Classification_EigenDecomposition_Stack_SEMESTER.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/home/jovyan/SInCohMap_EigenClassification/./Datasets/\n",
      "/home/jovyan/SInCohMap_EigenClassification/./Classification_maps/\n",
      "Appending /home/jovyan/SInCohMap_EigenClassification/./Datasets/WIELKOPOLSKA_RR_4x19_UTM_2016-01-01_2016-07-01_eigValuesVV.tiff file...\n",
      "Appending /home/jovyan/SInCohMap_EigenClassification/./Datasets/WIELKOPOLSKA_RR_4x19_UTM_2016-07-01_2016-12-31_eigValuesVV.tiff file...\n",
      "Appending /home/jovyan/SInCohMap_EigenClassification/./Datasets/WIELKOPOLSKA_RR_4x19_UTM_2016-01-01_2016-07-01_eigValuesVH.tiff file...\n",
      "Appending /home/jovyan/SInCohMap_EigenClassification/./Datasets/WIELKOPOLSKA_RR_4x19_UTM_2016-07-01_2016-12-31_eigValuesVH.tiff file...\n",
      "Elapsed time loading: \n",
      "2.474269151687622\n",
      "/home/jovyan/Training_Data\n",
      "\n",
      "        ################################\n",
      "        General info about the shapefile\n",
      "        ################################\n",
      "        \n",
      "Dataset driver is: ESRI Shapefile\n",
      "\n",
      "The shapefile has 1 layer(s)\n",
      "\n",
      "The layer is named: SInCohMap_WW_Validation_RR_WGS84_H33\n",
      "\n",
      "The layer's geometry is: Polygon\n",
      "\n",
      "Layer projection is: +proj=utm +zone=33 +datum=WGS84 +units=m +no_defs \n",
      "\n",
      "Layer has 2573 features\n",
      "\n",
      "Layer has 6 fields\n",
      "Their names are: \n",
      "\tClass_name - String\n",
      "\tclass lev1 - String\n",
      "\tCODE_LEV_1 - String\n",
      "\tCODE_LEV_3 - String\n",
      "\tLEVEL_1 - Integer64\n",
      "\tLEVEL_3 - Integer64\n",
      "../Training_Data/SInCohMap_WW_Validation_RR_WGS84_H33.shp\n",
      "/home/jovyan/SInCohMap_EigenClassification/./Datasets/WIELKOPOLSKA_RR_4x19_UTM_stacked_2016S1S2_VVVH.tiff\n",
      "Success\n",
      "Class 0.0 contains 10000 pixels\n",
      "Class 112.0 contains 67103 pixels\n",
      "Class 141.0 contains 29 pixels\n",
      "Class 142.0 contains 1308 pixels\n",
      "Class 211.0 contains 524363 pixels\n",
      "Class 311.0 contains 6689 pixels\n",
      "Class 312.0 contains 149963 pixels\n",
      "Class 313.0 contains 107627 pixels\n",
      "Class 321.0 contains 113179 pixels\n",
      "Class 411.0 contains 2874 pixels\n",
      "Class 512.0 contains 16865 pixels\n",
      "Elapsed time training: \n",
      "2.150825023651123\n",
      "\n",
      "Starting Random Forest classification, lean back and wait for the magic to happen :) \n",
      "GridSearchCV choose the best parameter combination for the classification as following:\n",
      "RandomForestClassifier(bootstrap=True, class_weight=None, criterion='gini',\n",
      "            max_depth=None, max_features=5, max_leaf_nodes=None,\n",
      "            min_impurity_split=1e-07, min_samples_leaf=1,\n",
      "            min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
      "            n_estimators=300, n_jobs=-1, oob_score=True, random_state=None,\n",
      "            verbose=0, warm_start=False)\n",
      "\n",
      "Elapsed time classifying: \n",
      "27229.88026547432\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.6/site-packages/matplotlib/font_manager.py:1320: UserWarning: findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans\n",
      "  (prop.get_family(), self.defaultFamily[fontext]))\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVgAAAFYCAYAAAAWbORAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f28e44038d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Classification report: \n",
      "precision = producer accuracy \n",
      "recall = user accuracy \n",
