Captcha_Solver.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Captcha_Solver.ipynb",
      "version": "0.3.2",
      "provenance": [],
      "collapsed_sections": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "metadata": {
        "id": "bH-GVGE7Vhs5",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "# Captcha Solver"
      ]
    },
    {
      "metadata": {
        "id": "Cx1LK1NAEcnI",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import cv2\n",
        "import keras\n",
        "import numpy as np\n",
        "from matplotlib import pyplot as plt"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "pXCj86wRh0PK",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "%%capture\n",
        "!unzip generated_captcha_images.zip"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "aiBIOMqhWDd6",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Data Processing"
      ]
    },
    {
      "metadata": {
        "id": "cAbh94sqWKAF",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Extracting Single letters from Captcha."
      ]
    },
    {
      "metadata": {
        "id": "7AFYg2Hsk_lA",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "import os \n",
        "import os.path, glob, imutils"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "JS0lvfQulKP3",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "captcha_image_folder = \"generated_captcha_images\"\n",
        "output_folder = \"extracted_letter_images\""
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "lecPCg9Pl6ON",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# Getting a list of all the captcha images we need to process\n",
        "captcha_image_files = glob.glob(os.path.join(captcha_image_folder,'*'))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "HfV8O0Fymor0",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "counts = {}"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ZWdQpJp_mvWO",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "# loop over the image paths\n",
        "for i,image_file in enumerate(captcha_image_files):\n",
        "    print(\"[INFO] processing image {}/{}\".format(i + 1, len(captcha_image_files)))\n",
        "    \n",
        "    # grab the base filename as the text\n",
        "    file_name = os.path.basename(image_file)\n",
        "    correct_text = os.path.splitext(file_name)[0]\n",
        "    \n",
        "    # Load the image\n",
        "    image = cv2.imread(image_file,cv2.IMREAD_GRAYSCALE)\n",
        "    \n",
        "    # Addding some extra padding around the image\n",
        "    image = cv2.copyMakeBorder(image,8,8,8,8,cv2.BORDER_REPLICATE)\n",
        "    \n",
        "    # threshold the image (convert it to pure black and white)\n",
        "    thresh = cv2.threshold(image, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]\n",
        "    \n",
        "    #Finding the contours\n",
        "    contours = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n",
        "    \n",
        "    contours = contours[0] if imutils.is_cv2() else contours[1]\n",
        "    letter_image_regions = []\n",
        "    \n",
        "    # Now we can loop through each of the four contours and extract the letter inside of each one\n",
        "    for contour in contours:\n",
        "        \n",
        "        (x,y,w,h) = cv2.boundingRect(contour)\n",
        "        if h == 0:\n",
        "            print(x,y,w,h)\n",
        "        if w/h > 1.25:\n",
        "            \n",
        "            # This contour is too wide to be a single letter!\n",
        "            # Split it in half into two letter regions!\n",
        "            \n",
        "            half_width = int(w/2)\n",
        "            letter_image_regions.append((x, y, half_width, h))\n",
        "            letter_image_regions.append((x + half_width, y, half_width, h))\n",
        "            \n",
        "        else:\n",
        "            letter_image_regions.append((x,y,w,h))\n",
        "            \n",
        "    if len(letter_image_regions)>4:\n",
        "        continue\n",
        "    \n",
        "    letter_image_regions = sorted(letter_image_regions, key=lambda x: x[0])\n",
        "    \n",
        "    for box,text in zip(letter_image_regions, correct_text):\n",
        "        \n",
        "        (x,y,w,h) = box\n",
        "        \n",
        "        letter_image = image[y-2: y+h+2,x-2:x+w+2]\n",
        "        saving_path = os.path.join(output_folder,text)\n",
        "        \n",
        "        # If the folder doesn't exists make it\n",
        "        if not os.path.exists(saving_path):\n",
        "            os.makedirs(saving_path)\n",
        "        count = counts.get(text,1)\n",
        "        \n",
        "        # Writing image to file\n",
        "        p = os.path.join(saving_path,f'{str(count).zfill(6)}.png')\n",
        "        cv2.imwrite(p,letter_image)        \n",
        "        counts[text]= count + 1"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "As-HZXYaYPfH",
        "colab_type": "code",
        "outputId": "eb413ba4-1330-4266-93f4-c824ea24e72c",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 54
        }
      },
      "cell_type": "code",
      "source": [
        "for i,j in counts.items():\n",
        "    print(f'{i}: {j}',end= ' ')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "R: 1228 B: 1204 5: 1235 M: 1197 E: 1264 F: 1182 X: 1210 Y: 1224 Z: 1232 D: 1200 W: 1194 4: 1202 U: 1206 6: 1167 2: 1209 8: 1216 3: 1210 T: 1224 N: 1252 Q: 1217 J: 1139 G: 1228 L: 1202 V: 1159 P: 1248 K: 1225 H: 1227 7: 1285 C: 1204 9: 1359 S: 1232 A: 1193 "
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "gtt3HOAB1VoT",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "mapping = {'A': 8, 'B': 9, 'C': 10, 'D': 11, 'E': 12, 'F': 13, 'G': 14,\n",
        "           'H': 15, 'J': 16, 'K': 17, 'L': 18, 'M': 19, 'N': 20, 'P': 21, \n",
        "           'Q': 22, 'R': 23, 'S': 24, 'T': 25, 'U': 26,'V': 27, 'W': 28,\n",
        "           'X': 29, 'Y': 30, 'Z': 31, '2':0,'3': 1, '4': 2,'5': 3, '6':4,\n",
        "           '7': 5,'8': 6,\n",
        "           '9': 7}"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "65ONyQzyYkML",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Building and Training model"
      ]
    },
    {
      "metadata": {
        "id": "0sLQ9oM9ZWmD",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Making training and validation data"
      ]
    },
    {
      "metadata": {
        "id": "OYwOXEL4yv8x",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "base_filename = 'extracted_letter_images'\n",
        "y = np.array([])\n",
        "len_letter = []\n",
        "images = []\n",
        "for folder in os.listdir(base_filename):\n",
        "    Subdir = os.path.join(base_filename,folder)\n",
        "    file_names = glob.glob( os.path.join( Subdir, '*' ))\n",
        "    n_files = len(file_names)\n",
        "    len_letter.append(n_files)\n",
        "    label = mapping[folder]\n",
        "    labels = np.full(n_files,label)\n",
        "    \n",
        "    if y.shape == (0,):\n",
        "        y = labels\n",
        "    else:\n",
        "        y = np.append(y,labels)\n",
        "    \n",
        "    for name in file_names:\n",
        "        image = cv2.imread(name, cv2.IMREAD_GRAYSCALE)\n",
        "        images.append(image)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "eFa8C46oZehb",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "#### Resizing images"
      ]
    },
    {
      "metadata": {
        "id": "ph8vqkId8xDM",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "img_width, img_height = 20, 20\n",
        "\n",
        "resized_images = [cv2.resize(img, (img_width, img_height), interpolation = cv2.INTER_CUBIC) for img in images]\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "ZDpxSzyy6JdQ",
        "colab_type": "code",
        "outputId": "2e07b007-085a-40a7-d98c-5b370329c0e8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "np.array(resized_images[0]).shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(20, 20)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "metadata": {
        "id": "kErKTTHE6hBE",
        "colab_type": "code",
        "outputId": "d3935967-9724-4762-a139-afaa1e97b2d6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "y.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(38942,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "metadata": {
        "id": "0AaG30T-ZkkB",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Splitting Data into training and validation"
      ]
    },
    {
      "metadata": {
        "id": "MM4-q9w7wcP2",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from keras.models import Sequential\n",
