{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# How To: Assess Image Quality Before Pipeline Design\n",
    "\n",
    "Run diagnostics on a plate image to objectively assess noise, contrast,\n",
    "and structure before choosing enhancers and detectors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from phenotypic.data import load_yeast_plate\n",
    "from phenotypic.plotting import PlotDiagnostics\n",
    "from phenotypic.util import ImageMetricsCalculator\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plate = load_yeast_plate()\n",
    "fig = PlotDiagnostics().inspect(plate)\n",
    "fig"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Decision Guide\n",
    "\n",
    "| Metric | Threshold | Action |\n",
    "|--------|-----------|--------|\n",
    "| Low SNR (< 10) | Noisy image | Add `StableDenoise` or `BlurGauss` |\n",
    "| Low RMS contrast | Faint colonies | Add `CLAHE` or `ContrastStretching` |\n",
    "| Low dynamic range | Under-exposed | Add `ContrastStretching` |\n",
    "| Low gradient mean | Soft edges | Add `UnsharpMask` or `FocusEdgeSobel` |\n",
    "| Long correlation length | Uneven illumination | Add `HomomorphicFilter` |"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "calculator = ImageMetricsCalculator(plate.detect_mat[:])\n",
    "metrics = {\n",
    "    \"noise\": calculator.compute_noise_metrics(),\n",
    "    \"contrast\": calculator.compute_contrast_metrics(),\n",
    "    \"structure\": calculator.compute_structure_metrics(),\n",
    "    \"background\": calculator.compute_background_metrics(),\n",
    "}\n",
    "for category, values in metrics.items():\n",
    "    print(f\"\\n{category}:\")\n",
    "    for key, val in values.items():\n",
    "        if isinstance(val, (int, float)):\n",
    "            print(f\"  {key}: {val:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.close(\"all\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3.11.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}