      "             precision    recall  f1-score   support\n",
      "\n",
      "Class 112.0       0.63      0.64      0.63       671\n",
      "Class 141.0       0.93      0.97      0.95        29\n",
      "Class 142.0       0.91      0.51      0.65       100\n",
      "Class 211.0       0.83      0.93      0.88      5244\n",
      "Class 311.0       0.95      0.21      0.34       100\n",
      "Class 312.0       0.56      0.81      0.66      1500\n",
      "Class 313.0       0.44      0.21      0.29      1076\n",
      "Class 321.0       0.45      0.22      0.30      1132\n",
      "Class 411.0       0.86      0.24      0.38       100\n",
      "Class 512.0       0.91      0.62      0.74       169\n",
      "\n",
      "avg / total       0.69      0.72      0.69     10121\n",
      "\n",
      "Feature ranking:\n",
      "1. feature 10 (0.041955)\n",
      "2. feature 26 (0.038348)\n",
      "3. feature 27 (0.036060)\n",
      "4. feature 25 (0.035322)\n",
      "5. feature 28 (0.031277)\n",
      "6. feature 9 (0.030939)\n",
      "7. feature 24 (0.030881)\n",
      "8. feature 8 (0.023795)\n",
      "9. feature 11 (0.023404)\n",
      "10. feature 12 (0.023375)\n",
      "11. feature 22 (0.022761)\n",
      "12. feature 23 (0.022624)\n",
      "13. feature 1 (0.021298)\n",
      "14. feature 0 (0.021243)\n",
      "15. feature 29 (0.019144)\n",
      "16. feature 7 (0.019060)\n",
      "17. feature 20 (0.017933)\n",
      "18. feature 40 (0.016967)\n",
      "19. feature 6 (0.016314)\n",
      "20. feature 34 (0.016175)\n",
      "21. feature 21 (0.015883)\n",
      "22. feature 36 (0.015302)\n",
      "23. feature 16 (0.015279)\n",
      "24. feature 17 (0.014861)\n",
      "25. feature 33 (0.014824)\n",
      "26. feature 19 (0.014753)\n",
      "27. feature 47 (0.014615)\n",
      "28. feature 35 (0.014200)\n",
      "29. feature 15 (0.014056)\n",
      "30. feature 32 (0.013937)\n",
      "31. feature 46 (0.013879)\n",
      "32. feature 37 (0.013864)\n",
      "33. feature 18 (0.013763)\n",
      "34. feature 56 (0.013677)\n",
      "35. feature 5 (0.013546)\n",
      "36. feature 49 (0.013339)\n",
      "37. feature 45 (0.013078)\n",
      "38. feature 55 (0.012960)\n",
      "39. feature 3 (0.012714)\n",
      "40. feature 4 (0.012657)\n",
      "41. feature 50 (0.012518)\n",
      "42. feature 48 (0.012484)\n",
      "43. feature 57 (0.012206)\n",
      "44. feature 2 (0.012147)\n",
      "45. feature 52 (0.012115)\n",
      "46. feature 54 (0.012113)\n",
      "47. feature 38 (0.012062)\n",
      "48. feature 39 (0.011941)\n",
      "49. feature 41 (0.011938)\n",
      "50. feature 13 (0.011937)\n",
      "51. feature 14 (0.011708)\n",
      "52. feature 31 (0.011580)\n",
      "53. feature 51 (0.011308)\n",
      "54. feature 53 (0.011183)\n",
      "55. feature 30 (0.010950)\n",
      "56. feature 44 (0.010694)\n",
      "57. feature 42 (0.010586)\n",
      "58. feature 43 (0.010469)\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfIAAAFgCAYAAABAP/uYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f28c5d75e80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Overall Classification accuracy: 0.716826\n",
      "\n",
      "Kappa score: 0.563290\n",
      "\n",
      "Mc Nemars test\n",
      "Chi-square 824.000000 P-value 0.000000\n",
      "\n",
      "Everything worked out just fine, congrats :)\n"
     ]
    }
   ],
   "source": [
    "from geo_utils import write_geotiff\n",
    "from landcover_classification import prepare_training_data, classification, validation, plot_classified_image \n",
    "import os, shutil\n",
    "import eig_wcs\n",
    "import numpy as np\n",
    "from osgeo import gdal\n",
    "import time\n",
    "\n",
    "start = time.time()\n",
    "\n",
    "## ---------------------------------------------------------------------------------------------------------------\n",
