3_extract_thermal_ROIs.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "7961e77b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pickle\n",
    "from flir_thermal.flirio import flirVideo\n",
    "import matplotlib.pyplot as plt\n",
    "import tkinter as tk\n",
    "import os"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "dc7537e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "mouse = 'KE182'\n",
    "\n",
    "dates = [#'231231',\n",
    "        #'240102',\n",
    "        #'240103',\n",
    "        #'240104',\n",
    "        '240105',\n",
    "        '240106',\n",
    "        #'240107', >> didn't run\n",
    "        '240108',\n",
    "        '240109',\n",
    "        '240110',\n",
    "        '240112',\n",
    "        '240114',\n",
    "        '240115',\n",
    "        '240116',\n",
    "        '240117',\n",
    "        '240118',\n",
    "        #'240119', >> no thermal\n",
    "        '240120',\n",
    "        '240121',\n",
    "        '240122',\n",
    "        #'240123',\n",
    "        #'240124',\n",
    "        #'240125',\n",
    "        #'240126',\n",
    "        #'240127',\n",
    "        #'240128',\n",
    "        #'240129',\n",
    "        #'240130',\n",
    "        ]\n",
    "\n",
    "\n",
    "# Done:\n",
    "\n",
    "base = str('/mnt/nasquatch/data/2p/kathryn/' + mouse)\n",
    "\n",
    "fs = 60\n",
    "\n",
    "# Generate a list of paths \n",
    "seq_paths = []\n",
    "ROI_paths = []\n",
    "LED_save_paths = []\n",
    "snout_save_paths = []\n",
    "background_save_paths = []\n",
    "plot_save_paths = []\n",
    "\n",
    "for d in range(len(dates)):\n",
    "    date_mouse = str(dates[d] + '_' + mouse)\n",
    "    \n",
    "    new_path = str(base + '/' + date_mouse + '/')\n",
    "    seq_paths.append(new_path)\n",
    "    \n",
    "    new_ROI = str(new_path + date_mouse + '_expt/thermal_ROI.plk')\n",
    "    ROI_paths.append(new_ROI)\n",
    "    \n",
    "    new_LED = str(new_path + date_mouse + '_expt/LED_temp.plk')\n",
    "    LED_save_paths.append(new_LED)\n",
    "    \n",
    "    new_snout = str(new_path + date_mouse + '_expt/snout_temp.plk')\n",
    "    snout_save_paths.append(new_snout)\n",
    "    \n",
    "    new_background = str(new_path + date_mouse + '_expt/background_temp.plk')\n",
    "    background_save_paths.append(new_background)\n",
    "    \n",
    "    new_plot = str(new_path + date_mouse + '_expt/' + date_mouse + '_plot_of_thermal_ROIs.tiff')\n",
    "    plot_save_paths.append(new_plot)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "78692082",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['/mnt/nasquatch/data/2p/kathryn/KE182/240105_KE182/240105_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240106_KE182/240106_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240108_KE182/240108_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240109_KE182/240109_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240110_KE182/240110_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240112_KE182/240112_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240114_KE182/240114_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240115_KE182/240115_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240116_KE182/240116_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240117_KE182/240117_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240118_KE182/240118_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240120_KE182/240120_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240121_KE182/240121_KE182_expt/snout_temp.plk',\n",
       " '/mnt/nasquatch/data/2p/kathryn/KE182/240122_KE182/240122_KE182_expt/snout_temp.plk']"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "snout_save_paths"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "eb9ab865",
   "metadata": {},
   "outputs": [],
   "source": [
    "for p in range(len(seq_paths)):\n",
    "    \n",
    "    \n",
    "    # check if we have already extracted temp data for this date:\n",
    "    if os.path.isfile(snout_save_paths[p]) == True:\n",
    "        print(str('Already have extracted thermal ROIs for ' + dates[p]))\n",
    "    else:\n",
    "        # Load ROIs: \n",
    "        with open(ROI_paths[p], 'rb') as f: \n",
    "            ROIs = pickle.load(f)\n",
    "\n",
    "        LED_coords = ROIs[0]\n",
    "        LED_size = ROIs[1]\n",
    "        snout_coords = ROIs[2]\n",
    "        snout_size = ROIs[3]\n",
    "        background_coords = ROIs[4]\n",
    "        background_size= ROIs[5]\n",
    "\n",
    "        # set ROI boundaries\n",
    "        LED_x = slice(LED_coords[0],(LED_coords[0]+LED_size))\n",
    "        LED_y = slice(LED_coords[1],(LED_coords[1]+LED_size))\n",
    "        snout_x = slice(snout_coords[0],(snout_coords[0]+snout_size))\n",
