biggest_area_selector.ipynb

biggest_area_selector.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "biggest_area_selector.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyPq2Hm3tJ9Rqc0ublJpzU+b",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/Feyn-Man/fefd9a38224f1f1e8ca1b18dcafb06d1/biggest_area_selector.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6h1Gg-p_c9BO"
      },
      "source": [
        "import numpy as np\r\n",
        "from scipy import ndimage"
      ],
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "kEfYfD7idHbF"
      },
      "source": [
        "def biggest_area_selector(arr):\r\n",
        "    \r\n",
        "    \"\"\"\r\n",
        "    Keep the connected component with the biggest area inside of a numpy array.\r\n",
        "\r\n",
        "    Example:\r\n",
        "    > arr = np.array([[1,1,1],\r\n",
        "                      [0,1,0],\r\n",
        "                      [0,0,0],\r\n",
        "                      [2,2,1]])\r\n",
        "    \r\n",
        "    > biggest_area_selector(arr)\r\n",
        "    array([[1, 1, 1],\r\n",
        "           [0, 1, 0],\r\n",
        "           [0, 0, 0],\r\n",
        "           [0, 0, 0]])\r\n",
        "    \"\"\"\r\n",
        "\r\n",
        "    label_arr, _ = ndimage.label(arr)\r\n",
        "    nonzero_labels = np.unique(label_arr)[np.unique(label_arr) > 0]\r\n",
        "    areas = np.bincount(label_arr.reshape((label_arr.size, )))\r\n",
        "    max_label = nonzero_labels[np.argmax(areas[1:])]\r\n",
        "    return np.where(label_arr == max_label, arr, 0)"
      ],
      "execution_count": 2,
      "outputs": []
    }
  ]
}