3.5-yolov2-train-hands-dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 雙手 (Hands)偵測 - YOLOv2模型訓練與調整\n",
    "\n",
    "在[3.0: YOLO物體偵測演算法概念與介紹](https://github.com/erhwenkuo/deep-learning-with-keras-notebooks/blob/master/3.0-yolo-algorithm-introduction.ipynb)的文章裡介紹了YOLO演算法一些重要概念與其它物體偵測演算法的不同之處。\n",
    "\n",
    "這一篇文章則是要手把手地介紹使用[basic-yolo-keras](https://github.com/erhwenkuo/basic-yolo-keras)來將Darknet預訓練的模型進行再訓練與調整來偵測人在做動的雙手 (Hands)。\n",
    "\n",
    "![hands-detection](http://cvrr.ucsd.edu/vivachallenge/wp-content/uploads/2015/01/example-1024x576.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## basic-yolo-keras 專案說明\n",
    "\n",
    "[basic-yolo-keras](https://github.com/experiencor/basic-yolo-keras)包含在Keras中使用Tensorflow後端的YOLOv2演算方的實現。它支持訓練YOLOv2網絡與不同的網絡結構，如MobileNet和InceptionV3。\n",
    "\n",
    "由於原始的專案[experiencor/basic-yolo-keras](https://github.com/experiencor/basic-yolo-keras)主要以英文來說明演算法的實現與再訓練， 對於一些剛入門深度學習的學習者來說，有一些不容易理解與入手的情況。\n",
    "\n",
    "因此在這個專案的方向在於文件說明的中文化以外，同時也會以一些便於使用與學習的角度進行源碼的調整與修正。\n",
    "\n",
    "### 需求\n",
    "\n",
    "- [Keras](https://github.com/fchollet/keras)\n",
    "- [Tensorflow](https://www.tensorflow.org/)\n",
    "- [Numpy](http://www.numpy.org/)\n",
    "- [h5py](http://www.h5py.org/) (For Keras model serialization.)\n",
    "- [Pillow](https://pillow.readthedocs.io/) (For rendering test results.)\n",
    "- [Python 3](https://www.python.org/)\n",
    "- [pydot-ng](https://github.com/pydot/pydot-ng) (Optional for plotting model.)\n",
    "\n",
    "### 安裝\n",
    "\n",
    "```bash\n",
    "git clone https://github.com/erhwenkuo/basic-yolo-keras.git\n",
    "cd basic-yolo-keras\n",
    "\n",
    "pip install numpy h5py pillow\n",
    "pip install tensorflow-gpu  # CPU-only: conda install -c conda-forge tensorflow\n",
    "pip install keras # Possibly older release: conda install keras\n",
    "...\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 資料集說明\n",
    "\n",
    "VIVA手部檢測資料集是由駕駛員和乘客雙手位置的二維邊界框(Bounding Box)組成，包括從手部動作和常見遮擋等自然駕駛的情境裡透過54個視頻採集而來。 這個資料集包括了7個可能的視角的手部圖像(也包括第一人稱視角)。 有些數據已經被我們的測試平台捕獲，有些則由YouTube提供。\n",
    "\n",
    "資料集的網站: http://cvrr.ucsd.edu/vivachallenge/index.php/hands/hand-detection/"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 專案的檔案路徑佈局\n",
    "\n",
    "1. 從[YOLO官方網站](http://pjreddie.com/darknet/yolo/)下載Darknet模型的設置檔與權重檔到專案根目錄。例如:使用MS COCO資料集訓練的預訓練模型\n",
    "    * 下載[YOLOv2 608x608 設置檔(yolo.cfg)](https://github.com/pjreddie/darknet/blob/master/cfg/yolo.cfg)\n",
    "    * 下載[YOLOv2 608x608 權重檔(yolo.weights)](https://pjreddie.com/media/files/yolo.weights)\n",
    "2. 在`basic-yolo-keras`的目錄裡產生一個子目錄`data`與`data/hands`\n",
    "3. 從[IVA手部檢測資料集官網]點撃[Download the dataset! (6.3 GB)](http://cvrr.ucsd.edu/vivachallenge/data/LISA_HD_Static.zip)來下 載`LISA_HD_Static.zip`。\n",
    "4. 解壓縮`LISA_HD_Static.zip`並複製其中的`detectiondata\\train`與`detectiondata\\test`兩個目錄到`basic-yolo-keras/data/hands`的目錄下\n",
    "\n",
    "最後你的目錄結構看起來像這樣: (這裡只列出來在這個範例會用到的相關檔案與目錄)\n",
    "```\n",
    "basic-yolo-keras/\n",
    "├── xxxx.ipynb\n",
    "├── yolo.weights\n",
    "├── backend.py\n",
    "├── preprocessing.py\n",
    "├── utils.py\n",
    "├── font/\n",
    "│   └── FiraMono-Medium.otf\n",
    "├── yolo.weights\n",
    "└── data/\n",
    "    └── hands/\n",
    "        ├── train/\n",
    "        │   ├── pos/         <--- 訓練圖像的位置\n",
    "        │   └── posGt/       <--- 標註檔的位置\n",
    "        └── test/\n",
    "            └── pos/         <--- 驗證圖像的位置\n",
    "    \n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### STEP 1. 載入相關函式庫"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:33:44.138409Z",
     "start_time": "2017-11-26T12:33:41.531465Z"
    },
    "code_folding": [],
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend.\n"
     ]
    }
   ],
   "source": [
    "# Utilities相關函式庫\n",
    "import os\n",
    "import random\n",
    "from tqdm import tqdm\n",
    "\n",
    "# 多維向量處理相關函式庫\n",
    "import numpy as np\n",
    "\n",
    "# 讓Keras只使用GPU來進行訓練\n",