        "from keras.layers import Conv2D,Dense,Flatten,MaxPooling2D,Dropout"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "NfLTV-FjxQ0Q",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X_train,X_test,y_train,y_test = train_test_split(np.array(resized_images),y, train_size=0.8, test_size=0.2)\n",
        "\n",
        "X_train = X_train.reshape(len(X_train),20,20,1).astype('float32')\n",
        "X_test = X_test.reshape(len(X_test),20,20,1).astype('float32')\n",
        "\n",
        "y_train = keras.utils.to_categorical(y_train, 32)\n",
        "y_test = keras.utils.to_categorical(y_test, 32)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "yZ_eBut_2sGL",
        "colab_type": "code",
        "outputId": "c7e1f77d-0e8d-40f7-a3c2-959891325c7d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "len(X_train), len(y_train)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(31153, 31153)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "metadata": {
        "id": "3d6VcfHG3iIt",
        "colab_type": "code",
        "outputId": "1e7da9cf-5cf4-45eb-bcd1-c821c6f253b4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "cell_type": "code",
      "source": [
        "len(X_test), len(y_test)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(7789, 7789)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        }
      ]
    },
    {
      "metadata": {
        "id": "U6imL8AqZ8bx",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Building Model"
      ]
    },
    {
      "metadata": {
        "id": "iuvksHUb3wk1",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model = Sequential()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "RD8puwSP4Tr-",
        "colab_type": "code",
        "outputId": "48c813f2-21ef-4eec-f903-06f5d185aec1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 141
        }
      },
      "cell_type": "code",
      "source": [
        "model.add(Conv2D(32,(3,3),input_shape = (20,20,1),activation = 'relu', name = 'Conv2D_1'))\n",
        "model.add(Conv2D(32,(3,3),activation = 'relu', name = 'Conv2D_2'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2), name = 'MaxPool2D_1'))\n",
        "model.add(Dropout(0.25, name = 'Dropout_1'))\n",
        "\n",
        "model.add(Conv2D(64,(3,3),activation = 'relu', name = 'Conv2d_3'))\n",
        "model.add(Conv2D(64,(3,3),activation = 'relu', name = 'Conv2d_4'))\n",
        "model.add(MaxPooling2D(pool_size=(2, 2), name = 'MaxPool2D_2'))\n",
        "model.add(Flatten(name = 'Flatten'))\n",
        "\n",
        "model.add(Dense(256,activation='relu', name = 'Dense_1'))\n",
        "model.add(Dropout(0.5, name = 'Dropout_2'))\n",
        "model.add(Dense(64,activation='relu',name = 'Dense_2'))\n",
        "model.add(Dense(32,activation='softmax', name = 'Dense_3'))\n",
        "\n",
        "model.compile(loss = 'categorical_crossentropy',optimizer='adam',metrics = ['accuracy'])"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Colocations handled automatically by placer.\n",
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "mjdNk8go4o6w",
        "colab_type": "code",
        "outputId": "22bd69da-e795-4da4-c904-78e6ef8d719d",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 555
        }
      },
      "cell_type": "code",
      "source": [
        "model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "Conv2D_1 (Conv2D)            (None, 18, 18, 32)        320       \n",
            "_________________________________________________________________\n",
            "Conv2D_2 (Conv2D)            (None, 16, 16, 32)        9248      \n",
            "_________________________________________________________________\n",
            "MaxPool2D_1 (MaxPooling2D)   (None, 8, 8, 32)          0         \n",
            "_________________________________________________________________\n",
            "Dropout_1 (Dropout)          (None, 8, 8, 32)          0         \n",
            "_________________________________________________________________\n",
            "Conv2d_3 (Conv2D)            (None, 6, 6, 64)          18496     \n",
            "_________________________________________________________________\n",
            "Conv2d_4 (Conv2D)            (None, 4, 4, 64)          36928     \n",
            "_________________________________________________________________\n",
            "MaxPool2D_2 (MaxPooling2D)   (None, 2, 2, 64)          0         \n",
            "_________________________________________________________________\n",
