
        "        dist = np.sqrt((self.eye_detected[0][0]-self.eye_detected[1][0])**2+(self.eye_detected[0][1]-self.eye_detected[1][1])**2)\n",
        "        center_x = self.image.shape[1]//2\n",
        "        center_y = self.image.shape[0]//2\n",
        "        shift_x = int(dist*self.x_factor)//2\n",
        "        shift_y = int(dist*self.y_factor)//2\n",
        "        start_x = center_x - shift_x\n",
        "        start_x = max(start_x,0)\n",
        "        end_x = center_x+shift_x\n",
        "        end_x = min(end_x,self.image.shape[1])\n",
        "        start_y = center_y - shift_y\n",
        "        start_y = max(start_y,0)\n",
        "        end_y = center_y + shift_y\n",
        "        end_y = min(end_y,self.image.shape[0])\n",
        "        self.image = self.image[start_y:end_y,start_x:end_x]\n",
        "  def smoothing(self):\n",
        "    self.image = cv2.bilateralFilter(self.image, self.smoothing_params[0], self.smoothing_params[1], self.smoothing_params[2])\n",
        "  \n",
        "\n",
        "  def resize(self):\n",
        "    self.image = cv2.resize(self.image,self.resize_shape, interpolation = cv2.INTER_AREA)\n",
        "  def getImage(self):\n",
        "    return self.image\n",
        "  def summary(self):\n",
        "    print(\"Total Images:\",self.total)\n",
        "    if (self.rotate_state==True and self.crop_state==True):\n",
        "      print(\"Total Face Detected: \",self.face_count//2)\n",
        "      print(\"Total Eye Detected:\",self.eye_count//2)\n",
        "    else:\n",
        "      print(\"Total Face Detected: \",self.face_count)\n",
        "      print(\"Total Eye Detected:\",self.eye_count)\n",
        "  \n",
        "  def main(self):\n",
        "    x=[]\n",
        "    for i in self.images:\n",
        "      self.modification(i)\n",
        "      plt.imshow(self.image)\n",
        "      plt.show()\n",
        "      if self.rotate_state==True:\n",
        "        self.rotation()\n",
        "      print(\"After rotation\")\n",
        "      plt.imshow(self.image)\n",
        "      plt.show()\n",
        "      if self.crop_state == True:\n",
        "        self.crop()\n",
        "      print(\"After cropping\")\n",
        "      plt.imshow(self.image)\n",
        "      plt.show()\n",
        "      if self.smoothing_state == True:\n",
        "        self.smoothing()\n",
        "      print(\"After smoothing\")\n",
        "      plt.imshow(self.image)\n",
        "      plt.show()\n",
        "      if self.resize_state ==True:\n",
        "        self.resize()\n",
        "      print(\"After resizing\")\n",
        "      plt.imshow(self.image)\n",
        "      plt.show()\n",
        "      x.append(self.image)\n",
        "    x = np.array(x)\n",
        "    return x\n"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}