{
 "metadata": {
  "name": "",
  "signature": "sha256:6c055a506f9a0cfccbfd514af3f400081ff70957843f6abdc7ccb0fe51ec5757"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": [],
     "source": [
      "> This is one of the 100 recipes of the [IPython Cookbook](http://ipython-books.github.io/), the definitive guide to high-performance scientific computing and data science in Python.\n"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "# 11.3. Segmenting an image"
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "1. Let's import the packages."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import numpy as np\n",
      "import matplotlib.pyplot as plt\n",
      "from skimage.data import coins\n",
      "from skimage.filter import threshold_otsu\n",
      "from skimage.segmentation import clear_border\n",
      "from skimage.morphology import closing, square\n",
      "from skimage.measure import regionprops, label\n",
      "from skimage.color import lab2rgb\n",
      "import matplotlib as mpl\n",
      "%matplotlib inline"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "2. We create a function to display a grayscale image."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def show(img, cmap=None):\n",
      "    cmap = cmap or plt.cm.gray\n",
      "    plt.figure(figsize=(4,2));\n",
      "    plt.imshow(img, cmap=cmap);\n",
      "    plt.axis('off');\n",
      "    plt.show();"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "3. We retrieve a test image bundled in scikit-image, showing various coins on a plain background."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "img = coins()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "show(img)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "4. The first step to segment the image consists in finding an intensity threshold separating the (bright) coins from the (dark) background. **Otsu's method** defines a simple algorithm to find such a threshold automatically."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "threshold_otsu(img)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "show(img>107)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "5. There appears to be a problem in the top left corner of the image, with part of the background being too bright. Let's use the notebook widgets to find a better threshold."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "from IPython.html import widgets\n",
      "@widgets.interact(t=(10, 240))\n",
      "def threshold(t):\n",
      "    show(img>t)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "6. The threshold 120 looks better. The next step consists in cleaning the binary image by smoothing the coins and removing the border. Scikit-image contains a few functions for these purposes."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "img_bin = clear_border(closing(img>120, square(5)))\n",
      "show(img_bin)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "7. Next, we perform the segmentation task itself with the `label` function. This function detects the connected components in the image, and attributes a unique label to every component. Here, we color-code the labels in the binary image."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "labels = label(img_bin)\n",
      "show(labels, cmap=plt.cm.rainbow)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "8. Small artifacts in the image result in spurious labels that do not correspond to coins. Therefore we only keep components with more than 100 pixels. The `regionprops` function allows us to retrieve specific properties of the components (here, the area and the bounding box)."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "regions = regionprops(labels, \n",
      "                      ['Area', 'BoundingBox'])\n",
      "boxes = np.array([label['BoundingBox'] for label in regions \n",
      "                                       if label['Area'] > 100])\n",
      "print(\"There are {0:d} coins.\".format(len(boxes)))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "9. Finally, we show the label number on top of each component in the original image."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "plt.figure(figsize=(6,4));\n",
      "plt.imshow(img, cmap=plt.cm.gray);\n",
      "plt.axis('off');\n",
      "xs = boxes[:,[1,3]].mean(axis=1)\n",
      "ys = boxes[:,[0,2]].mean(axis=1)\n",
      "for i, box in enumerate(boxes):\n",
      "    plt.text(xs[i]-5, ys[i]+5, str(i))"
     ],
     "language": "python",
     "metadata": {},
     "outputs": []
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "> You'll find all the explanations, figures, references, and much more in the book (to be released later this summer).\n",
      "\n",
      "> [IPython Cookbook](http://ipython-books.github.io/), by [Cyrille Rossant](http://cyrille.rossant.net), Packt Publishing, 2014 (500 pages)."
     ]
    }
   ],
   "metadata": {}
  }
 ]
}