            "Flatten (Flatten)            (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "Dense_1 (Dense)              (None, 256)               65792     \n",
            "_________________________________________________________________\n",
            "Dropout_2 (Dropout)          (None, 256)               0         \n",
            "_________________________________________________________________\n",
            "Dense_2 (Dense)              (None, 64)                16448     \n",
            "_________________________________________________________________\n",
            "Dense_3 (Dense)              (None, 32)                2080      \n",
            "=================================================================\n",
            "Total params: 149,312\n",
            "Trainable params: 149,312\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "id": "vZFcxAg1ZqGb",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Training Model"
      ]
    },
    {
      "metadata": {
        "id": "QeR5orn75AC6",
        "colab_type": "code",
        "outputId": "e698ac19-4120-48d2-ca9a-1fd31d03f493",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 471
        }
      },
      "cell_type": "code",
      "source": [
        "model.fit(X_train,y_train,batch_size=32,validation_data=(X_test, y_test), epochs=10, verbose=1)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
            "Instructions for updating:\n",
            "Use tf.cast instead.\n",
            "Train on 31153 samples, validate on 7789 samples\n",
            "Epoch 1/10\n",
            "31153/31153 [==============================] - 11s 354us/step - loss: 2.8089 - acc: 0.6351 - val_loss: 0.1158 - val_acc: 0.9783\n",
            "Epoch 2/10\n",
            "31153/31153 [==============================] - 6s 200us/step - loss: 0.1584 - acc: 0.9603 - val_loss: 0.0454 - val_acc: 0.9924\n",
            "Epoch 3/10\n",
            "31153/31153 [==============================] - 6s 177us/step - loss: 0.0970 - acc: 0.9784 - val_loss: 0.0405 - val_acc: 0.9940\n",
            "Epoch 4/10\n",
            "31153/31153 [==============================] - 6s 188us/step - loss: 0.0809 - acc: 0.9826 - val_loss: 0.0380 - val_acc: 0.9946\n",
            "Epoch 5/10\n",
            "31153/31153 [==============================] - 6s 198us/step - loss: 0.0701 - acc: 0.9849 - val_loss: 0.0393 - val_acc: 0.9937\n",
            "Epoch 6/10\n",
            "31153/31153 [==============================] - 6s 207us/step - loss: 0.0646 - acc: 0.9868 - val_loss: 0.0370 - val_acc: 0.9945\n",
            "Epoch 7/10\n",
            "31153/31153 [==============================] - 6s 195us/step - loss: 0.0621 - acc: 0.9882 - val_loss: 0.0356 - val_acc: 0.9950\n",
            "Epoch 8/10\n",
            "31153/31153 [==============================] - 5s 175us/step - loss: 0.0577 - acc: 0.9894 - val_loss: 0.0330 - val_acc: 0.9967\n",
            "Epoch 9/10\n",
            "31153/31153 [==============================] - 5s 175us/step - loss: 0.0488 - acc: 0.9910 - val_loss: 0.0293 - val_acc: 0.9967\n",
            "Epoch 10/10\n",
            "31153/31153 [==============================] - 5s 175us/step - loss: 0.0605 - acc: 0.9895 - val_loss: 0.0326 - val_acc: 0.9960\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<keras.callbacks.History at 0x7f3f84984780>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "metadata": {
        "id": "fsfbOtu0aOPi",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Saving Model"
      ]
    },
    {
      "metadata": {
        "id": "3hDfE7OkIMPf",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "model.save(\"Captcha_Solver.h5\")"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "J2jOexiQaGp7",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "###  Evaluating Model"
      ]
    },
    {
      "metadata": {
        "id": "cWjerIKtOn7c",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        "def plot_model_history(model_name, history, epochs):\n",
        "    print(model_name)\n",
        "    plt.figure(figsize=(15, 5))\n",
        "\n",
        "    # summarize history for accuracy\n",
        "    plt.subplot(1, 2 ,1)\n",
        "    plt.plot(np.arange(0, len(history['acc'])), history['acc'], 'r')\n",
        "    plt.plot(np.arange(1, len(history['val_acc'])+1), history['val_acc'], 'g')\n",
        "    plt.xticks(np.arange(0, epochs+1, epochs/10))\n",
        "    plt.title('Training Accuracy vs. Validation Accuracy')\n",
