Python/CalculateSharpnessOneImage.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# QC script to calculate sharpness metric for images in a plate"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The following script takes an image from the OMERO server.\n",
    "It adds increasing levels of blurriness to it and then\n",
    "calculates a 'sharpness' score for the images.\n",
    "It highlights the start of an investigation for a workflow\n",
    "that might be used to identify out-of-focus images across\n",
    "a collection of images."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Workflow summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![Overview](SharpnessWorkflow.jpg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Import packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from skimage import feature\n",
    "from scipy.ndimage import convolve\n",
    "from scipy import misc\n",
    "from scipy import ndimage\n",
    "from ipywidgets import widgets\n",
    "from IPython.display import display"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create a connection to the OMERO Server"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from omero.gateway import BlitzGateway\n",
    "\n",
    "from getpass import getpass\n",
    "\n",
    "HOST = 'workshop.openmicroscopy.org'\n",
    "PORT = 4064\n",
    "conn = BlitzGateway(input(\"Username: \"),\n",
    "                    getpass(\"OMERO Password: \"),\n",
    "                    host=HOST, port=PORT)\n",
    "conn.connect()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Define the 3 algorithms we are going to test\n",
    "The algorithms were sourced from:\n",
    "* gradientBased - [stackoverflow: Detect which image is sharper](https://stackoverflow.com/questions/6646371/detect-which-image-is-sharper/26014796)\n",
    "* fourierBased - [Image Sharpness Measure for Blurred Images in Frequency Domain](https://doi.org/10.1016/j.proeng.2013.09.086)\n",
    "* edgeBased - Canny Edge Detection algorithm scipy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "class AlgorithmList:\n",
    "    def gradientBasedSharpnessMetric(self):\n",
    "        gy, gx = np.gradient(plane)\n",
    "        gnorm = np.sqrt(gx**2 + gy**2)\n",
    "        sharpness = np.average(gnorm)\n",
    "        return sharpness\n",
    "\n",
    "    def fourierBasedSharpnessMetric(self):\n",
    "        fftimage = np.fft.fft2(plane)\n",
    "        fftshift = np.fft.fftshift(fftimage)\n",
    "        fftshift = np.absolute(fftshift)\n",
    "        M = np.amax(fftshift)\n",
    "        Th = (fftshift > M // float(1000)).sum()\n",
    "        if 'image' in locals():\n",
    "            sharpness = Th // (float(image.getSizeX()) * float(image.getSizeY()))\n",
    "            return sharpness*10000\n",
    "        else:\n",
    "            return Th\n",
    "\n",
    "    def edgeBasedSharpnessMetric(self):\n",
    "        edges1 = feature.canny(plane, sigma=3)\n",
    "        kernel = np.ones((3, 3))\n",
    "        kernel[1, 1] = 0\n",
    "        sharpness = convolve(edges1, kernel, mode=\"constant\")\n",
    "        sharpness = sharpness[edges1 != 0].sum()\n",
    "        return sharpness\n",
    "\n",
    "\n",
    "print(\"loaded:\", dir(AlgorithmList))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Choose the algorithm to test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def dropdown_widget(Algorithm_list,\n",
    "                    dropdown_widget_name,\n",
    "                    displaywidget=False):\n",
    "\n",
    "    alg_sel = widgets.Dropdown(\n",
    "        options=Algorithm_list,\n",
    "        value=Algorithm_list[0],\n",
    "        description=dropdown_widget_name,\n",
    "        disabled=False,\n",
    "    )\n",
    "    if displaywidget is True:\n",
    "        display(alg_sel)\n",
    "    return alg_sel\n",
    "\n",
    "\n",
    "Algorithm = dropdown_widget(\n",
    "    ['Gradient', 'Fourier', 'Edge'],\n",
    "    'Algorithm: ', True\n",
    ")\n",
    "\n",
    "# SELECT THE METHOD THEN MOVE TO THE NEXT CELL WITHOUT RUNNING THE CELL AGAIN"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Example Image from Scipy to test Algorithms\n",
