imagej_server_demo.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## NB: Start an ImageJ server before you run this!\n",
    "\n",
    "```\n",
    "git clone git://github.com/imagej/imagej-server\n",
    "cd image-server\n",
    "mvn -Pexec\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import StringIO\n",
    "\n",
    "import imagej\n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "ij = imagej.IJ()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find modules that contain a specific string in its ID.\n",
    "create = ij.find('CreateImgFromImg')[0]\n",
    "invert = ij.find('InvertII')[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Details for CreateImgFromImg:\n",
      "{\n",
      "    \"inputs\": [\n",
      "        {\n",
      "            \"columnCount\": 6, \n",
      "            \"name\": \"in\", \n",
      "            \"maximumValue\": null, \n",
      "            \"defaultValue\": null, \n",
      "            \"required\": true, \n",
      "            \"label\": \"\", \n",
      "            \"widgetStyle\": \"\", \n",
      "            \"stepSize\": null, \n",
      "            \"choices\": null, \n",
      "            \"softMinimum\": null, \n",
      "            \"minimumValue\": null, \n",
      "            \"genericType\": \"net.imglib2.img.Img<capture of ?>\", \n",
      "            \"softMaximum\": null\n",
      "        }\n",
      "    ], \n",
      "    \"identifier\": \"command:net.imagej.ops.create.img.CreateImgFromImg\", \n",
      "    \"outputs\": [\n",
      "        {\n",
      "            \"columnCount\": 6, \n",
      "            \"name\": \"out\", \n",
      "            \"maximumValue\": null, \n",
      "            \"defaultValue\": null, \n",
      "            \"required\": true, \n",
      "            \"label\": \"\", \n",
      "            \"widgetStyle\": \"\", \n",
      "            \"stepSize\": null, \n",
      "            \"choices\": null, \n",
      "            \"softMinimum\": null, \n",
      "            \"minimumValue\": null, \n",
      "            \"genericType\": \"net.imglib2.img.Img<capture of ?>\", \n",
      "            \"softMaximum\": null\n",
      "        }\n",
      "    ]\n",
      "}\n",
      "Details for InvertII:\n",
      "{\n",
      "    \"inputs\": [\n",
      "        {\n",
      "            \"columnCount\": 6, \n",
      "            \"name\": \"out\", \n",
      "            \"maximumValue\": null, \n",
      "            \"defaultValue\": null, \n",
      "            \"required\": true, \n",
      "            \"label\": \"\", \n",
      "            \"widgetStyle\": \"\", \n",
      "            \"stepSize\": null, \n",
      "            \"choices\": null, \n",
      "            \"softMinimum\": null, \n",
      "            \"minimumValue\": null, \n",
      "            \"genericType\": \"net.imglib2.IterableInterval<capture of ?>\", \n",
      "            \"softMaximum\": null\n",
      "        }, \n",
      "        {\n",
      "            \"columnCount\": 6, \n",
      "            \"name\": \"in\", \n",
      "            \"maximumValue\": null, \n",
      "            \"defaultValue\": null, \n",
      "            \"required\": true, \n",
      "            \"label\": \"\", \n",
      "            \"widgetStyle\": \"\", \n",
      "            \"stepSize\": null, \n",
      "            \"choices\": null, \n",
      "            \"softMinimum\": null, \n",
      "            \"minimumValue\": null, \n",
      "            \"genericType\": \"net.imglib2.IterableInterval<capture of ?>\", \n",
      "            \"softMaximum\": null\n",
      "        }\n",
      "    ], \n",
      "    \"identifier\": \"command:net.imagej.ops.image.invert.InvertII\", \n",
      "    \"outputs\": [\n",
      "        {\n",
      "            \"columnCount\": 6, \n",
      "            \"name\": \"out\", \n",
      "            \"maximumValue\": null, \n",
      "            \"defaultValue\": null, \n",
      "            \"required\": true, \n",
      "            \"label\": \"\", \n",
      "            \"widgetStyle\": \"\", \n",
      "            \"stepSize\": null, \n",
      "            \"choices\": null, \n",
      "            \"softMinimum\": null, \n",
      "            \"minimumValue\": null, \n",
      "            \"genericType\": \"net.imglib2.IterableInterval<capture of ?>\", \n",
      "            \"softMaximum\": null\n",
      "        }\n",
      "    ]\n",
      "}\n"
     ]
    }
   ],
   "source": [
    "# Check details of a module. Names of \"inputs\" and \"outputs\" are usually important.\n",
    "print('Details for CreateImgFromImg:')\n",
    "print(json.dumps(ij.detail(create), indent=4))\n",
    "\n",
    "print('Details for InvertII:')\n",
    "print(json.dumps(ij.detail(invert), indent=4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "u'object:jdyv1ot5l09mdjcs'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Upload an image.\n",
    "img_in = ij.upload('data/images/cells-16bit.tif')\n",
    "img_in"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Execute modules.\n",
    "result = ij.run(create, {'in': img_in})\n",
    "img_out = result['out']\n",
    "\n",
    "result = ij.run(invert, {'in': img_in, 'out': img_out})\n",
    "img_out = result['out']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Retrieve images.\n",
    "cells_png = ij.retrieve(img_out, format='png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def data2image(data):\n",
    "    return Image.open(StringIO.StringIO(data))\n",
    "\n",
    "def show(image):\n",
    "    plt.imshow(image, interpolation='nearest', cmap='gray')\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ae1c0d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# convert image data to PIL image.\n",
    "cells_pil = data2image(cells_png)\n",
    "\n",
    "# We can also convert a PIL image to numpy array as needed.\n",
    "cells_np = np.array(cells_pil)\n",
    "show(cells_np)"
   ]
  }
 ],
 "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": "2.7.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}