satellite_pose_estimation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "#importing libraries for image processing\n",
    "\n",
    "import keras\n",
    "import numpy as np\n",
    "import random\n",
    "from PIL import Image\n",
    "import cv2\n",
    "import os\n",
    "import pandas as pd\n",
    "import warnings\n",
    "import matplotlib.pyplot as plt\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "#importing libraries for building model\n",
    "\n",
    "from __future__ import print_function\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten\n",
    "from keras.layers import Conv2D, MaxPooling2D\n",
    "from keras import backend as K\n",
    "from keras import losses"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Loads the images from training directory\n",
    "\n",
    "def load_images_from_folder(folder):\n",
    "    images = []\n",
    "    for filename in os.listdir(folder):\n",
    "        img = cv2.imread(os.path.join(folder,filename))\n",
    "        if img is not None:\n",
    "            images.append(img)\n",
    "    return images\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Pre-processing the input\n",
    "\n",
    "def preprocess_images(images):\n",
    "    images = np.array(images, dtype=np.float32)\n",
    "    images = images.astype('float32')\n",
    "    images /= 255\n",
    "    return images\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Read the coordinates of the joints (labels)\n",
    "\n",
    "def read_labels(label_file):\n",
    "    df = pd.read_csv(label_file)\n",
    "    numpy_matrix = df.as_matrix()\n",
    "    return numpy_matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Model\n",
    "\n",
    "def build_model():\n",
    "    model = Sequential()\n",
    "    model.add(Conv2D(55, (55,96),\n",
    "                     activation='relu',\n",
    "                     input_shape=input_shape))\n",
    "    model.add(MaxPooling2D(pool_size=(2, 2), strides=(2, 2)))\n",
    "    model.add(Conv2D(27, (27,32), activation='relu'))\n",
    "    model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "    model.add(Conv2D(13, (13,12), activation='relu'))\n",
    "    model.add(MaxPooling2D(pool_size=(2, 2)))\n",
    "    model.add(Flatten())\n",
    "    model.add(Dropout(0.25))    \n",
    "    model.add(Dense(2048, activation='relu'))\n",
    "    model.add(Dense(2048, activation='relu'))\n",
    "    model.add(Dropout(0.5))\n",
    "    model.add(Dense(12, activation='relu'))\n",
    "    return model\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#directories for train_data and test_data\n",
    "\n",
    "train_img_path = '/home/nani/Satellite_pose_Estimation/training_data/X_train'\n",
    "validation_img_path = '/home/nani/Satellite_pose_Estimation/validation_data/X_validation'\n",
    "test_img_path = '/home/nani/Satellite_pose_Estimation/test_data/X_test'\n",
    "\n",
    "train_lbl_path = '/home/nani/Satellite_pose_Estimation/training_data/y_train/y_train.csv'\n",
    "validation_lbl_path = '/home/nani/Satellite_pose_Estimation/validation_data/y_validation/y_validation.csv'\n",
    "test_lbl_path = '/home/nani/Satellite_pose_Estimation/test_data/y_test/y_test.csv'\n",
    "\n",
    "input_shape = (256, 256, 3)\n",
    "\n",
    "## input image dimensions\n",
    "img_rows, img_cols = 256, 256\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/nani/anaconda3/envs/satellite_pose_estimation/lib/python3.7/site-packages/ipykernel_launcher.py:5: FutureWarning: Method .as_matrix will be removed in a future version. Use .values instead.\n",
      "  \"\"\"\n"
     ]
    }
   ],
   "source": [
    "#Read training and testing data\n",
    "\n",
    "#X\n",
    "train_images = load_images_from_folder(train_img_path)\n",
    "validation_images = load_images_from_folder(validation_img_path)\n",
    "test_images  = load_images_from_folder(test_img_path)\n",
    "\n",
    "#y\n",
    "y_train = read_labels(train_lbl_path)\n",
    "y_validation = read_labels(validation_lbl_path)\n",
