vizCow

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Complex shapes in Underworld: Voxels, binvox and the Cow\n",
    "\n",
    "This notebook describes a possible method for importing complex geometries onto Underworld particles\n",
    "\n",
    "Given a supported 3D model file format (based on binvox):\n",
    "  VRML V2.0: almost fully supported\n",
    "  UG, OBJ, OFF, DXF, XGL, POV, BREP, PLY, JOT: only polygons supported\n",
    "\n",
    "The workflow is:\n",
    " 1. Use **binvox** to convert 3D model file into a voxel binary file.\n",
    " 2. Use numpy to read in the voxel file.\n",
    " 3. Setup an uw particle distribution.\n",
    " 4. Use scipy to interpolate (ndimage) the voxel field onto the particles"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Input 3D model:  \n",
    "downloaded from http://tf3dm.com/3d-model/cow-39025.html\n",
    "\n",
    "**Note**: We can also use 3D software like 'blender' to manipulate the model file\n",
    "\n",
    "Binvox  \n",
    "downloaded http://www.patrickmin.com/binvox/\n",
    "\n",
    "Run this command on the input image  \n",
    "`./binvox -c -t vtk <filename>.obj`\n",
    "\n",
    "This will output a file named `<filename>.vtk`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from scipy import ndimage\n",
    "import underworld as uw\n",
    "import glucifer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# read in the vtk file, create a binary file from it\n",
    "prefix = '/home/julian/Downloads/theCow/cleanCow'\n",
    "with open(prefix+'.vtk') as fp, open(prefix+'.binary','w') as ff:\n",
    "    \n",
    "    for line in fp:\n",
    "        if line.startswith(\"DIMENSIONS\"):\n",
    "            shape = line.partition(\" \")[2].rstrip().split(\" \")\n",
    "        if line.startswith(\"LOOKUP_TABLE default\"):\n",
    "            # stop reading ascii\n",
    "            break\n",
    "            \n",
    "    # create new binary file\n",
    "    for line in fp:\n",
    "        ff.write(line)     "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# size of binary voxel field\n",
    "shape = [int(ii) for ii in shape]\n",
    "shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# read in binary information into a numpy array\n",
    "wtf = np.fromfile(prefix+'.binary', dtype=np.uint8 )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# UW part - create particle distribution across geometry\n",
    "mesh = uw.mesh.FeMesh_Cartesian(elementRes=(10,10,10), \n",
    "                                minCoord=(0.0,0.0,0.0),\n",
    "                                maxCoord=(150.0,260.0,150.0) )\n",
    "\n",
    "swarm = uw.swarm.Swarm(mesh)\n",
    "colour = swarm.add_variable('double',count=1)\n",
    "layout = uw.swarm.layouts.GlobalSpaceFillerLayout(swarm, particlesPerCell=50)\n",
    "swarm.populate_using_layout(layout)\n",
    "\n",
    "colour.data[:]=0.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# reshape the coordinates appropriately\n",
    "cow = wtf.reshape((shape[0],shape[1],shape[2]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# evaluate the cow coordinates at the particle coordinates - order is 0 but can be [0 - 5] for spline interpolation\n",
    "mapping = ndimage.map_coordinates(cow, \n",
    "                              [swarm.particleCoordinates.data[:,0].T-50,\n",
    "                               swarm.particleCoordinates.data[:,1].T,\n",
    "                               swarm.particleCoordinates.data[:,2].T], \n",
    "                               order=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# how many unique numbers \n",
    "np.unique(mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "np.histogram(mapping)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "colour.data[:]=mapping.reshape(-1,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# output as xdmf file to visualise\n",
    "sH = swarm.save('swarm.h5')\n",
    "cH = colour.save('colour.h5')\n",
    "\n",
    "colour.xdmf('colour', cH, 'colour',sH,'swarm')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.11+"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}