    "## ---------------------------------------------------------------------------------------------------------------\n",
    "## -------- CLASSIFICATION: VV EIGENVALUES BASED --------------------------------------------------------------------\n",
    "## ---------------------------------------------------------------------------------------------------------------\n",
    "## ---------------------------------------------------------------------------------------------------------------\n",
    "selSet = \"WIELKOPOLSKA_RR_4x19_UTM\"\n",
    "\n",
    "class_method = 'rf' # or 'svm'\n",
    "\n",
    "RR_reference_dir = '../Training_Data/'\n",
    "#vector_path = os.path.join(RR_reference_dir,'SInCohMap_Donyana_Validation_RR_WGS84_H30_vf.shp')\n",
    "#vector_path = os.path.join(RR_reference_dir,'SInCohMap_STyrol_Validation_RR_WGS84_H32.shp')\n",
    "vector_path = os.path.join(RR_reference_dir,'SInCohMap_WW_Validation_RR_WGS84_H33.shp')\n",
    "\n",
    "idir = os.path.join(os.getcwd(), './Datasets/')\n",
    "odir = os.path.join(os.getcwd(),'./Classification_maps/')\n",
    "\n",
    "print(idir)\n",
    "print(odir)\n",
    "\n",
    "#geo_proj = inout.getProjection(selSet)\n",
    "#geo_transf = inout.getGeoTransform(selSet)\n",
    "\n",
    "datafiles = [os.path.join(idir,selSet + '_' + '2016-01-01_2016-07-01' + '_eigValuesVV.tiff'), \\\n",
    "             os.path.join(idir,selSet + '_' + '2016-07-01_2016-12-31' + '_eigValuesVV.tiff'), \\\n",
    "             os.path.join(idir,selSet + '_' + '2016-01-01_2016-07-01' + '_eigValuesVH.tiff'), \\\n",
    "             os.path.join(idir,selSet + '_' + '2016-07-01_2016-12-31' + '_eigValuesVH.tiff')]\n",
    "\n",
    "tmdatafile = os.path.join(idir,selSet + '_stacked_2016S1S2_VVVH.tiff')\n",
    "                          \n",
    "tmdata = []\n",
    "for ff in datafiles:\n",
    "    print('Appending ' + ff + ' file...')\n",
    "    raster_dataset = gdal.Open(ff, gdal.GA_ReadOnly)\n",
    "    geo_proj = raster_dataset.GetProjection()\n",
    "    geo_transf = raster_dataset.GetGeoTransform()\n",
    "    for b in range(1, raster_dataset.RasterCount + 1):\n",
    "        data = raster_dataset.GetRasterBand(b)\n",
    "        tmdata.append(data.ReadAsArray())\n",
    "                          \n",
    "tmdata = np.dstack(tmdata)\n",
    "write_geotiff(tmdatafile, tmdata, geo_transf, geo_proj)\n",
    "\n",
    "column      = 'LEVEL_3'\n",
    "nr_points   = 10000\n",
    "sampling_methodology = 'proportional'\n",
    "\n",
    "load_end = time.time()\n",
    "print('Elapsed time loading: ')\n",
    "print(load_end - start)\n",
    "\n",
    "## SAMPLE\n",
    "[training_samples, training_labels, bands_data, projection, geo_transform, is_test, test_labels, test_samples] = prepare_training_data(vector_path, tmdatafile, column, nr_points, sampling_methodology)\n",
    "\n",
    "training_end = time.time()\n",
    "print('Elapsed time training: ')\n",
    "print(training_end - load_end)\n",
    "\n",
    "## CLASSIFY\n",
    "[classified_image, classifier] = classification(training_samples, training_labels, test_labels, test_samples, bands_data, geo_proj, geo_transf, classifier=class_method)\n",
    "\n",
    "classif_end = time.time()\n",
    "print('Elapsed time classifying: ')\n",
    "print(classif_end-training_end)\n",
    "\n",
    "oimage = 'classified_' + column + '_S' + str(nr_points) + sampling_methodology + '_' + class_method + '_' + os.path.basename(tmdatafile)\n",
    "shutil.move('classified_image.tiff',os.path.join(odir, oimage))\n",
    "\n",
    "## VALIDATE\n",
    "\n",
    "validation(classified_image, is_test, test_labels, training_samples, classifier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}