    "y_test = read_labels(test_lbl_path)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "#Preprocessing the input data\n",
    "\n",
    "x_train = preprocess_images(train_images)\n",
    "x_validation = preprocess_images(validation_images) \n",
    "x_test = preprocess_images(test_images)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/nani/anaconda3/envs/satellite_pose_estimation/lib/python3.7/site-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Colocations handled automatically by placer.\n",
      "WARNING:tensorflow:From /home/nani/anaconda3/envs/satellite_pose_estimation/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n"
     ]
    }
   ],
   "source": [
    "#Build model\n",
    "model = build_model()\n",
    "model.compile(loss=keras.losses.mean_squared_error,\n",
    "              optimizer='adam',\n",
    "              metrics=['accuracy'])\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WARNING:tensorflow:From /home/nani/anaconda3/envs/satellite_pose_estimation/lib/python3.7/site-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
      "Instructions for updating:\n",
      "Use tf.cast instead.\n",
      "Train on 80 samples, validate on 10 samples\n",
      "Epoch 1/10\n",
      "80/80 [==============================] - 189s 2s/step - loss: 97619.5122 - acc: 0.0750 - val_loss: 17131.7930 - val_acc: 0.2000\n",
      "Epoch 2/10\n",
      "80/80 [==============================] - 183s 2s/step - loss: 17430.4893 - acc: 0.0875 - val_loss: 17093.0156 - val_acc: 0.2000\n",
      "Epoch 3/10\n",
      "80/80 [==============================] - 183s 2s/step - loss: 17311.2886 - acc: 0.1125 - val_loss: 16774.9766 - val_acc: 0.0000e+00\n",
      "Epoch 4/10\n",
      "80/80 [==============================] - 184s 2s/step - loss: 16350.3839 - acc: 0.0750 - val_loss: 14098.5215 - val_acc: 0.0000e+00\n",
      "Epoch 5/10\n",
      "80/80 [==============================] - 186s 2s/step - loss: 12312.5363 - acc: 0.0500 - val_loss: 8692.7793 - val_acc: 0.2000\n",
      "Epoch 6/10\n",
      "80/80 [==============================] - 183s 2s/step - loss: 6479.2479 - acc: 0.1375 - val_loss: 5227.1904 - val_acc: 0.2000\n",
      "Epoch 7/10\n",
      "80/80 [==============================] - 184s 2s/step - loss: 3763.9762 - acc: 0.1500 - val_loss: 2559.9434 - val_acc: 0.2000\n",
      "Epoch 8/10\n",
      "80/80 [==============================] - 183s 2s/step - loss: 1967.5698 - acc: 0.0750 - val_loss: 1793.2426 - val_acc: 0.1000\n",
      "Epoch 9/10\n",
      "80/80 [==============================] - 184s 2s/step - loss: 1183.4892 - acc: 0.2250 - val_loss: 1622.2321 - val_acc: 0.4000\n",
      "Epoch 10/10\n",
      "80/80 [==============================] - 182s 2s/step - loss: 1043.1078 - acc: 0.4000 - val_loss: 1674.1647 - val_acc: 0.4000\n"
     ]
    }
   ],
   "source": [
    "#Fit the data\n",
    "history = model.fit(x_train, y_train,\n",
    "          batch_size=10,\n",
    "          epochs=10,\n",
    "          verbose=1,\n",
    "          validation_data=(x_validation, y_validation))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r",
      "10/10 [==============================] - 7s 725ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[500.95465087890625, 0.699999988079071]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Evaluate test_Data\n",
    "model.evaluate(x_test, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['val_loss', 'val_acc', 'loss', 'acc'])\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# list all data in history\n",
    "print(history.history.keys())\n",
    "# summarize history for accuracy\n",
    "plt.plot(history.history['acc'])\n",
    "plt.plot(history.history['val_acc'])\n",
    "plt.title('model accuracy')\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()\n",
    "# summarize history for loss\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('model loss')\n",
    "plt.ylabel('loss')\n",
    "plt.xlabel('epoch')\n",
    "plt.legend(['train', 'validation'], loc='upper left')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}