    "os.environ[\"CUDA_DEVICE_ORDER\"] = \"PCI_BUS_ID\"\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"] = \"\"\n",
    "\n",
    "# 圖像處理相關函式庫\n",
    "import cv2\n",
    "import imgaug as ia\n",
    "from imgaug import augmenters as iaa\n",
    "import colorsys\n",
    "import matplotlib.pyplot as plt\n",
    "from PIL import Image, ImageDraw, ImageFont\n",
    "%matplotlib inline\n",
    "\n",
    "# 序列/反序列化相關函式庫\n",
    "import pickle\n",
    "\n",
    "# 深度學習相關函式庫\n",
    "from keras.models import Sequential, Model\n",
    "from keras.layers import Reshape, Activation, Conv2D, Input, MaxPooling2D, BatchNormalization, Flatten, Dense, Lambda\n",
    "from keras.layers.advanced_activations import LeakyReLU\n",
    "from keras.callbacks import EarlyStopping, ModelCheckpoint, TensorBoard\n",
    "from keras.optimizers import SGD, Adam, RMSprop\n",
    "from keras.layers.merge import concatenate\n",
    "import keras.backend as K\n",
    "import tensorflow as tf\n",
    "\n",
    "# 專案相關函式庫\n",
    "from preprocessing import parse_annotation, BatchGenerator\n",
    "from utils import WeightReader, decode_netout, draw_boxes, normalize\n",
    "from utils import draw_bgr_image_boxes, draw_rgb_image_boxes, draw_pil_image_boxes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 設定相關設定與參數"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 專案的根目錄路徑\n",
    "ROOT_DIR = os.getcwd()\n",
    "\n",
    "# 訓練/驗證用的資料目錄\n",
    "DATA_PATH = os.path.join(ROOT_DIR, \"data\")\n",
    "\n",
    "# 資料集目錄\n",
    "DATA_SET_PATH = os.path.join(DATA_PATH, \"hands\")\n",
    "TRAIN_DATA_PATH = os.path.join(DATA_SET_PATH, \"train\")\n",
    "\n",
    "TRAIN_IMGS_PATH = os.path.join(TRAIN_DATA_PATH, \"pos\")\n",
    "TRAIN_ANNOTATION_PATH = os.path.join(TRAIN_DATA_PATH, \"posGt\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 定義圖像的類別\n",
    "\n",
    "在這個圖像資料集裡有4種類別:\n",
    "1. `leftHand_driver`\n",
    "2. `rightHand_driver`\n",
    "3. `leftHand_passenger`\n",
    "4. `rightHand_passenger`\n",
    "\n",
    "但在這次的訓練中, 我們只要判斷是左/右手就行了:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['left_hand', 'right_hand']\n"
     ]
    }
   ],
   "source": [
    "# 圖像類別的Label-encoding\n",
    "map_categories = {0: 'left_hand', 1: 'right_hand'}\n",
    "\n",
    "# 取得所有圖像的圖像類別列表\n",
    "labels=list(map_categories.values())\n",
    "\n",
    "print(labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 設定YOLOv2模型的設定與參數"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:33:52.507849Z",
     "start_time": "2017-11-26T12:33:52.487930Z"
    },
    "collapsed": true,
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "LABELS = labels # 圖像類別\n",
    "\n",
    "IMAGE_H, IMAGE_W = 416, 416 # 模型輸入的圖像長寬\n",
    "GRID_H,  GRID_W  = 13 , 13\n",
    "BOX              = 5\n",
    "CLASS            = len(LABELS)\n",
    "CLASS_WEIGHTS    = np.ones(CLASS, dtype='float32')\n",
    "OBJ_THRESHOLD    = 0.5\n",
    "NMS_THRESHOLD    = 0.45 # NMS非極大值抑制 , 說明(https://chenzomi12.github.io/2016/12/14/YOLO-nms/)\n",
    "ANCHORS          = [0.57273, 0.677385, 1.87446, 2.06253, 3.33843, 5.47434, 7.88282, 3.52778, 9.77052, 9.16828]\n",
    "\n",
    "NO_OBJECT_SCALE  = 1.0\n",
    "OBJECT_SCALE     = 5.0\n",
    "COORD_SCALE      = 1.0\n",
    "CLASS_SCALE      = 1.0\n",
    "\n",
    "BATCH_SIZE       = 16\n",
    "WARM_UP_BATCHES  = 0\n",
    "TRUE_BOX_BUFFER  = 50"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Darknet預訓練權重檔與訓練/驗證資料目錄"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:33:56.177021Z",
     "start_time": "2017-11-26T12:33:56.172746Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "wt_path = 'yolo.weights' \n",
    "\n",
    "train_image_folder = TRAIN_IMGS_PATH\n",
    "train_annot_folder = TRAIN_ANNOTATION_PATH\n",
    "valid_image_folder = TRAIN_IMGS_PATH\n",
    "valid_annot_folder = TRAIN_ANNOTATION_PATH"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### STEP 2. 構建YOLOv2網絡結構模型"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:33:58.179391Z",
     "start_time": "2017-11-26T12:33:58.175696Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# the function to implement the orgnization layer (thanks to github.com/allanzelener/YAD2K)\n",
