resize_speed.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a60562de",
   "metadata": {},
   "source": [
    "# Load & resize photos banchmark\n",
    "Before running benchmark, download the data from the repository https://github.com/ria-com/nomeroff-net/tree/master/banchmarks/images to the images folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "89845aea",
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "import os\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "398680f0",
   "metadata": {},
   "outputs": [],
   "source": [
    "dirName = \"./images\"\n",
    "testData = {\n",
    "    \"0.jpeg\": [\"AI5255EI\"],\n",
    "    \"1.jpeg\": [\"AT6883CM\"],\n",
    "    \"2.jpeg\": [\"AT1515CK\"],\n",
    "    \"3.jpeg\": [\"BX0578CE\"],\n",
    "    \"4.jpeg\": [\"AC4249CB\"],\n",
    "    \"5.jpeg\": [\"BC3496HC\"],\n",
    "    \"6.jpeg\": [\"BC3496HC\"],\n",
    "    \"7.jpeg\": [\"AO1306CH\"],\n",
    "    \"8.jpeg\": [\"AE1077CO\"],\n",
    "    \"9.jpeg\": [\"AB3391AK\"],\n",
    "    \"10.jpeg\": [\"BE7425CB\"],\n",
    "    \"11.jpeg\": [\"BE7425CB\"],\n",
    "    \"12.jpeg\": [\"AB0680EA\"],\n",
    "    \"13.jpeg\": [\"AB0680EA\"],\n",
    "    \"14.jpeg\": [\"BM1930BM\"],\n",
    "    \"15.jpeg\": [\"AI1382HB\"],\n",
    "    \"16.jpeg\": [\"AB7333BH\"],\n",
    "    \"17.jpeg\": [\"AB7642CT\"],\n",
    "    \"18.jpeg\": [\"AC4921CB\"],\n",
    "    \"19.jpeg\": [\"BC9911BK\"],\n",
    "    \"20.jpeg\": [\"BC7007AK\"],\n",
    "    \"21.jpeg\": [\"AB5649CI\"],\n",
    "    \"22.jpeg\": [\"AX2756EK\"],\n",
    "    \"23.jpeg\": [\"AA7564MX\"],\n",
    "    \"24.jpeg\": [\"AM5696CK\"],\n",
    "    \"25.jpeg\": [\"AM5696CK\"],\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "60096413",
   "metadata": {},
   "outputs": [],
   "source": [
    "def print_result(end_time, j):\n",
    "    print(f\"Processed {j} photos\")\n",
    "    print(f\"Time {end_time}\")\n",
    "    print(f\"One photo process {end_time/j} seconds\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "908f5f96",
   "metadata": {},
   "source": [
    "### Pillow"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3f232443",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed 26 photos\n",
      "Time 2.232311725616455\n",
      "One photo process 0.08585814329294059 seconds\n"
     ]
    }
   ],
   "source": [
    "from PIL import Image\n",
    "\n",
    "i = 0\n",
    "start_time = time.time()\n",
    "for fname in testData.keys():\n",
    "    img_path = os.path.join(dirName, fname)\n",
    "    im = Image.open(img_path)\n",
    "    (width, height) = (im.width // 2, im.height // 2)\n",
    "    img = np.asarray(im.resize((width, height)))\n",
    "    i += 1\n",
    "end_time = time.time() - start_time\n",
    "print_result(end_time, i)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7ff9f59",
   "metadata": {},
   "source": [
    "### Pillow-SIMD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "229a0bc7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed 26 photos\n",
      "Time 0.8349943161010742\n",
      "One photo process 0.03211516600388747 seconds\n"
     ]
    }
   ],
   "source": [
    "from PIL import Image\n",
    "\n",
    "i = 0\n",
    "start_time = time.time()\n",
    "for fname in testData.keys():\n",
    "    img_path = os.path.join(dirName, fname)\n",
    "    im = Image.open(img_path)\n",
    "    (width, height) = (im.width // 2, im.height // 2)\n",
    "    img = np.asarray(im.resize((width, height)))\n",
    "    i += 1\n",
    "end_time = time.time() - start_time\n",
    "print_result(end_time, i)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "855a2586",
   "metadata": {},
   "source": [
    "### OpenCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8b8545a2",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed 26 photos\n",
      "Time 1.2619729042053223\n",
      "One photo process 0.048537419392512396 seconds\n"
     ]
    }
   ],
   "source": [
    "import cv2\n",
    "\n",
    "i = 0\n",
    "start_time = time.time()\n",
    "for fname in testData.keys():\n",
    "    img_path = os.path.join(dirName, fname)\n",
    "    img = cv2.imread(img_path)\n",
    "    (width, height) = (img.shape[1] // 2, img.shape[0] // 2)\n",
    "    # resize image\n",
    "    img = cv2.resize(img, [width, height], interpolation = cv2.INTER_AREA)\n",
    "    i += 1\n",
    "end_time = time.time() - start_time\n",
    "print_result(end_time, i)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2d0b2993",
   "metadata": {},
   "source": [
    "### TurboJPEG"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "8aa4bcb3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed 26 photos\n",
      "Time 0.6053831577301025\n",
      "One photo process 0.023283967605003945 seconds\n"
     ]
    }
   ],
   "source": [
    "from turbojpeg import TurboJPEG\n",
    "jpeg = TurboJPEG()\n",
    "\n",
    "i = 0\n",
    "start_time = time.time()\n",
    "for fname in testData.keys():\n",
    "    img_path = os.path.join(dirName, fname)\n",
    "    with open(img_path, 'rb') as in_file:\n",
    "        img = jpeg.decode(in_file.read())\n",
    "        (width, height) = (img.shape[1] // 2, img.shape[0] // 2)\n",
    "        img = cv2.resize(img, [width, height], interpolation = cv2.INTER_AREA)\n",
    "    i += 1\n",
    "end_time = time.time() - start_time\n",
    "print_result(end_time, i)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2f56e4ee",
   "metadata": {},
   "source": [
    "### TurboJPEG superfast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "184e840e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Processed 26 photos\n",
      "Time 0.4682457447052002\n",
      "One photo process 0.018009451719430778 seconds\n"
     ]
    }
   ],
   "source": [
    "from turbojpeg import TurboJPEG\n",
    "jpeg = TurboJPEG()\n",
    "\n",
    "i = 0\n",
    "start_time = time.time()\n",
    "for fname in testData.keys():\n",
    "    img_path = os.path.join(dirName, fname)\n",
    "    with open(img_path, 'rb') as in_file:\n",
    "        # Only for scaling_factor (13, 8), (7, 4), (3, 8), (1, 2), (2, 1), (15, 8), (3, 4), (5, 8), (5, 4), (1, 1), (1, 8), (1, 4), (9, 8), (3, 2), (7, 8), (11, 8)\n",
    "        img = jpeg.decode(in_file.read(), scaling_factor=(1, 2))\n",
    "    i += 1\n",
    "end_time = time.time() - start_time\n",
    "print_result(end_time, i)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "41c38274",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python ",
   "language": "python",
   "name": "py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}