003 - Color Spaces.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### RGB es BGR en OpenCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2\n",
    "import numpy as np\n",
    "\n",
    "image=cv2.imread(\"images/bus_stop/1.jpg\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos a mirar los niveles de colores individuales para el primer pixel (0,0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "232 236 241\n",
      "(800, 1200, 3)\n"
     ]
    }
   ],
   "source": [
    "#BRG values for the firt 0,0 pixel\n",
    "B,G,R=image[0,0]\n",
    "print(B,G,R)\n",
    "print(image.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Que pasas cuando se convierte a escala de grises?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(800, 1200)\n",
      "237\n"
     ]
    }
   ],
   "source": [
    "gray_img=cv2.cvtColor(image,cv2.COLOR_BGR2GRAY)\n",
    "print(gray_img.shape)\n",
    "print(gray_img[0,0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ahora son solo 2 dimensiones, cada pixel  tiene un solo valor (antes 3) con un rango entre 0 y 255"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Otro espacio de color util es HSV\n",
    "Muy util en color filtering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# H:0-180, S:0-255, V:0-255\n",
    "\n",
    "image=cv2.imread(\"images/cafeteria/01-breeze-outdoor-dining.jpg\")\n",
    "\n",
    "hsv_image=cv2.cvtColor(image,cv2.COLOR_BGR2HSV)\n",
    "\n",
    "cv2.imshow(\"HSV image\",hsv_image)\n",
    "cv2.imshow(\"Hue channel\",hsv_image[:,:,0])\n",
    "cv2.imshow(\"Saturation channel\",hsv_image[:,:,1])\n",
    "cv2.imshow(\"Value channel\",hsv_image[:,:,2])\n",
    "\n",
    "cv2.waitKey()\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(960, 1280)\n"
     ]
    }
   ],
   "source": [
    "#### Vamos a explorar canales individuales en RGB\n",
    "image=cv2.imread(\"images/classroom/1017896781_f3c075c551_o.jpg\")\n",
    "\n",
    "#OpenCV's split funtion splites the image into each color index\n",
    "B,G,R=cv2.split(image)\n",
    "\n",
    "print(B.shape)\n",
    "\n",
    "cv2.imshow(\"Red\",R)\n",
    "cv2.imshow(\"Green\",G)\n",
    "cv2.imshow(\"Blue\",B)\n",
    "cv2.imshow(\"Original\",image)\n",
    "\n",
    "cv2.waitKey(0)\n",
    "cv2.destroyAllWindows()\n",
    "\n",
    "#Vamos a re-hacer la imagen original\n",
    "merged=cv2.merge([B,G,R])\n",
    "cv2.imshow(\"Merged\",merged)\n",
    "\n",
    "#Vamos ahora a amplificar el color azul\n",
    "merged=cv2.merge([B+100,G,R])\n",
    "cv2.imshow(\"Merged with Blue amplified\",merged)\n",
    "\n",
    "cv2.waitKey(0)\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Vamos a crear una matriz de ceros con las dimensiones de la imagen\n",
    "zeros=np.zeros(image.shape[:2],dtype=\"uint8\")\n",
    "\n",
    "cv2.imshow(\"Red\",cv2.merge([zeros,zeros,R]))\n",
    "cv2.imshow(\"Green\",cv2.merge([zeros,G,zeros]))\n",
    "cv2.imshow(\"Blue\",cv2.merge([B,zeros,zeros]))\n",
    "\n",
    "cv2.waitKey()\n",
    "cv2.destroyAllWindows()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(960, 1280)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "image.shape[:2]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}