Ford_Video_Analysis.ipynb

{
 "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OqSmFS3lZj9Y"
      },
      "source": [
        "# <center> Video Image Data </center>\n",
        "#### CMSE 495 Ford Group\n",
        "\n",
        "This tutorial teaches the user how to input a video file, such a mp4 and convert each frame of the video into a jpeg image using python, primarily in a Jupyter notebook."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uey1neRTkRwd"
      },
      "source": [
        "[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/pathakis/DataTools_Tutorial_Demo/blob/main/Video-Image-Data-Tutorial/Ford_Video_Analysis.ipynb)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GQIJxk_kdjT1"
      },
      "source": [
        "<b> Environment Setup (Makefile):</b>\n",
        "- Use the command 'make innit' automatically set up the environment for you.\n",
        "\n",
        "<b> Environment Setup (Manual):</b>\n",
        "- Set up new environment using pip/conda (Conda Recommended). Use command \n",
        "\n",
        "    <code> conda create -n envs python=3.10 </code>\n",
        "\n",
        "- Activate your new environment. Use command \n",
        "\n",
        "    <code> conda activate envs</code>\n",
        "\n",
        "- Install the requisite packages.Use command \n",
        "\n",
        "    <code> pip install opencv-python</code> or,\n",
        "\n",
        "    <code> conda install -c conda-forge opencv</code>\n",
        "\n",
        "<b> Usage Instructions:</b>\n",
        "\n",
        "- The example call shows the format in which this func may be used.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f0aUW4PLdobE"
      },
      "source": [
        "This process uses 2  packages called [os](https://docs.python.org/3/library/os.html)  and [cv2](https://pypi.org/project/opencv-python/). Os provides  miscellaneous operating system interfaces such as opening and reading the files."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "JGkN_k3BgXV8"
      },
      "outputs": [],
      "source": [
        "# !pip install opencv-python"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "PCbpVR-HZzmt"
      },
      "outputs": [],
      "source": [
        "import cv2\n",
        "import os\n",
        "import glob\n",
        "import urllib.request\n",
        "\n",
        "\n",
        "def video_to_frames(file_path, directory_path, greyscale = False):\n",
        "\n",
        "  '''This function will change a video file to a frames'''\n",
        "    \n",
        "    #opening the video\n",
        "  vidcap = cv2.VideoCapture(file_path) \n",
        "    \n",
        "  dirname = directory_path\n",
        "  os.makedirs(dirname, exist_ok=True)\n",
        "  \n",
        "  #capturing a frame as well as a boolean value representing whether an image was properly opened\n",
        "  success,image = vidcap.read()\n",
        "  \n",
        "  count = 0\n",
        "  \n",
        "  while success:\n",
        "      \n",
        "    #this is specifically for foam_segmented.avi\n",
        "    if greyscale:\n",
        "        image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)\n",
        "        \n",
        "    #writing the image to a the directory path that was specified, \n",
        "    #if the path specified does not exist then it will be created\n",
        "    #this finctionality was added so that the images could be stored in a separate folder\n",
        "    #example of output file names: 1.jpg, 2.jpg, 3.jpg, and so on\n",
        "    cv2.imwrite(os.path.join(dirname, str(count)+\".jpg\"), image)\n",
        "    success,image = vidcap.read()\n",
        "    count += 1\n",
        "    #All the frames will be added in order\n",
        "    cv2.waitKey(1) \n",
        "  \n",
        "  #releasing the threads\n",
        "    vidcap.release()\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "hBfLcvMhh6v7"
      },
      "source": [
        "DEMO FOR THE **avi_to_frames** \n",
        "\n",
        "1. To download a sample avi file that you want to work with, use the following code `urllib.request.urlretrieve('https://file-examples.com/wp-content/uploads/2018/04/file_example_AVI_480_750kB.avi', 'testing.mp4')` \n",
        "\n",
        "2. After the video has been downloaded `avi_frames(./testing.mp4, path_to_where_you_want_the_frames, False)` this will create a folder with frames from the video."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "-N-hJD11jJjo"
      },
      "outputs": [],
      "source": [
        "# Making a Video From Frames\n",
        "def frames_to_video(directory_path, fps, width, height):\n",
        "  fourcc = cv2.VideoWriter_fourcc(*'mp4v')\n",
        "  video = cv2.VideoWriter('video.avi', fourcc, fps, (width, height))\n",
        "  num_frames = len([name for name in os.listdir(directory_path) if os.path.isfile(name)])\n",
        "\n",
        "  for j in range(num_frames):\n",
        "    img = cv2.imread(str(j) + '.jpg')\n",
        "    video.write(img)\n",
        "\n",
        "  cv2.destroyAllWindows()\n",
        "  video.release()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sIhvc4DplLkE"
      },
      "source": [
        "DEMO FOR THE **frames_to_video** \n",
        "\n",
        "1. We will be working with the frames that we created using `avi_to_frames`. If you have not created those frames feel free to look at the steps above.\n",
        "\n",
        "2. Run the following command (**make the necessary changes in the function call**)`frames_to_video(where_the_frames_are, fps, width, height)`\n",
        "\n",
        "3. The video will show up in the current directory."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bnfzxPNJeZVS"
      },
      "source": [
        "**The code below will put the image arrays into a list.** This snippet of code utilizes glob but packages like os can also be used."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "mLbqPW7SeJd-"
      },
      "outputs": [],
      "source": [
        "path = glob.glob(\"./*.jpg\")\n",
        "images = []\n",
        "for img in path:\n",
        "  n = cv2.imread(img)\n",
        "  images.append(n)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GIq-_h4wdxWM"
      },
      "source": [
        "<b> References:</b>\n",
        "- [Managing Environments](https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html)\n",
        "- [Open CV in python](https://pypi.org/project/opencv-python/)\n",
        "- [Colab Button](https://www.youtube.com/watch?v=RoGZIbwzG5w)"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "name": "Ford_Video_Analysis.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "base",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.9.15"
    },
    "vscode": {
      "interpreter": {
        "hash": "4f3567101b31d35f97cf2856951ebbba1e09a3f852422478e736adda2bb3beee"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}