    "def space_to_depth_x2(x):\n",
    "    return tf.space_to_depth(x, block_size=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:34:01.523076Z",
     "start_time": "2017-11-26T12:34:00.007828Z"
    },
    "code_folding": [],
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "input_image = Input(shape=(IMAGE_H, IMAGE_W, 3))\n",
    "true_boxes  = Input(shape=(1, 1, 1, TRUE_BOX_BUFFER , 4))\n",
    "\n",
    "# Layer 1\n",
    "x = Conv2D(32, (3,3), strides=(1,1), padding='same', name='conv_1', use_bias=False)(input_image)\n",
    "x = BatchNormalization(name='norm_1')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "\n",
    "# Layer 2\n",
    "x = Conv2D(64, (3,3), strides=(1,1), padding='same', name='conv_2', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_2')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "\n",
    "# Layer 3\n",
    "x = Conv2D(128, (3,3), strides=(1,1), padding='same', name='conv_3', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_3')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 4\n",
    "x = Conv2D(64, (1,1), strides=(1,1), padding='same', name='conv_4', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_4')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 5\n",
    "x = Conv2D(128, (3,3), strides=(1,1), padding='same', name='conv_5', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_5')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "\n",
    "# Layer 6\n",
    "x = Conv2D(256, (3,3), strides=(1,1), padding='same', name='conv_6', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_6')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 7\n",
    "x = Conv2D(128, (1,1), strides=(1,1), padding='same', name='conv_7', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_7')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 8\n",
    "x = Conv2D(256, (3,3), strides=(1,1), padding='same', name='conv_8', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_8')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "\n",
    "# Layer 9\n",
    "x = Conv2D(512, (3,3), strides=(1,1), padding='same', name='conv_9', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_9')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 10\n",
    "x = Conv2D(256, (1,1), strides=(1,1), padding='same', name='conv_10', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_10')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 11\n",
    "x = Conv2D(512, (3,3), strides=(1,1), padding='same', name='conv_11', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_11')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 12\n",
    "x = Conv2D(256, (1,1), strides=(1,1), padding='same', name='conv_12', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_12')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 13\n",
    "x = Conv2D(512, (3,3), strides=(1,1), padding='same', name='conv_13', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_13')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "skip_connection = x\n",
    "\n",
    "x = MaxPooling2D(pool_size=(2, 2))(x)\n",
    "\n",
    "# Layer 14\n",
    "x = Conv2D(1024, (3,3), strides=(1,1), padding='same', name='conv_14', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_14')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 15\n",
    "x = Conv2D(512, (1,1), strides=(1,1), padding='same', name='conv_15', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_15')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 16\n",
    "x = Conv2D(1024, (3,3), strides=(1,1), padding='same', name='conv_16', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_16')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 17\n",
    "x = Conv2D(512, (1,1), strides=(1,1), padding='same', name='conv_17', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_17')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 18\n",
    "x = Conv2D(1024, (3,3), strides=(1,1), padding='same', name='conv_18', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_18')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 19\n",