    "        snout_y = slice(snout_coords[1],(snout_coords[1]+snout_size))\n",
    "        background_x = slice(background_coords[0],(background_coords[0]+background_size))\n",
    "        background_y = slice(background_coords[1],(background_coords[1]+background_size))\n",
    "\n",
    "\n",
    "        all_vids = {}\n",
    "        # Load all .seq video files in this folder \n",
    "        for f in os.listdir(seq_paths[p]): \n",
    "            if '.seq' in f: # find all .seq files\n",
    "                this_vid_path = str(seq_paths[p] + '/' + f) # File to load \n",
    "                # this takes forever, but not a ton of memory:\n",
    "                this_vid = flirVideo(this_vid_path) # Load video info \n",
    "\n",
    "        LED_temp  = []\n",
    "        snout_temp = []\n",
    "        background_temp = []\n",
    "\n",
    "        # by frame in this video, extract ROIs, average over space, append to our vars\n",
    "        for frame in range(1, this_vid.num_frames + 1):\n",
    "            index, this_frame, times = this_vid.read_video(frames = np.arange(frame, frame+1))\n",
    "            this_LED_temp = np.nanmean(np.nanmean(this_frame[0, LED_y, LED_x], axis = 0), axis = 0)\n",
    "            LED_temp.append(this_LED_temp)\n",
    "            this_s_temp = np.nanmean(np.nanmean(this_frame[0, snout_y, snout_x], axis = 0), axis = 0)\n",
    "            snout_temp.append(this_s_temp)\n",
    "            this_b_temp = np.nanmean(np.nanmean(this_frame[0, background_y, background_x], axis = 0), axis = 0)\n",
    "            background_temp.append(this_b_temp)\n",
    "\n",
    "        # Plot temp variables + save\n",
    "        f, ax = plt.subplots(1, 1)\n",
    "        plt.rcParams['figure.dpi'] = 600\n",
    "        plt.rcParams['figure.figsize'] = [4, 2]                     \n",
    "        ds = 10 # plot every ds-th point \n",
    "        plt.plot(LED_temp[::ds], c = 'blue')\n",
    "        plt.plot(snout_temp[::ds], c = 'red')\n",
    "        plt.plot(background_temp[::ds], c = 'green')\n",
    "        plt.ylim(15, 40)  \n",
    "        ax.legend(['IR LED', 'Snout', 'Background'])                \n",
    "        plt.ylabel('Temp (°C)')\n",
    "        plt.rcParams['font.size'] = 6                  \n",
    "        plt.savefig(plot_save_paths[p])\n",
    "\n",
    "        # Save our temperature variables: \n",
    "        with open(LED_save_paths[p], 'wb') as pic:\n",
    "            pickle.dump(LED_temp, pic)\n",
    "        pic.close()        \n",
    "        with open(snout_save_paths[p], 'wb') as pic:\n",
    "             pickle.dump(snout_temp, pic)\n",
    "        pic.close()\n",
    "        with open(background_save_paths[p], 'wb') as pic:\n",
    "             pickle.dump(background_temp, pic)\n",
    "        pic.close()       "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "cc50700c",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "bc5009c0",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "f42c9e44",
   "metadata": {},
   "outputs": [],
   "source": [
    " \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "86bf62e0",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d71cbc4a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "500db1a4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "31b5ac6b",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "89347bc5",
   "metadata": {},
   "outputs": [],
   "source": [
    "                          \n",
    "                    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "982d3c78",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "141b5dea",
   "metadata": {},
   "outputs": [],
   "source": [
    "    # Sort the video dict alphbetically, to put the runs in order \n",
    "    myKeys = list(all_vids.keys())\n",
    "    myKeys.sort()\n",
    "    sorted_all_vids = {i: all_vids[i] for i in myKeys}\n",
    "    del all_vids # free up memory \n",
    "    sorted_keys = sorted_all_vids.keys()\n",
    "\n",
    "    # Concatenate all videos in order \n",
    "    concat_vid = np.empty(shape = [0,256,320]) # ineffecient allocation\n",
    "    for this_key in sorted_keys:\n",
    "        this_vid = sorted_all_vids[this_key] # pull the video array\n",
    "        concat_vid = np.append(concat_vid, this_vid, axis = 0) # append it to our concatenated list\n",
    "        del this_vid\n",
    "    \n",
    "    del sorted_all_vids\n",
    "    \n",
    "    # Concat vid is frames by x by y \n",
    "    # now, extract thermal ROIs and average in space \n",
    "    LED_temp = np.mean(np.mean(concat_vid[:, LED_x, LED_y],1),1)\n",
    "    snout_temp = np.mean(np.mean(concat_vid[:, snout_x, snout_y],1),1)\n",
    "    background_temp = np.mean(np.mean(imarray[:, background_x, background_y],1),1)\n",
    "                         \n",
    "     \n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}