walls.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "Yx3R0SBwBiJB",
    "tags": []
   },
   "outputs": [],
   "source": [
    "import cv2 as cv\n",
    "from matplotlib import pyplot as plt\n",
    "from ipywidgets import interactive\n",
    "import image\n",
    "import mapa\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "import glob\n",
    "from collections import Counter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load image, resize and convert"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "FctJO_IKCFCH",
    "tags": []
   },
   "outputs": [],
   "source": [
    "img = cv.imread('./iamges/new-walls.jpg')\n",
    "img = cv.cvtColor(img, cv.COLOR_BGR2RGB)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 623
    },
    "id": "tohOt1oKCsuK",
    "outputId": "81a8be4d-a39c-4b25-c3d0-01c82e629120",
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "calibration = np.load('rig/calib.npz')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "mtx = calibration['mtx']\n",
    "dist = calibration['dist']\n",
    "rvecs = calibration['rvecs']\n",
    "tvecs = calibration['tvecs']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "h, w = img.shape[:2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "newCameraMtx, roi = cv.getOptimalNewCameraMatrix(mtx, dist, (w, h), 1, (w, h))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "undistortedImg = cv.undistort(img, mtx, dist, None, newCameraMtx)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(undistortedImg)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot histograms of color spaces, both rgb and hsv"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Color tweaking\n",
    "\n",
    "Find parameters for extracting the mask (high and low colors for range limits) using sliders and looking for the best match."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "hsv_img = cv.cvtColor(img, cv.COLOR_RGB2HSV)\n",
    "def mask(h, sl, vl, sh, vh):\n",
    "    m = cv.inRange(\n",
    "        hsv_img,\n",
    "        (h-10, sl, vl),\n",
    "        (h+10, sh, vh),\n",
    "    )\n",
    "    plt.imshow(m)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.imshow(hsv_img[:,:,0])\n",
    "plt.colorbar()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "interactive(\n",
    "    mask,\n",
    "    h=(0, 179),\n",
    "    sl=(0, 255),\n",
    "    vl=(0, 255),\n",
    "    sh=(0, 255),\n",
    "    vh=(0, 255),\n",
    ")"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Create the mask"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "BLUE_LOWER = (90, 95, 87)\n",
    "BLUE_UPPER = (110, 255, 255)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ROI corner finder\n",
    "\n",
    "Find maze corners in specific regions of interest"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " method reference; to index tPlot corners to a new image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "image_class = image.Image(undistortedImg)\n",
    "image_class.set_areas(\n",
    "    image.Area(50, 150),\n",
    "    image.Area(350, 450),\n",
    "    image.Area(125, 225),\n",
    "    image.Area(450, 550),\n",
    ")\n",
    "image_class.set_origin_points((0, 99), (0, 99))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(image_class.img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "hsv_class = image_class.rgb_to_hsv()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(hsv_class.img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "mask_class = hsv_class.make_mask(BLUE_LOWER, BLUE_UPPER)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(mask_class.img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(np.max(mask_class.img))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "np.where(mask_class.img.flatten())[0].size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "corners = mask_class.find_corners()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "corners"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "debug_class = image_class.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "debug_class.paint_corners(corners)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(debug_class.img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "fix1_class = image_class.fix_perspective(corners, (10, 10), (24, 24))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "fix1_class.set_areas(\n",
    "    image.Area(0, 50),\n",
    "    image.Area(250, 300),\n",
    "    image.Area(0, 50),\n",
    "    image.Area(250, 300),\n",
    ")\n",
    "fix1_class.set_origin_points(\n",
    "    (0, 49),\n",
    "    (0, 49),\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(fix1_class.img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#plt.imshow(cv.cvtColor(fix1_class.img, cv.COLOR_RGB2HSV)[:,:,0])\n",
    "plt.imshow(fix1_class.img[:,:,0] > 100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(fix1_class.img[10:19, 0:9])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "fix1_mapa = mapa.Mapa.from_image(fix1_class.img, (10, 10), (24, 24))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "map_image = image.Image.from_map(fix1_mapa)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "print(fix1_class.img.dtype, fix1_class.img.shape)\n",
    "#print(type(map_image.img))\n",
    "print(map_image.img.dtype, map_image.img.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "plt.imshow(map_image.img)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "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": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}