obtain_thresholds.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import cv2\n",
    "\n",
    "from obtain_thresholds import read_img_data_from_disk, k_fold_cross_validation\n",
    "from Localization import plate_detection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = read_img_data_from_disk('./dataset/ColourVariations')\n",
    "\n",
    "randomise = np.arange(X.shape[0])\n",
    "X = X[randomise]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "X.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "k_fold_cross_validation(X, 5, 3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./dataset/bounding_boxes.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "category_1 = df.iloc[:32]\n",
    "path = './dataset/Frames/Category_I/'\n",
    "avg_intersection_over_union_i = 0\n",
    "\n",
    "for i in range(32):\n",
    "    row = category_1.iloc[i]\n",
    "    bounding_box = BoundingBox(row['Left'], row['Top'], row['Right'], row['Bottom'])\n",
    "\n",
    "    frame = cv2.imread(path + row['Image'])\n",
    "    plates = plate_detection(frame)\n",
    "\n",
    "    if len(plates[1]):\n",
    "        avg_intersection_over_union_i += (bounding_box.intersection_over_union(plates[1][0]))\n",
    "print(f'Detected correctly Category I: {avg_intersection_over_union_i / 32}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "category_2 = df.iloc[32:60]\n",
    "path = './dataset/Frames/Category_II/'\n",
    "avg_intersection_over_union_ii = 0\n",
    "\n",
    "for i in range(28):\n",
    "    row = category_2.iloc[i]\n",
    "    bounding_box = BoundingBox(row['Left'], row['Top'], row['Right'], row['Bottom'])\n",
    "\n",
    "    frame = cv2.imread(path + row['Image'])\n",
    "    plates = plate_detection(frame)\n",
    "\n",
    "    if len(plates[1]):\n",
    "        avg_intersection_over_union_ii += (bounding_box.intersection_over_union(plates[1][0]))\n",
    "print(f'Detected correctly Category II: {avg_intersection_over_union_ii / 28}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "category_3 = df.iloc[60:]\n",
    "path = './dataset/Frames/Category_III/'\n",
    "avg_intersection_over_union_iii = 0\n",
    "\n",
    "for i in range(0, 10, 2):\n",
    "    row_1 = category_3.iloc[i]\n",
    "    row_2 = category_3.iloc[i+1]\n",
    "\n",
    "    bounding_box_1 = BoundingBox(row_1['Left'], row_1['Top'], row_1['Right'], row_1['Bottom'])\n",
    "    bounding_box_2 = BoundingBox(row_2['Left'], row_2['Top'], row_2['Right'], row_2['Bottom'])\n",