    "x = Conv2D(1024, (3,3), strides=(1,1), padding='same', name='conv_19', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_19')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 20\n",
    "x = Conv2D(1024, (3,3), strides=(1,1), padding='same', name='conv_20', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_20')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 21\n",
    "skip_connection = Conv2D(64, (1,1), strides=(1,1), padding='same', name='conv_21', use_bias=False)(skip_connection)\n",
    "skip_connection = BatchNormalization(name='norm_21')(skip_connection)\n",
    "skip_connection = LeakyReLU(alpha=0.1)(skip_connection)\n",
    "skip_connection = Lambda(space_to_depth_x2)(skip_connection)\n",
    "\n",
    "x = concatenate([skip_connection, x])\n",
    "\n",
    "# Layer 22\n",
    "x = Conv2D(1024, (3,3), strides=(1,1), padding='same', name='conv_22', use_bias=False)(x)\n",
    "x = BatchNormalization(name='norm_22')(x)\n",
    "x = LeakyReLU(alpha=0.1)(x)\n",
    "\n",
    "# Layer 23\n",
    "x = Conv2D(BOX * (4 + 1 + CLASS), (1,1), strides=(1,1), padding='same', name='conv_23')(x)\n",
    "output = Reshape((GRID_H, GRID_W, BOX, 4 + 1 + CLASS))(x)\n",
    "\n",
    "# small hack to allow true_boxes to be registered when Keras build the model \n",
    "# for more information: https://github.com/fchollet/keras/issues/2790\n",
    "output = Lambda(lambda args: args[0])([output, true_boxes])\n",
    "\n",
    "model = Model([input_image, true_boxes], output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:34:03.819802Z",
     "start_time": "2017-11-26T12:34:03.786125Z"
    },
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "Layer (type)                    Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      "input_1 (InputLayer)            (None, 416, 416, 3)  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "conv_1 (Conv2D)                 (None, 416, 416, 32) 864         input_1[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "norm_1 (BatchNormalization)     (None, 416, 416, 32) 128         conv_1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_1 (LeakyReLU)       (None, 416, 416, 32) 0           norm_1[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_1 (MaxPooling2D)  (None, 208, 208, 32) 0           leaky_re_lu_1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv_2 (Conv2D)                 (None, 208, 208, 64) 18432       max_pooling2d_1[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "norm_2 (BatchNormalization)     (None, 208, 208, 64) 256         conv_2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_2 (LeakyReLU)       (None, 208, 208, 64) 0           norm_2[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_2 (MaxPooling2D)  (None, 104, 104, 64) 0           leaky_re_lu_2[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv_3 (Conv2D)                 (None, 104, 104, 128 73728       max_pooling2d_2[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "norm_3 (BatchNormalization)     (None, 104, 104, 128 512         conv_3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_3 (LeakyReLU)       (None, 104, 104, 128 0           norm_3[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv_4 (Conv2D)                 (None, 104, 104, 64) 8192        leaky_re_lu_3[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "norm_4 (BatchNormalization)     (None, 104, 104, 64) 256         conv_4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_4 (LeakyReLU)       (None, 104, 104, 64) 0           norm_4[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv_5 (Conv2D)                 (None, 104, 104, 128 73728       leaky_re_lu_4[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "norm_5 (BatchNormalization)     (None, 104, 104, 128 512         conv_5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_5 (LeakyReLU)       (None, 104, 104, 128 0           norm_5[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_3 (MaxPooling2D)  (None, 52, 52, 128)  0           