Example classifying raster imagery using scikit-learn for imagery classification

Unsupervised imagery classification.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unsupervised classification of imagery using `scikit-learn`\n",
    "\n",
    "\n",
    "This example shows how to classify imagery (for example from LANDSAT) using [`scikit-learn`](http://scikit-learn.org/stable/). There are many classification methods available, but for this example we will use [K-Means](http://scikit-learn.org/stable/modules/clustering.html#k-means) as it's simple and fast. For imagery I grabbed the [North Carolina dataset](http://grassbook.org/datasets/datasets-3rd-edition/) raster sample and I'm using the red, green and blue bands of the landsat 7 imagery within this pacakge. I'm using [`rasterio`](https://github.com/mapbox/rasterio) to read in the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import rasterio\n",
    "import sklearn.cluster"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "red_path = r\"C:\\projects\\quick_scripts\\gis_se\\landsat\\ncrast\\lsat7_2002_30.tif\"\n",
    "green_path = r\"C:\\projects\\quick_scripts\\gis_se\\landsat\\ncrast\\lsat7_2002_20.tif\"\n",
    "blue_path = r\"C:\\projects\\quick_scripts\\gis_se\\landsat\\ncrast\\lsat7_2002_10.tif\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3L, 475L, 527L)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with rasterio.open(red_path) as red, rasterio.open(green_path) as green, rasterio.open(blue_path) as blue:\n",
    "    data = np.array([red.read(1), green.read(1), blue.read(1)])\n",
    "\n",
    "data.shape # Note that this is a three band image giving us a 3 dimensional array."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x155ef898>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {