earth-observation-for-journalism/true-color-image/02 Spectral Indices.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before running this notebook the files have to be downloaded. See [01a Download from Scihub.ipynb](01a Download from Scihub.ipynb) for further information.\n",
    "\n",
    "The calculation in this notebook depends on three parameters:\n",
    "\n",
    "- `product_path` is the path to a previously downloaded product\n",
    "- `index_to_calculate` is one of the supported indices; these are 'ndvi'\n",
    "  - ndvi -- normalized difference in vegetation\n",
    "  - nbr -- normalized burn ratio\n",
    "  - bsi -- bare soil index\n",
    "- `fill_value`, the value which is used to represent invalid pixels to handle division by zero.\n",
    "\n",
    "Change the values below and select _Kernel → Restart and Run All Cells_ to re-evaluate all cells in this notebook.\n",
    "The path of the output file will be printed below the last cell."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "downloads = ['input/forest_fires/S2A_MSIL2A_20190613T101031_N0212_R022_T33UUT_20190614T125329.zip',\n",
    " 'input/forest_fires/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.zip',\n",
    " 'input/forest_fires/S2A_MSIL2A_20190626T102031_N0212_R065_T33UUT_20190626T125319.zip']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "product_path = Path(downloads[1])\n",
    "index_to_calculate = 'nbr'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Preview of product\n",
    "\n",
    "Uncomment the following lines to plot a true-color image of the downloaded data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#from sentinel_helpers import scihub_band_paths\n",
    "#import rasterio.plot as rplot\n",
    "#\n",
    "#tci = scihub_band_paths(product_path, 'TCI', '60m')[0]\n",
    "#with r.open(tci) as src:\n",
    "#    rplot.show(src)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define Formulas\n",
    "\n",
    "We define the formulas as data so we can substitute the bands with actual values later on and execute the operations when needed.\n",
    "We use a lisp-like language with prefix notation for this.\n",
    "\n",
    "### Operators\n",
    "\n",
    "Index calculations take the following form:\n",
    "\n",
    "$$\n",
    "NDVI = (B08 - B04) / (B08 + B04)\n",
    "$$\n",
    "\n",
    "This means they do pixel-wise operations on bands which we can express using numpy ndarrays.\n",
    "The ndarrays have overloaded operators, so for two arrays `array_a` and `array_b` with an identical shape `array_a + array_b` adds each cell in `array_a` with the corresponding cell in `array_b`.\n",
    "The exact behavior of numpy arrays in this regard is called [broadcasting](https://numpy.org/doc/stable/user/basics.broadcasting.html).\n",
    "\n",
    "We define the basic arithmetic operations `+`, `-`, `*` and `/` as functions taking variadic arguments:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from functools import reduce\n",
    "\n",
    "def add(nums):\n",
    "    return reduce(lambda a, b: a + b, nums)\n",
    "\n",
    "def sub(nums):\n",
    "    return reduce(lambda a, b: a - b, nums)\n",
    "\n",
    "def div(nums):\n",
    "    return reduce(lambda a, b: a / b, nums)\n",
    "\n",
    "def mul(nums):\n",
    "    return reduce(lambda a, b: a * b, nums)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Indices"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This section contains the formula definitions for a selection of indices.\n",
    "These definitions are not exhaustive - there are many spectral indices which are not implemented in this notebook - however the general shape of this formula allows for enough flexibility to implement other indices as well.\n",
    "\n",
    "The formulas are defined in a lisp-like prefix notation: `(add, 1, 2, 3)` translates to `1 + 2 + 3`.\n",
    "Each element in a formula can be either a function, a string or a tuple.\n",
    "A string represents a band number, a function represents an arithmetic operation and a tuple allows us to express nested formulas.\n",
    "\n",
    "We first define some spectral indices using these basic elements:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "indices = {\n",
    "    # normalized difference in vegetation\n",
    "    'ndvi': (div, (sub, 'B08', 'B04'), (add, 'B08', 'B04')),\n",
    "    # normalized burn ratio\n",
    "    'nbr': (div, (sub, 'B08', 'B12'), (add, 'B08', 'B12')),\n",
    "    # bare soil index\n",
    "    'bsi': (div, (sub, (add, 'B11', 'B04'), (add, 'B08', 'B02')),\n",
    "                 (add, (add, 'B11', 'B04'), (add, 'B08', 'B02')))\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "… and verify that the `index_to_calculate` above has a matching implementation:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "supported_indices = ', '.join(indices.keys())\n",
    "assert index_to_calculate in supported_indices, f'Only the following indices are supported: {supported_indices}'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Resolving the Formulas\n",
    "\n",
    "We define a function that returns all of the bands referenced in a formula.\n",
    "This is necessary when resolving these references to data sets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_bands(formula):\n",
    "    bands = set()\n",
    "    for element in formula:\n",
    "        if type(element) == tuple:\n",
    "            # recur for subformulas\n",
    "            for band in get_bands(element):\n",
    "                bands.add(band)\n",
    "        elif type(element) == str:\n",
    "            bands.add(element)\n",
    "    return bands"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The resolving process needs a `band_map` in the form of `band_num` → `numpy.array`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_formula(band_map, formula):\n",
    "    op = formula[0]\n",
    "    args = []\n",
    "    for element in formula[1:]:\n",
    "        if type(element) == tuple:\n",
    "            # recur on sub-formula\n",
    "            args.append(evaluate_formula(band_map, element))\n",
    "        elif type(element) == str:\n",
    "            # substitute band number\n",
    "            args.append(band_map[element])\n",
    "        else:\n",
    "            # just append the number\n",
    "            args.append(element)\n",
    "    return op(args)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We also define a function that makes working with the prefix notation a bit easier to avoid errors that can occur when translating the formulas:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prefix_to_infix_str(formula):\n",
    "    '''\n",
    "    Returns a human-readable string of the above data-based notation. Useful\n",
    "    for debugging.\n",
    "    '''\n",
    "    result = ['(']\n",
    "    \n",
    "    # operation\n",
    "    op = formula[0]\n",
    "    if op == add:\n",
    "        sym = ' + '\n",
    "    elif op == sub:\n",
    "        sym = ' - '\n",
    "    elif op == div:\n",
    "        sym = ' / '\n",
    "    elif op == mul:\n",
    "        sym = ' * '\n",
    "        \n",
    "    operands = formula[1:]\n",
    "    for idx, operand in enumerate(operands, start=1):\n",
    "        if type(operand) == tuple:\n",
    "            result.append(prefix_to_infix_str(operand))\n",
    "        else:\n",
    "            result.append(str(operand))\n",
    "        if idx < len(operands):\n",
    "            result.append(sym)\n",
    "        \n",
    "    result.append(')')\n",
    "    return ''.join(result)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Small test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "formula = (add, *range(5), (mul, *range(1, 4))) # == (add, 0, 1, 2, 3, 4, (mul, 1, 2, 3))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'(0 + 1 + 2 + 3 + 4 + (1 * 2 * 3))'"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prefix_to_infix_str(formula)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "16"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "evaluate_formula({}, formula)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Extraction of relevant band file paths\n",
    "\n",
    "We start by extracting the information which bands are referenced by the index formula given by `index_to_calculate`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['B12', 'B08']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bands = get_bands(indices[index_to_calculate])\n",
    "bands = list(bands)\n",
    "bands"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Resampling\n",
    "\n",
    "Some bands are not available in all resolutions.\n",
    "\n",
    "The band `B08` is available only at 10m and the band `B12` only at 20m:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['zip+file://input/forest_fires/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.zip!/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.SAFE/GRANULE/L2A_T33UUT_A020611_20190603T101041/IMG_DATA/R10m/T33UUT_20190603T101031_B08_10m.jp2',\n",
       " 'zip+file://input/forest_fires/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.zip!/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.SAFE/GRANULE/L2A_T33UUT_A020611_20190603T101041/IMG_DATA/R20m/T33UUT_20190603T101031_B12_20m.jp2',\n",
       " 'zip+file://input/forest_fires/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.zip!/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.SAFE/GRANULE/L2A_T33UUT_A020611_20190603T101041/IMG_DATA/R60m/T33UUT_20190603T101031_B12_60m.jp2']"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sentinel_helpers import scihub_band_paths\n",
    "band_paths = list(sorted(scihub_band_paths(product_path, bands)))\n",
    "band_paths"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This can be solved by upsampling the lower bands to the highest available resolution.\n",
    "First we discard duplicate bands which are already available in higher resolution:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['zip+file://input/forest_fires/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.zip!/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.SAFE/GRANULE/L2A_T33UUT_A020611_20190603T101041/IMG_DATA/R10m/T33UUT_20190603T101031_B08_10m.jp2',\n",
       " 'zip+file://input/forest_fires/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.zip!/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.SAFE/GRANULE/L2A_T33UUT_A020611_20190603T101041/IMG_DATA/R20m/T33UUT_20190603T101031_B12_20m.jp2']"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "highest_resolution_band_paths = []\n",
    "covered_bands = set()\n",
    "for band in band_paths:\n",
    "    band_num = band.split('_')[-2]\n",
    "    if band_num in covered_bands:\n",
    "        continue\n",
    "    else:\n",
    "        highest_resolution_band_paths.append(band)\n",
    "        covered_bands.add(band_num)\n",
    "highest_resolution_band_paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def resolution(band_path):\n",
    "    '''\n",
    "    Return the resolution encoded in a raster file's path.\n",
    "    '''\n",
    "    return int(band_path.split('_')[-1].split('.')[0].replace('m', ''))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The target resolution is given by the raster with the highest resolution. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "10"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "target_resolution = list(sorted(resolution(band) for band in highest_resolution_band_paths))[0]\n",
    "target_resolution"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Generate the raster cloud mask\n",
    "\n",
    "Spectral indices can get distorted by highly reflective clouds.\n",
    "These clouds need to be excluded.\n",
    "\n",
    "In order to construct a cloud mask that we can operate on, we need to know the transformation needed from the coordinates given by the raster's coordinate reference system to the pixel coordinates of the highest resolution raster.\n",
    "\n",
    "These are encoded as metadata in the raster file:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'zip+file://input/forest_fires/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.zip!/S2A_MSIL2A_20190603T101031_N0212_R022_T33UUT_20190603T114652.SAFE/GRANULE/L2A_T33UUT_A020611_20190603T101041/IMG_DATA/R10m/T33UUT_20190603T101031_B08_10m.jp2'"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sentinel_helpers import scihub_cloud_mask\n",
    "import rasterio as r\n",
    "import matplotlib.pyplot as pplt\n",
    "\n",
    "highres_raster_path = [band_path for band_path in highest_resolution_band_paths if resolution(band_path) == target_resolution][0]\n",
    "highres_raster_path"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((10980, 10980),\n",
       " Affine(10.0, 0.0, 300000.0,\n",
       "        0.0, -10.0, 5800020.0))"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "with r.open(highres_raster_path) as src:\n",
    "    target_transform = src.transform\n",
    "    target_shape = src.shape\n",
    "\n",
    "target_shape, target_transform"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fee81f18190>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# pixels with clouds are True, pixels without are False\n",
    "raster_cloud_mask = scihub_cloud_mask(product_path,\n",
    "                                      rasterize=True,\n",
    "                                      target_shape=target_shape,\n",
    "                                      target_transform=target_transform)\n",
    "\n",
    "pplt.imshow(raster_cloud_mask)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Calculate the index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import numpy.ma as ma\n",
    "\n",
    "from rasterio.enums import Resampling\n",
    "import rasterio.plot as rplt\n",
    "\n",
    "from sentinel_helpers import RasterReaderList, scihub_cloud_mask, scihub_normalize_range\n",
    "from tqdm.notebook import tqdm\n",
    "\n",
    "from tempfile import NamedTemporaryFile\n",
    "\n",
    "out_dir = Path('output/spectral_indices/')\n",
    "out_dir.mkdir(exist_ok=True, parents=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "with r.open(highest_resolution_band_paths[0]) as src:\n",
    "    data = src.read(1)\n",
    "    rplt.show(src)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.007747626915637307"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(data & raster_cloud_mask).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ef5d3e0206384ecb9dc60dd3712cfa2d",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HBox(children=(HTML(value='Resampling datasets'), FloatProgress(value=0.0, max=2.0), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "with RasterReaderList(highest_resolution_band_paths) as readers:\n",
    "    # build the band_map as described above and scale up where needed\n",
    "    band_map = {}\n",
    "    for reader in tqdm(readers, desc='Resampling datasets'):\n",
    "        band_num = reader.name.split('_')[-2]\n",
    "        data_resolution = resolution(reader.name)\n",
    "        scale_factor = int(data_resolution / target_resolution)\n",
    "        out_shape = (\n",
    "            int(reader.height * scale_factor),\n",
    "            int(reader.width * scale_factor)\n",
    "        )\n",
    "        \n",
    "        band_map[band_num] = (np.clip(\n",
    "            # we read only the first band to obtain a two-dimensional ndarray\n",
    "            reader.read(1, out_shape=out_shape, resampling=Resampling.bilinear, masked=True),\n",
    "        0, 10000) / 10000).astype('float32')\n",
    "        \n",
    "        if scale_factor == 1:\n",
    "            out_name = Path(reader.name).name.replace(band_num, index_to_calculate.upper())\n",
    "            out_name = out_name.replace('.jp2', '.tif')\n",
    "            out_path = out_dir / out_name\n",
    "            out_meta = reader.meta.copy()\n",
    "    \n",
    "    # ignore numpy division errors (i.e. divide by 0) in this context;\n",
    "    # divide by zero results in `nan`, which we fill with zeros. clouds are\n",
    "    # masked using the raster_cloud_mask created above\n",
    "    with np.errstate(divide='ignore', invalid='ignore'):\n",
    "        index = evaluate_formula(band_map, indices[index_to_calculate])\n",
    "        # we need to invert the cloud mask (`~`) because we want to hide those pixels that are cloudy\n",
    "        tile_mask = ~raster_cloud_mask | index.mask\n",
    "        index = ma.masked_array(index, tile_mask)\n",
    "    \n",
    "    out_meta.update({\n",
    "        'count': 1,\n",
    "        'driver': 'GTiff',\n",
    "        'dtype': 'float32'\n",
    "    })\n",
    "    \n",
    "    with r.open(out_path, 'w+', **out_meta) as dst:\n",
    "        dst.write(index.data, 1)\n",
    "        # note that the mask we write has an inverted interpretation of \n",
    "        # True / False due to the `.msk` file format\n",
    "        dst.write_mask(~index.mask)\n",
    "        print(f'Wrote result to {out_path}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to debug the cloud / tile mask, uncomment these lines\n",
    "# pplt.imshow(tile_mask)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with r.open(out_path, 'r') as src:\n",
    "    rplt.show(src)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# writing the mask explicitly works:\n",
    "# dst.write(index.data, 1)\n",
    "# dst.write_mask(~index.mask)"
   ]
  }
 ],
 "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": "3.8.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}