leaky_re_lu_5[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv_6 (Conv2D)                 (None, 52, 52, 256)  294912      max_pooling2d_3[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "norm_6 (BatchNormalization)     (None, 52, 52, 256)  1024        conv_6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_6 (LeakyReLU)       (None, 52, 52, 256)  0           norm_6[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv_7 (Conv2D)                 (None, 52, 52, 128)  32768       leaky_re_lu_6[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "norm_7 (BatchNormalization)     (None, 52, 52, 128)  512         conv_7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_7 (LeakyReLU)       (None, 52, 52, 128)  0           norm_7[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv_8 (Conv2D)                 (None, 52, 52, 256)  294912      leaky_re_lu_7[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "norm_8 (BatchNormalization)     (None, 52, 52, 256)  1024        conv_8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_8 (LeakyReLU)       (None, 52, 52, 256)  0           norm_8[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_4 (MaxPooling2D)  (None, 26, 26, 256)  0           leaky_re_lu_8[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "conv_9 (Conv2D)                 (None, 26, 26, 512)  1179648     max_pooling2d_4[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "norm_9 (BatchNormalization)     (None, 26, 26, 512)  2048        conv_9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_9 (LeakyReLU)       (None, 26, 26, 512)  0           norm_9[0][0]                     \n",
      "__________________________________________________________________________________________________\n",
      "conv_10 (Conv2D)                (None, 26, 26, 256)  131072      leaky_re_lu_9[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "norm_10 (BatchNormalization)    (None, 26, 26, 256)  1024        conv_10[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_10 (LeakyReLU)      (None, 26, 26, 256)  0           norm_10[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_11 (Conv2D)                (None, 26, 26, 512)  1179648     leaky_re_lu_10[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_11 (BatchNormalization)    (None, 26, 26, 512)  2048        conv_11[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_11 (LeakyReLU)      (None, 26, 26, 512)  0           norm_11[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_12 (Conv2D)                (None, 26, 26, 256)  131072      leaky_re_lu_11[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_12 (BatchNormalization)    (None, 26, 26, 256)  1024        conv_12[0][0]                    \n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_12 (LeakyReLU)      (None, 26, 26, 256)  0           norm_12[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_13 (Conv2D)                (None, 26, 26, 512)  1179648     leaky_re_lu_12[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_13 (BatchNormalization)    (None, 26, 26, 512)  2048        conv_13[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_13 (LeakyReLU)      (None, 26, 26, 512)  0           norm_13[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "max_pooling2d_5 (MaxPooling2D)  (None, 13, 13, 512)  0           leaky_re_lu_13[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv_14 (Conv2D)                (None, 13, 13, 1024) 4718592     max_pooling2d_5[0][0]            \n",
      "__________________________________________________________________________________________________\n",