    "",
    "Now create increasing levels of Gaussian blur on an image\n",
    "obtained from Scipy package.\n",
    "Calculate the sharpness of the original and blurred images\n",
    "using the chosen algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "method = Algorithm.value\n",
    "if method == 'Gradient':\n",
    "    sharpness_method = AlgorithmList().gradientBasedSharpnessMetric\n",
    "elif method == 'Fourier':\n",
    "    sharpness_method = AlgorithmList().fourierBasedSharpnessMetric\n",
    "elif method == 'Edge':\n",
    "    sharpness_method = AlgorithmList().edgeBasedSharpnessMetric\n",
    "\n",
    "resultArray = np.zeros((5, 2), dtype=float)\n",
    "plt.figure(figsize=(20, 15))\n",
    "cntr = 1\n",
    "for sigValue in range(0, 20, 4):\n",
    "    face = misc.face(gray=True)\n",
    "    plane = ndimage.gaussian_filter(face, sigma=sigValue)\n",
    "    plt.subplot(1, 5, cntr)\n",
    "    plt.imshow(plane, cmap=plt.cm.gray)\n",
    "    plt.axis('off')\n",
    "    sharpness = sharpness_method()\n",
    "    resultArray[cntr - 1, 1] = sharpness\n",
    "    resultArray[cntr - 1, 0] = sigValue\n",
    "    cntr = cntr + 1\n",
    "\n",
    "plt.show()\n",
    "plt.figure(figsize=(15, 8))\n",
    "plt.plot(resultArray[:, 0], resultArray[:, 1], 'ro')\n",
    "plt.title(method)\n",
    "plt.xlabel('Levels of gaussian blur')\n",
    "plt.ylabel('Sharpness score')\n",
    "plt.show()\n",
    "plt.gcf().clear()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fetch OMERO Image from the server"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ex: Select an Image and enter its Id\n",
    "imageId = 9397\n",
    "image = conn.getObject(\"Image\", imageId)\n",
    "print(image.getName(), image.getDescription())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now create increasing levels of Gaussian blur on an OMERO image.\n",
    "Calculate the sharpness of the original and blurred images using\n",
    "the chosen algorithm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "pixels = image.getPrimaryPixels()\n",
    "image_plane = pixels.getPlane(0, 0, 0)\n",
    "\n",
    "resultArray = np.zeros((5, 2), dtype=float)\n",
    "plt.figure(figsize=(20, 15))\n",
    "cntr = 1\n",
    "for sigValue in range(0, 20, 4):\n",
    "    face = misc.face(gray=True)\n",
    "    plane = ndimage.gaussian_filter(image_plane, sigma=sigValue)\n",
    "    plt.subplot(1, 5, cntr)\n",
    "    plt.imshow(plane, cmap=plt.cm.gray)\n",
    "    plt.axis('off')\n",
    "    sharpness = sharpness_method()\n",
    "    resultArray[cntr - 1, 1] = sharpness\n",
    "    resultArray[cntr - 1, 0] = sigValue\n",
    "    cntr = cntr + 1\n",
    "\n",
    "plt.show()\n",
    "plt.figure(figsize=(15, 8))\n",
    "plt.plot(resultArray[:, 0], resultArray[:, 1], 'ro')\n",
    "plt.title(method)\n",
    "plt.xlabel('Levels of gaussian blur')\n",
    "plt.ylabel('Sharpness score')\n",
    "plt.show()\n",
    "plt.gcf().clear()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Close the connection to the OMERO server"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "conn.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### License\n",
    "Copyright (C) 2019 University of Dundee. All Rights Reserved.\n",
    "This program is free software; you can redistribute it and/or modify it\n",
    "under the terms of the GNU General Public License as published by the\n",
    "Free Software Foundation; either version 2 of the License, or\n",
    "(at your option) any later version.\n",
    "This program is distributed in the hope that it will be useful, but\n",
    "WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY\n",
    "or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for\n",
    "more details. You should have received a copy of the GNU General\n",
    "Public License along with this program; if not, write to the\n",
    "Free Software Foundation,\n",
    "Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "OMERO Python 3",
   "language": "python",
   "name": "python3"
  },
  "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
}