        "    plt.xlabel('Num of Epochs')\n",
        "    plt.ylabel('Accuracy')\n",
        "    plt.legend(['train', 'validation'], loc='best')\n",
        "\n",
        "    plt.subplot(1, 2, 2)\n",
        "    plt.plot(np.arange(1, len(history['loss'])+1), history['loss'], 'r')\n",
        "    plt.plot(np.arange(1, len(history['val_loss'])+1), history['val_loss'], 'g')\n",
        "    plt.xticks(np.arange(0, epochs+1, epochs/10))\n",
        "    plt.title('Training Loss vs. Validation Loss')\n",
        "    plt.xlabel('Num of Epochs')\n",
        "    plt.ylabel('Loss')\n",
        "    plt.legend(['train', 'validation'], loc='best')\n",
        "\n",
        "\n",
        "    plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "metadata": {
        "id": "R4y9ToeIOr_u",
        "colab_type": "code",
        "outputId": "433c6a17-b2c6-421a-ad15-4d5143e6ac1b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 367
        }
      },
      "cell_type": "code",
      "source": [
        "plot_model_history('model',model.history.history,32)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "model\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA4kAAAFNCAYAAACtyCreAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "rTIBIH7tfWFY",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "## Testing the model on Outside Data"
      ]
    },
    {
      "metadata": {
        "id": "gJDwKgBlfa-a",
        "colab_type": "text"
      },
      "cell_type": "markdown",
      "source": [
        "### Captcha Preprocessing"
      ]
    },
    {
      "metadata": {
        "id": "uC496L4cuKpI",
        "colab_type": "code",
        "outputId": "7e2a31fb-63cb-4251-9145-919ec9e3c696",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 173
        }
      },
      "cell_type": "code",
      "source": [
        "from skimage.transform import resize\n",
        "img_size = 20\n",
        "\n",
        "predictions = []\n",
        "\n",
        "img = cv2.imread('captcha.jpg', cv2.IMREAD_GRAYSCALE)\n",
        "plt.imshow(img)\n",
        "image = cv2.copyMakeBorder(img,8,8,8,8,cv2.BORDER_REPLICATE)\n",
        "\n",
        "thresh = cv2.threshold(image, 0, 255, cv2.THRESH_BINARY_INV | cv2.THRESH_OTSU)[1]\n",
        "\n",
        "contours = cv2.findContours(thresh.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)\n",
        "\n",
        "contours = contours[0] if imutils.is_cv2() else contours[1]\n",
        "letter_image_regions = []\n",
        "\n",
        "for contour in contours:\n",
        "\n",
        "    (x,y,w,h) = cv2.boundingRect(contour)\n",
        "    if h == 0:\n",
        "        print(x,y,w,h)\n",
        "    if w/h > 1.25:\n",
        "\n",
        "        half_width = int(w/2)\n",
        "        letter_image_regions.append((x, y, half_width, h))\n",
        "        letter_image_regions.append((x + half_width, y, half_width, h))\n",
        "\n",
        "    else:\n",
        "        letter_image_regions.append((x,y,w,h))\n",
        "\n",
        "if len(letter_image_regions)>4:\n",
        "    print('Sorry! but the captcha is more than 4 letters!!')\n",
        "\n",
        "else:\n",
        "    letter_image_regions = sorted(letter_image_regions, key=lambda x: x[0])\n",
        "\n",
        "    for box,text in zip(letter_image_regions, correct_text):\n",
        "\n",
        "        (x,y,w,h) = box\n",
        "\n",
        "        letter_image = image[y-2: y+h+2,x-2:x+w+2]\n",
        "\n",
        "        X = cv2.resize(letter_image, (img_width, img_height), interpolation = cv2.INTER_CUBIC)\n",
        "        \n",
        "        X = X.reshape(1,20,20,1).astype('float32')\n",
        "        prediction = model.predict(X).argmax(axis = 1)[0]\n",
        "        for key,value in mapping.items():\n",
        "            if value == prediction:\n",
        "                prediction = key\n",
        "                break\n",
        "        predictions.append(prediction)\n",
        "    print(f\"The Captcha Text is {''.join(predictions)}\")"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "The Captcha Text is 8746\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAACLCAYAAABx0GIrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "metadata": {
        "id": "NkwYSgoAa58f",
        "colab_type": "code",
        "colab": {}
      },
      "cell_type": "code",
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}