      "norm_14 (BatchNormalization)    (None, 13, 13, 1024) 4096        conv_14[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_14 (LeakyReLU)      (None, 13, 13, 1024) 0           norm_14[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_15 (Conv2D)                (None, 13, 13, 512)  524288      leaky_re_lu_14[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_15 (BatchNormalization)    (None, 13, 13, 512)  2048        conv_15[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_15 (LeakyReLU)      (None, 13, 13, 512)  0           norm_15[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_16 (Conv2D)                (None, 13, 13, 1024) 4718592     leaky_re_lu_15[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_16 (BatchNormalization)    (None, 13, 13, 1024) 4096        conv_16[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_16 (LeakyReLU)      (None, 13, 13, 1024) 0           norm_16[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_17 (Conv2D)                (None, 13, 13, 512)  524288      leaky_re_lu_16[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_17 (BatchNormalization)    (None, 13, 13, 512)  2048        conv_17[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_17 (LeakyReLU)      (None, 13, 13, 512)  0           norm_17[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_18 (Conv2D)                (None, 13, 13, 1024) 4718592     leaky_re_lu_17[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_18 (BatchNormalization)    (None, 13, 13, 1024) 4096        conv_18[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_18 (LeakyReLU)      (None, 13, 13, 1024) 0           norm_18[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_19 (Conv2D)                (None, 13, 13, 1024) 9437184     leaky_re_lu_18[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "norm_19 (BatchNormalization)    (None, 13, 13, 1024) 4096        conv_19[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_21 (Conv2D)                (None, 26, 26, 64)   32768       leaky_re_lu_13[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_19 (LeakyReLU)      (None, 13, 13, 1024) 0           norm_19[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "norm_21 (BatchNormalization)    (None, 26, 26, 64)   256         conv_21[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_20 (Conv2D)                (None, 13, 13, 1024) 9437184     leaky_re_lu_19[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_21 (LeakyReLU)      (None, 26, 26, 64)   0           norm_21[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "norm_20 (BatchNormalization)    (None, 13, 13, 1024) 4096        conv_20[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "lambda_1 (Lambda)               (None, 13, 13, 256)  0           leaky_re_lu_21[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_20 (LeakyReLU)      (None, 13, 13, 1024) 0           norm_20[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "concatenate_1 (Concatenate)     (None, 13, 13, 1280) 0           lambda_1[0][0]                   \n",
      "                                                                 leaky_re_lu_20[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "conv_22 (Conv2D)                (None, 13, 13, 1024) 11796480    concatenate_1[0][0]              \n",
      "__________________________________________________________________________________________________\n",
      "norm_22 (BatchNormalization)    (None, 13, 13, 1024) 4096        conv_22[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "leaky_re_lu_22 (LeakyReLU)      (None, 13, 13, 1024) 0           norm_22[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "conv_23 (Conv2D)                (None, 13, 13, 35)   35875       leaky_re_lu_22[0][0]             \n",
      "__________________________________________________________________________________________________\n",
      "reshape_1 (Reshape)             (None, 13, 13, 5, 7) 0           conv_23[0][0]                    \n",
      "__________________________________________________________________________________________________\n",
      "input_2 (InputLayer)            (None, 1, 1, 1, 50,  0                                            \n",
      "__________________________________________________________________________________________________\n",
      "lambda_2 (Lambda)               (None, 13, 13, 5, 7) 0           reshape_1[0][0]                  \n",
      "                                                                 input_2[0][0]                    \n",