    "\n",
    "    frame = cv2.imread(path + row_1['Image'])\n",
    "    plates = plate_detection(frame)\n",
    "\n",
    "    if len(plates[1]) == 1:\n",
    "        b1 = bounding_box_1.intersection_over_union(plates[1][0])\n",
    "        b2 = bounding_box_2.intersection_over_union(plates[1][0])\n",
    "\n",
    "        if b1 < b2:\n",
    "            b1 = b2\n",
    "\n",
    "        avg_intersection_over_union_iii += b1\n",
    "    elif len(plates[1]) == 2:\n",
    "        b1 = bounding_box_1.intersection_over_union(plates[1][0])\n",
    "        b2 = bounding_box_2.intersection_over_union(plates[1][0])\n",
    "        b3 = bounding_box_1.intersection_over_union(plates[1][1])\n",
    "        b4 = bounding_box_2.intersection_over_union(plates[1][1])\n",
    "\n",
    "        op1 = b1\n",
    "        op2 = b2\n",
    "\n",
    "        if b3 > op1:\n",
    "            op1 = b3\n",
    "\n",
    "        if b4 > op2:\n",
    "            op2 = b4\n",
    "\n",
    "        avg_intersection_over_union_iii += op2 + op1\n",
    "print(f'Detected correctly Category III: {avg_intersection_over_union_iii / 10}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from test_pipeline import localize_and_rotate\n",
    "from pre_processing_data import read_image\n",
    "from Segment import remove_black_rows, dilate_or_erode, clear_top_bottom\n",
    "from pre_processing_data import read_reference_characters\n",
    "from character_recognition import recognise_character\n",
    "import Helpers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "frame = cv2.imread('./dataset/Frames/Category_I/plate12.jpg')\n",
    "plates = localize_and_rotate(frame)\n",
    "plate = cv2.cvtColor(plates[0], cv2.COLOR_BGR2GRAY)\n",
    "plate = Helpers.isodata_thresholding(plate)\n",
    "plate = clear_top_bottom(plate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "plate = np.array(plate, np.uint8)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x159d1561120>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAh8AAACICAYAAABHoYkDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/bCgiHAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAoPklEQVR4nO3deXyM1/4H8M/sWSYb2SUhIbEWERKDqqKitNbbRbVVlJBQpHUv97a099fS1i1tNajWpe1VsbSqaqnUEjSxJBJrEBFCZBGafZvl/P6ITo2IzEzmmWeW7/v1mtdLZs5zznfmmJnvPM9ZBIwxBkIIIYQQMxHyHQAhhBBC7AslH4QQQggxK0o+CCGEEGJWlHwQQgghxKwo+SCEEEKIWVHyQQghhBCzouSDEEIIIWZFyQchhBBCzIqSD0IIIYSYFSUfhBBCCDErzpKPhIQEtGvXDg4ODoiKisKJEye4aooQQgghVoST5GPz5s2Ij4/H4sWLcerUKfTo0QPR0dEoLi7mojlCCCGEWBEBFxvLRUVFoU+fPvjiiy8AABqNBoGBgZg9ezYWLFjwyGM1Gg1u3boFFxcXCAQCU4dGCCGEEA4wxlBRUQF/f38IhY8+tyE2deP19fVIT0/HwoULtfcJhUIMHToUqampjcrX1dWhrq5O+3d+fj66dOli6rAIIYQQYgY3btxAQEDAI8uYPPkoKSmBWq2Gj4+Pzv0+Pj64ePFio/JLly7Fe++91+j+ARgBMSSmDo8Qoqdl549BLtA9MTpsz2x4pgnRK+Y0cgbVNXGkcYpio+A/8jrYAjckbN5g0LFlGiEWdIsySRwlr0dia/xnBh2zoTQCx/tJTdI+IdZKBSWOYjdcXFyaLWvy5MNQCxcuRHx8vPbv8vJyBAYGQgwJxALzJB83F/bD0ikbzNJWS6weOwqac40TOGujGhyBOV9uMvi4WibBurBgDiIifxL7+iDmyBEAQC/nxh8gB59LwN3xUnSUaLAkMxLp4S0fNnblu3Cs6LsZ7STpCBRpcHqHHGGOrgbXM+dKbotjAYAg8SmEyQxr/22380g43xm/dWv+QxcAxAFtEHPwoDHhWZwVsydCuvck32FwomcG0F9+me8wHur9yyPhMTKb7zB03futos+QCZMnH56enhCJRCgqKtK5v6ioCL6+vo3Ky2QyyGQyU4dhEKULwyjnal5j0McqiYjvEExC7SA06vWuY0qsAyUfnBKJHtk3wRK5tgeGuZ5DOrq3uMnWrSrvtekAABjkqDGqHtO9hw3/PHISSvGGx0VcOTEQ1yJrmj9A/OjX2Zpc/2Qfcj7wwm9bItHmoxS+wzGpIS7nMcxJyXcYD9W123q8vGcSXJ/O4TsUo5h8totUKkVERAT279+vvU+j0WD//v1QKBSmbo5Ymbqn+6DVgmsGH1egqkTfD+aYPiDCq+zP+uLtsF18h2ESMoEEK/yT+Q7D7GZ7XMenfmmobqPmOxS70l4ix6yQQ3yHYTROLrvEx8dj0qRJ6N27NyIjI/Hpp5+iqqoKkydP5qI5YiVqxkRCM6MEP3ZIMvjYCiaA9yrb+lVFgPGPH7eZMwDEtuQsUyBUchSAnO9QbBInyccLL7yA27dvY9GiRSgsLETPnj2xd+/eRoNQiX251V+EK91/NOiYw7VA3OmXUFMjRXtkchMYIYQ8QBRUBWchLffAFc4GnM6aNQuzZs3iqnpiZQThXSEMrDLomN9rNZiUPA1hk9M5iooQQh6u3QtnkHnF3WLHfFg72tuFcE4c3BYVS2tweeC3eh+TVV+Nl3+dQYmHjUu7E4QCVSXfYRBCzIySD8I570138bsBl1tK1FUY892bCJtJ+wHZOulT1zH6rO2MBbuuUvEdAiFWgfd1PoiNM3CJfDXTIPr/3kK7tY1XwyXEkp2vr0F8u358h0GIVaDkg3BqYtYNvOpaonf5oVOmw/NXSjwIIcSWUfJBOPOvq5noL9PAkKt7ApNvc0gIIcTSUPJhhIFnx0I+l4eX7oqFLaXbjDaiSogE+s2RHz7qZQir6iDLPQfj1rck1iz45+mQ3BXh8mursaXSDRtGDzOqnu2/fQ+ZgdsyDJo2DU5XS/Uu/+TWU5jfSndVyY0VrbHx2UEATLfaZHTWMxDNcmy2XNbfXZEbvc7g+jeUe2PL2CfAJCLs2ZsIAHjq+dcguVMFn/WFWB90RK96It6bCZ/DJehUchW0zJj5zC3ojeznAgBc4zsUo1DyYYTSGgc4ZmXxHYZFCz0pQ4BY/2WqhZfzoC4v5zAiYqlavS2FZ2UJBPVKPHl4GsQ1KgizMoyqS80YYMAwowGzY+By8CzU1fovdJYU8zjO/qcNvm17GADwj6KeODU3HMLL+sesLijCk1Om6dz31hffYaRTLQCg27GJCPg/QJ11vtm6On/YASEVMbj6ty/1bv/Ngl44M7dHw+ssEODJKdOwd90q9Fl5CtUaKWI9kwE4P7KO/nNiIK3QwCftMtQld/Ru25aFL4lF/0np+KLNcc7bKlM6QnX1GuftcIWSDwMNODMOrdbSindNEchkKNoSjF/bJAJ67kocviQWPtU0s8VesYzz2l/M0us3WlRX/yVzsXPBxwgQ6/cedT1w2aDEAwAEv2fi+pJIhAd1AwA4F6rhnGzYlw2rq2u0Gdu//z0Zb7s0ZE5+Z2rAMvRLZtSXrqDjWjFCoF8C8vzVIShY3gFOR+7FzBike082bF9wL3FLRmSz9fj8nA5WV0dnO+7pujIWbTdn43R+T4T7R3DenlOxGnJwn+RwhZIPAxVkeaPD7mN8h2GxBGIxMvok6lW2UlOLnolz0f6LFNBQD2IK3qtSUPZ3EQI4bsdh54l72+CZjvu3xg+01py7iIADkcDfmi+bnhuEDj82/tIydPsCes/qavfddahu34bT9ttw4jsYK0DrfBCTEbq4IH9GD73K3lRVoteRGLR/ixI5QgixN5R8EJMQubvhzrhuOPPmqmbL5ior8cKFVxE84bQZIiOEEGJpKPkgJlHfsz1OLF2tV9lPiodAPvwqxxERQgixVJR8kBYTiMVQOYv0KlutqUdJ/aNH0RNCCLFtlHyQFrvzah8c+uorvcr2T38Vf/S/y3FEhBBCLBklH6RFrv9bgZPv63e5JWRbDLxHX+Q4IkIIIZaOkg/SIsywfeMIIYQQWueDGC/7iyjseuYTQI9Z7Z2+nolOK6/QgkSEEELozAcxntCjHp2l+i2nI7srgPr2bY4jIoQQYg0o+SBGyV2iwNsRu/QqG7ItBv6/0SBTQgghDSj5IEbp0v8qXnMt1qtswAEGzTkaaEoIIaQBJR/EYH9MUqB/6yt6lR10bgyc8ms4jogQQog1oQGnxCD10b3xj39txHh5ebNl5xb0hlOcEOrss2aIjBBCiLWg5IMYZMWaBPSUyfQqeznaDeoS615GXeTjDXh6mLROlncLmooKk9ZJiK0S+/kCAFQFhTxHQkyJkg9CmiB0csLFRcG4OvZLk9bbd/4MuG2k3XyJ/RI6OEDg6KhX2Ysf+0EgYgiLq2uyjLq0FGDMRNERc6DkwxiCeytr3f+fXaDnalvW+ga59/yEAgPiFwj1f10eZAGvU/73bXE10rSJBwCAFmYj9kwgwMXl3ZE7Zq1hx51v+qHoMa8AJ+jyrjWh5MNAOS+sgfp5DWbcfBx5UVUNdwpF2H3jZLPHnlfWY367vhxHaHoCmQy7rqYCAEQCB72P25n5q9FtdvouDiELUo0+3pId+3gNurSLReD7KXyHQojZ3d0ZitxeBiYezfj1p+/Qf04M5FuPm7Rewh1KPowgEgixKuAwSvJq7rtP3uxx3aUO2JB3VPv364rnocq/xUmMpiYSGD4xyphj/pTx8gpUvKTC1JznoX7SOl4jffV+ZyaCvk0D/+d27EPw7tfROf4yqgZ2QvJa037pEcMJDDl7aljF3NSrp69TNrfo+GHp0+E/9oKJorF8lHwYSSIQwU/cfMLxoPuPmZOchFomAQC8s+o1+K6wvF/CYl8fxB5NNnu7cqED5EJgY4dtOJLtCQCoZRKsCws2eyymJq5lYMp6vsOwHyoh1OXlEFfT4v4AcPCJlRi53fa+6FZ8tBIxnnPgvYqfz1Fjvg/uty9iLU5me+vclxDW0SIuQXOBkg8eDXNSAlACAJQz/4cjL4dhx8leCJtxgt/A7icSYaRTLW/Ne4icMMq5GgBQx5RYB+tPPoh5ff3UOuxMD0cb2SG+Q7EIQWI52re6A1tbfSdSJoGqZd//vPITyzFKXK1z34E0ES4NcrTJ2XGUfFiI8fJyjJenYcTQ05iRMAWhcfxfuxSFhqB0Jd9R/EUMEYp3dAIA+E29DXXJHZ4jItZgiKMaQxzT+A6DEIN96peGyMTnoFS1AQC4rXGBbE/z4wutgd2vcPrHJAV6PXGJ7zC0hjkp8d3Tq3Hl077I+YTfwanq1nL83v1HXmO4n0ggREafRGT0SUTWkhCI2wXxHRIhxAD5/+iHV9pZ0JldK3AifKv2c68sWMJ3OCZj92c+SodXIzH4AN9h6OjvIETO82tQralHv+tzAQC+X6aD1TU9z93e5D7zFUJLZyL0awnUl3P4DocQooenX0jFbI/rfIdhtcr71YCJ+wEAnIo0cNlsvesF2X3yYcmchFJkLlgFAOhXPANuO89CU1VllrZFPt4o6G3ZF1CzX16NMPVMdFingfpKLt/hEEIIp3IGrwcGN/z7/ZJOOHi7P8QH0vkNykh2f9nFWqQsXwMW1s5s7ZU+GYKMf64yW3vGujxpNS7O9oGoa0eI2wbyHQ4h5BHS7gThh0pXXFaa/kfU77UaSCpsc2bIw7zteRFzvtwEUedQvkMxCiUfVkQtl0Lk4QGBRMppOwKZDEpH61mG8+pza7A7aTPYBtNPpSzT1OAPdXXzBfVwU1WJPFUlhDTjk9gp6VPXsTYsBMN/iTdpvQWqSix4aya8VtvmwoRNGeVcjYS96yHyMO3+U+ZAl12sSNLm9QCA7p/Ewu8T7uay33gzAudncXPWQ800je5ryWJkjQgEJpsXz5gAvbbMg+tVIdIWftHi+mL6vQDVzXy4wHqv0xJiEuzhnwXGmjxuBpzSTDxDUCCASGC6GLkSLJFjx7nf8ExAb6taE4SSD2I2IdtiEDrngZHuAiF+vWmaa5a7O+7GsrPt8Vs3F5PUd/8iTCMSIlpeIctveR2E2IDQWccxYrYJ3lN/YudMV9c9G64fafHCYeYiEYiw+2Y6ngkdAE21ac7Ucs3uL7u0n3wJHTbO5DsMgxyZ9wmurLCePWL6xc/AiE4DETY/syEzv/+mUWNEp4EoVpvmGnC8RzZmZl8xSV06HozbmBsh5C+meE9x+N6a2m0EDtVYz1ekqCUbefLA7s98aGprIWxYZBTL7rbHwed6cdaWqpUz9m3d0OJ63ISO2DfuPxjmORsdXsloeWAciR73KkRlNXC7fgbqe9n4H5MUePvtb/BbWVdc6t3wwqvLy/HakFcBUcMbXf1FDX7t/ItRbYoEQviL/zDNEyCE2C11eTk+jh6Nkz+dwfxWDdP5w0++CL8Fxg3aYjIJ9uz+HgAw7G+TIP6j4TOx5D8Na3nYG7tPPu43wfU08v/ngbv1TihSlJukzruTFeg/q2FFuuJaFzwxfbpJNrdqL5GjdavKFtfzoBvv9MO7r2w0+vgyTQ3GvP4GAECWloE/draF80edIDp0CgDg9dt1rCiaCFGtGiKc0h53/1odgre6ots7E3Gur3FxdJMwBB13/mvXYUIIMYL6Si6SYh7HbpeG+a2+tyqgzso2rjKBAE9OmQYAaPOfbHjKGj6/322VCsA0i4d1OVyDi6P8rWLDUko+AAT/XIXw/Fjt3wI1gxdMM2raM6MMyf+NBACI6hla7zqG8A9iceyfn0EmsKzV6q6/1w9zn9uB5+VlRh2fVV+NiR+9Ca+9Da8dA6BO9IYs9yZU98qo8m9B2swbg2Wch9+KcITFvYrLA781OA4noRSrAg4jcsdEeI/NBjQ0vYQQYhzB75n4c35hi4afMgbp3oYfollBCmgkDZdIkhGpLVIRosGVCWuMbuITv1MY+N+xkM8Lg/rC5ZZEyzlKPgDg2Bl4czQBQZN5Ad6Zuvd5J6SgW9s3wEQM/xubgL4OIm4aN5Cv4hZmuBs3KPJwLTD9u7fQdrXuLByPb1K1iYchhEcyECTqhS7il3Gh3/8MPl4iECGjTyKGiyLBKPkghFgQz7UP/3HrFxqCYNdpyB35ldF1H35sO4Z6T4HIwjctpuSDJyF/b/jP95JDLNZE//feDrfWaVe1A2bveg2hi007/Vd06BQCqx/DYwtfglSsQnrEFpPWTwghlkSdfRVdFlXjMc+XAABno77nOSLuWM9QXhsVGnccMftfM2pUdQf3EmgG9DR9UAZadnU4QudwdOroxFn4j70Anyk0iJQQYvtUBYXwH3sB/mMvIOamAkpmm2duDfrGe/fddyEQCHRunTp10j5eW1uLuLg4tG7dGnK5HOPHj0dRUZHJg7Y1YTEnMfXn6ThTX2vQcd8HH8QTq45BFNaeo8ial6OsRH6JO2/tE0KIrboWWYMidQ32VssMWmm5MkAKkasrh5G1nME/t7t27YqCggLt7ejRo9rH5s2bh507d2Lr1q1ITk7GrVu3MG7cOJMGbKs6zDuGCWvjUWLgehdve15E3O5dHEXVvKG/zkPIS5lmaSvv3vLkhhC2cucmGEIIMYPrKid8PvJZvFc8UO9jjn28BrfHd+UwqpYzOPkQi8Xw9fXV3jw9PQEAZWVlWLduHZYvX47BgwcjIiIC69evR0pKCo4do+Wk9RGwJAUj//UW32FYJPXt25gWNADTggagTFOj93G7M/ZB7OfLYWSEEMKdf4f0gvoSBwsn8szg5CM7Oxv+/v4ICQnBxIkTkZeXBwBIT0+HUqnE0KFDtWU7deqEoKAgpKY2PW21rq4O5eXlOjdCTOnntN0Q9O7GdxiEEELuMSj5iIqKwoYNG7B3716sXr0aubm5ePzxx1FRUYHCwkJIpVK4u7vrHOPj44PCwsIm61y6dCnc3Ny0t8BA2hadmJZJN64jxAZs7vAL+mTa5kBG0mDfvz9B3qJ+fIfRJIOm2j799NPaf3fv3h1RUVFo27YttmzZAkdHR6MCWLhwIeLj/9peuby8nBIQHkgO+eG79hsBOOt9TOh3M9F5eQ7oI4wQ6yITSOAjKQdgfVuxE/14iJygkVjunlIt+kno7u6OsLAwXLlyBb6+vqivr0dpaalOmaKiIvj6Nn3NXSaTwdXVVedGzK+Px3V4i/RPPABAWiqAuqiYo4gIIYTYqhYlH5WVlcjJyYGfnx8iIiIgkUiwf/9+7eOXLl1CXl4eFApFiwMl3Nr70UC8XfwY32FwottnsRBeK+A7DEIIIfcYlHy89dZbSE5OxrVr15CSkoKxY8dCJBJhwoQJcHNzw9SpUxEfH4+DBw8iPT0dkydPhkKhQN++1rP9u71y/f4Yjhbzt14Il9p+kwN1yR2+wyCEEHKPQWM+bt68iQkTJuDOnTvw8vLCgAEDcOzYMXh5eQEAVqxYAaFQiPHjx6Ourg7R0dFYtWoVJ4ETC9C7DKrBERAfSOc7EmICIi8vXJvUju8wCCF2wKDkIzEx8ZGPOzg4ICEhAQkJCS0KiliH84qNCL4zDWEH+I6kabPyo8CU1rtvjjmp2/vh/Cz6sUAI4R7NQSRWRdS1o0Hlr/RjUN+5y1E0hBBCjEG72loQgUwGlYEzlpVMjXxlK24CsjAiLy/sTtoMwLhp3YQQYo1Erq6QCCoMOqZEXQVRvYCjiFqOkg8Lcu2fEbg4zbDT3gsK++BchIajiCyHyMsLu08n8R2GTWMCy/2gIsRuCQSYkZ6GUc76byynZhqMeOctBG5I4TCwlqHkw0Jkb4jA1WF0vZ3wo+LFvkhZvobvMAghD/jq+hEEieUGHRP98uvwONj0tiaWwK7HfGj2B2JD3lFkfx7FdyhGGXBmHC4M5PcSxPkRCbj7SxinbQh7dMa69O2ctkEIIZbIqC9py13YVMtukw+vFHdsCkuEn1iO/aM/QVz2ZQgOtOEllps/dMWhwZ8ZfFy9WgRNVZXJ4nCeokRY8iSDjnESSrGrx3pIDvmZLI771Y3og3/8uBl+Bmb+lZpajBjyHJiynpO4zKVkZxjisi8jLvsyrnwXzkkbRW/0w6dLV3JSNyFWRyDQvufuv91caP59UmZmX4GfyMns7ZqD3SYfAz0uw/PecuLBEjlGOVfj6w6b0TldDJ9U8yzxLkv2Red0MbZFfGXwaTUuqG7mQ1khNfg4b5EzvgrZCuwPMGk8pa8qMH7ZPgxyNG5Mizor26Tx8KFjq9sY5VyNUc7V+GHAGuQmdjdp/Tfe6Yd/ztqISJnEpPUSYrUEQu177v7b0ikbkLvEPKt1C52d0TldjDHOlTa7MaZdjvm4+n1PDHH6AoDuF36AWI5P/dJQx5Tou+OVRse5rXWBbNdJo9sVubuh4Ju/zhAcCPkvPEROACwns23/vRpdWr2MC/3+Z9BxfmI5NoZtxsBtMQj827kWx1ESo8Cg6ccx2+O6wcfmKisx/sP58IJlX/M0VE+ZDFuivsLkHbpnp3wn5EFTrf9gNADI/iIKboFleCN0B56Xlxkcy09Vcnyw9BW0srHXmJCmjHKuRum4rVjRdWijx3zGXwFTqVpUf310b5TGVgIAxCIN9vhtMaqeLgmxCL6Yg5ZFwz27TD5+6rca7SVNn2mQCSTI6NN4QbWev8XCpwXtChwcHqjX+KRj2o3+0Gz3BGDaX/eiQ6cg6NcPMOIMo6fIGRmK9ejy6Wx0mHccYMZdeCyZrkDvqZn4xO+UUcff1sjgtdo2vxR7ymSN/m8G/2c6hPVCdFx1G+rLOXrVM+nxI1jsdcHoOE5XB6HVf23zNSakKa+6luDVh3w3hKyIQad/ZUFdXm5UvbXPRkIVV4KM7j+2NEQEb7oFVWFRi+vhml0mH8aqVFQDAuOv+ykN2zS2SW/c6oNT67rD82tuPvw9LqvxYu5gJAYbvnSpTCBBzvNr0ONaLAQPXC3xOl0DYXJGo2OK3tB9TZ96LRXLfBuX00dWfTVePDILoTAucbFGuWPWAgDaYwZcruqXHvdw+tbo9nZVO+CbQwMRimNG10GILbk6/kt0LYiFpNK44x2GF+OYCRKP8JMvwq/COvaxouTDAFcGbQAG8R0FcGhTH/it5W7+tvO24yiq6I33Pi40+tfx6b83njbcNXUivOR9dO5jAgEyF5hmivH5+hq8mPE6Ql+1n8TjfjnPm2eq7LKrwxE6hxIPQu5nCVsT+L50A2oTTkLgEiUfVub3Wg0kldzPo5L+moaUst7Aj8afmn/QecVGgMPxWvNzx8N/rOniJbZP1SkIwrQqq58VRYi1sc1htDaqQFWJhfEz4PklXWt/UKWmFmV1DnyHYfMqNbWoqDN8RpSl2vfDNxB0DOE7DEJaLE9VafQ4Oz5Q8mFFJo+OgeNPJ8zXoIZBzTRQM8tfvv2xpDjIh1/lOwyb99iu2fB89jLfYRBC7lOtqce0oAEGz3rjEyUfpGknzmJEQASeaW/+xXUM0XltLMImp/Mdhs3rujIWYTHGTzUnhJhejrISYwOtb5VuGvNhJUYMHAuWw8N4BsbALPhUXsS7M9F2fZo1rCZMCCEmtaXSDRt6DwSYcVN8+UTJhwUr09RgwlOvAgDUOVd4u57H6uowYshz2Ja0EU5Cy7ne3y9+Brx3nYOGBgtyrueHsWj7XRbUfAdCCIHi9Hh4xAshqFNCXX6N73CMQpddLJw6K7thmXCezz6os7IxcmoscpRGTmQ3sb7zZ8B99wVoKir4DsXmhb8fC/9N2VD/8QffoRBi98KSJ8F9gRTqrGyorl7jOxyj2WXyMemDeByqscynvvh2V4R/EIvwD2Lx5AfxfIejQ/prGl54fz72VfO7D0iPj2Phsf2M0asJEv11+zwWflsuQX37Nt+hEGIeGjXCP4hFpaaW70gaCfltCtqtBDSns/gOpcXs8rJL669SEec1A+9M2oQXXSzr11xKSQi8E7hbQKylWn+Vinme01DnqcHzg1OwxOeMWdotUVeh75Y3AQAdvjgBTQv3UbAWFzZ3xntTi1q0FLqx2m+egbCEs1DT2SViZ7wTUhAeOA9MwvDlmK8wxJHfC47Bu1+HqEyMDj/UQJBymtdYTMUukw8ACFiSgkXyF1E77ge85lrMdzj4R1FP7L7WBcoz7miLG3yH80gBSxuSo11zB2DnE90Q7nsT37Y9zElbh2uBuNMvoaZGivZvNqyqaU+DS30+T8E26SBs698TkX55WBd0lNP2/lBXY2Da6wCA0PlpNp/k9Up7AX5llbD8yeTNc8qvwaBzY3Co2098h2ITQhY0rKc0XTQNK0Z+i1HO5p/G+tjxlwAAXRbfgir/ltnb55KAWdhUhvLycri5uWEQRkMs4P70fu4SBYY+9dc+Ii7iWnzkk8l5u/db+UdbfLNiBFpztFcL12qfjcRTHxzG254XTVrv77UavJw8jabR3lM1PgrdFzT86hEL1Pjc33TTXrPqq7Hy9mBcq2wF9ZPW/SEnS/bF1yE/wFvU/GZKI7o+aVNjWYQ9OmPPnk2PLHO4Fki80xe/5wfDd4z1n743h+yVURjeV/eMg6Ow3ujNL5ujZGrMyh+Aa1G1vI/3M4SKKXEIO1BWVgZXV9dHlrX75ONB4nZBmLLvEABgvNy0YwqUTI2fqzwa3f/pgglw+vG4SdsyN6bogZhvtkMk0GCMs+GDUn+ucoKS6Z6Ie2v/iwibacZF1ayI0MUFr6c//PTrY7IChEma/uLdWy1DlUamc9/ic8+izbjzJo2RT61/98B4r+aT1nWR4VCXlpkhIvMQdQ7F1B2/PrLM/L0TEPqGdX/eWAKRlxem/t7wOrbkuyKzrg45Si+d+wpVbvi5S+sWxccHSj5MZNX1o5AI/vrbUyg1eqqpmmmQVOOIFR06myg6yyR0csKXF/cBAILEcgBAsboKcoHkoa9dnqohUYnp+jTNXDGRnGUKHHhhWZOPzxz4ElTX8swYESG2bfm1VLgIdceFuAlFcBM6PvK4YnUVhi2bD5/PLXecnyGsOvkoKyuDu7s7BmAExOA3+XhQ4aaOSI3YZtSx68t98XNvr+YL2pDl51MRLJFjePxruPtiNU5EbtF5PF9ViVldONxpjhBCeHJjYRTSJn/9yDLD5k+G8440M0XEPRWUOIrdKC0thZub2yPLWlzycfPmTQQGBvIdBiGEEEKMcOPGDQQEBDyyjMUlHxqNBpcuXUKXLl1w48aNZk/dEMtRXl6OwMBA6jcrQn1mnajfrJOt9xtjDBUVFfD394dQ+Oi1tCxuqq1QKESbNm0AAK6urjbZQbaO+s36UJ9ZJ+o362TL/dbc5ZY/WeYyn4QQQgixWZR8EEIIIcSsLDL5kMlkWLx4MWQyWfOFicWgfrM+1GfWifrNOlG//cXiBpwSQgghxLZZ5JkPQgghhNguSj4IIYQQYlaUfBBCCCHErCj5IIQQQohZWVzykZCQgHbt2sHBwQFRUVE4cYJ2NeXT4cOH8eyzz8Lf3x8CgQA//fSTzuOMMSxatAh+fn5wdHTE0KFDkZ2drVPm7t27mDhxIlxdXeHu7o6pU6eistLwnW+JfpYuXYo+ffrAxcUF3t7eGDNmDC5duqRTpra2FnFxcWjdujXkcjnGjx+PoqIinTJ5eXkYOXIknJyc4O3tjfnz50OlUpnzqdiV1atXo3v37toFqBQKBfbs2aN9nPrM8n344YcQCASYO3eu9j7qt4ezqORj8+bNiI+Px+LFi3Hq1Cn06NED0dHRKC4u5js0u1VVVYUePXogISHhoY9//PHH+Pzzz7FmzRocP34czs7OiI6ORm1trbbMxIkTcf78eSQlJeGXX37B4cOHMX36dHM9BbuTnJyMuLg4HDt2DElJSVAqlRg2bBiqqqq0ZebNm4edO3di69atSE5Oxq1btzBu3Djt42q1GiNHjkR9fT1SUlLwzTffYMOGDVi0aBEfT8kuBAQE4MMPP0R6ejrS0tIwePBgjB49GufPnwdAfWbpTp48iS+//BLdu3fXuZ/6rQnMgkRGRrK4uDjt32q1mvn7+7OlS5fyGBX5EwC2fft27d8ajYb5+vqyZcuWae8rLS1lMpmMbdq0iTHG2IULFxgAdvLkSW2ZPXv2MIFAwPLz880Wuz0rLi5mAFhycjJjrKGPJBIJ27p1q7ZMVlYWA8BSU1MZY4zt3r2bCYVCVlhYqC2zevVq5urqyurq6sz7BOyYh4cH+/rrr6nPLFxFRQULDQ1lSUlJ7IknnmBz5sxhjNF77VEs5sxHfX090tPTMXToUO19QqEQQ4cORWpqKo+Rkabk5uaisLBQp8/c3NwQFRWl7bPU1FS4u7ujd+/e2jJDhw6FUCjE8ePHzR6zPSorKwMAtGrVCgCQnp4OpVKp02+dOnVCUFCQTr899thj8PHx0ZaJjo5GeXm59pc44Y5arUZiYiKqqqqgUCiozyxcXFwcRo4cqdM/AL3XHsViNpYrKSmBWq3W6QAA8PHxwcWLF3mKijxKYWEhADy0z/58rLCwEN7e3jqPi8VitGrVSluGcEej0WDu3Lno378/unXrBqChT6RSKdzd3XXKPthvD+vXPx8j3Dh79iwUCgVqa2shl8uxfft2dOnSBZmZmdRnFioxMRGnTp3CyZMnGz1G77WmWUzyQQgxvbi4OJw7dw5Hjx7lOxSih44dOyIzMxNlZWXYtm0bJk2ahOTkZL7DIk24ceMG5syZg6SkJDg4OPAdjlWxmMsunp6eEIlEjUYBFxUVwdfXl6eoyKP82S+P6jNfX99GA4ZVKhXu3r1L/cqxWbNm4ZdffsHBgwcREBCgvd/X1xf19fUoLS3VKf9gvz2sX/98jHBDKpWiQ4cOiIiIwNKlS9GjRw989tln1GcWKj09HcXFxejVqxfEYjHEYjGSk5Px+eefQywWw8fHh/qtCRaTfEilUkRERGD//v3a+zQaDfbv3w+FQsFjZKQpwcHB8PX11emz8vJyHD9+XNtnCoUCpaWlSE9P15Y5cOAANBoNoqKizB6zPWCMYdasWdi+fTsOHDiA4OBgnccjIiIgkUh0+u3SpUvIy8vT6bezZ8/qJI5JSUlwdXVFly5dzPNECDQaDerq6qjPLNSQIUNw9uxZZGZmam+9e/fGxIkTtf+mfmsC3yNe75eYmMhkMhnbsGEDu3DhAps+fTpzd3fXGQVMzKuiooJlZGSwjIwMBoAtX76cZWRksOvXrzPGGPvwww+Zu7s727FjBztz5gwbPXo0Cw4OZjU1Ndo6hg8fzsLDw9nx48fZ0aNHWWhoKJswYQJfT8nmzZw5k7m5ubFDhw6xgoIC7a26ulpbZsaMGSwoKIgdOHCApaWlMYVCwRQKhfZxlUrFunXrxoYNG8YyMzPZ3r17mZeXF1u4cCEfT8kuLFiwgCUnJ7Pc3Fx25swZtmDBAiYQCNi+ffsYY9Rn1uL+2S6MUb81xaKSD8YYW7lyJQsKCmJSqZRFRkayY8eO8R2SXTt48CAD0Og2adIkxljDdNt33nmH+fj4MJlMxoYMGcIuXbqkU8edO3fYhAkTmFwuZ66urmzy5MmsoqKCh2djHx7WXwDY+vXrtWVqampYbGws8/DwYE5OTmzs2LGsoKBAp55r166xp59+mjk6OjJPT0/25ptvMqVSaeZnYz+mTJnC2rZty6RSKfPy8mJDhgzRJh6MUZ9ZiweTD+q3hxMwxhg/51wIIYQQYo8sZswHIYQQQuwDJR+EEEIIMStKPgghhBBiVpR8EEIIIcSsKPkghBBCiFlR8kEIIYQQs6LkgxBCCCFmRckHIYQQQsyKkg9CCCGEmBUlH4QQQggxK0o+CCGEEGJWlHwQQgghxKz+H1J8IDF+vLkPAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(plate)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "contours, _ = cv2.findContours(plate, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_area = 0\n",
    "plate_area = plate.shape[0] * plate.shape[1]\n",
    "for contour in contours:\n",
    "    contour_area = cv2.contourArea(contour)\n",
    "    if contour_area < plate_area * 0.5:\n",
    "        mean_area += contour_area\n",
    "\n",
    "mean_area /= len(contours)    \n",
    "threshold_area = mean_area * 0.5\n",
    "\n",
    "bounding_boxes = []\n",
    "\n",
    "for contour in contours:\n",
    "    contour_area = cv2.contourArea(contour)\n",
    "    if contour_area >= plate_area * 0.7:\n",
    "        continue\n",
    "\n",
    "    if contour_area > threshold_area:\n",
    "        x, y, w, h = cv2.boundingRect(contour)\n",
    "        bounding_boxes.append((plate[y:y+h, x:x+w], x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "bounding_boxes.sort(key=lambda x: x[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for bounding_box in bounding_boxes:\n",
    "    plt.imshow(bounding_box[0])\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "reference_characters = read_reference_characters('./dataset/Lab07-Dataset')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.07665598290598291, '5')"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "recognise_character(reference_characters, bounding_boxes[0][0])"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}