      "==================================================================================================\n",
      "Total params: 50,583,811\n",
      "Trainable params: 50,563,139\n",
      "Non-trainable params: 20,672\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary() # 打印模型"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### STEP 3. 載入預訓練的模型權重"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Load the weights originally provided by YOLO**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:34:06.976188Z",
     "start_time": "2017-11-26T12:34:06.232200Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "weight_reader = WeightReader(wt_path)  # 初始讀取Darknet預訓練權重檔物件"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:34:11.559043Z",
     "start_time": "2017-11-26T12:34:08.310987Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "handle norm_1 start\n",
      "shape:  (32,)\n",
      "handle norm_1 completed\n",
      "handle conv_1 start\n",
      "handle conv_1 completed\n",
      "handle norm_2 start\n",
      "shape:  (64,)\n",
      "handle norm_2 completed\n",
      "handle conv_2 start\n",
      "handle conv_2 completed\n",
      "handle norm_3 start\n",
      "shape:  (128,)\n",
      "handle norm_3 completed\n",
      "handle conv_3 start\n",
      "handle conv_3 completed\n",
      "handle norm_4 start\n",
      "shape:  (64,)\n",
      "handle norm_4 completed\n",
      "handle conv_4 start\n",
      "handle conv_4 completed\n",
      "handle norm_5 start\n",
      "shape:  (128,)\n",
      "handle norm_5 completed\n",
      "handle conv_5 start\n",
      "handle conv_5 completed\n",
      "handle norm_6 start\n",
      "shape:  (256,)\n",
      "handle norm_6 completed\n",
      "handle conv_6 start\n",
      "handle conv_6 completed\n",
      "handle norm_7 start\n",
      "shape:  (128,)\n",
      "handle norm_7 completed\n",
      "handle conv_7 start\n",
      "handle conv_7 completed\n",
      "handle norm_8 start\n",
      "shape:  (256,)\n",
      "handle norm_8 completed\n",
      "handle conv_8 start\n",
      "handle conv_8 completed\n",
      "handle norm_9 start\n",
      "shape:  (512,)\n",
      "handle norm_9 completed\n",
      "handle conv_9 start\n",
      "handle conv_9 completed\n",
      "handle norm_10 start\n",
      "shape:  (256,)\n",
      "handle norm_10 completed\n",
      "handle conv_10 start\n",
      "handle conv_10 completed\n",
      "handle norm_11 start\n",
      "shape:  (512,)\n",
      "handle norm_11 completed\n",
      "handle conv_11 start\n",
      "handle conv_11 completed\n",
      "handle norm_12 start\n",
      "shape:  (256,)\n",
      "handle norm_12 completed\n",
      "handle conv_12 start\n",
      "handle conv_12 completed\n",
      "handle norm_13 start\n",
      "shape:  (512,)\n",
      "handle norm_13 completed\n",
      "handle conv_13 start\n",
      "handle conv_13 completed\n",
      "handle norm_14 start\n",
      "shape:  (1024,)\n",
      "handle norm_14 completed\n",
      "handle conv_14 start\n",
      "handle conv_14 completed\n",
      "handle norm_15 start\n",
      "shape:  (512,)\n",
      "handle norm_15 completed\n",
      "handle conv_15 start\n",
      "handle conv_15 completed\n",
      "handle norm_16 start\n",
      "shape:  (1024,)\n",
      "handle norm_16 completed\n",
      "handle conv_16 start\n",
      "handle conv_16 completed\n",
      "handle norm_17 start\n",
      "shape:  (512,)\n",
      "handle norm_17 completed\n",
      "handle conv_17 start\n",
      "handle conv_17 completed\n",
      "handle norm_18 start\n",
      "shape:  (1024,)\n",
      "handle norm_18 completed\n",
      "handle conv_18 start\n",
      "handle conv_18 completed\n",
      "handle norm_19 start\n",
      "shape:  (1024,)\n",
      "handle norm_19 completed\n",
      "handle conv_19 start\n",
      "handle conv_19 completed\n",
      "handle norm_20 start\n",
      "shape:  (1024,)\n",
      "handle norm_20 completed\n",
      "handle conv_20 start\n",
      "handle conv_20 completed\n",
      "handle norm_21 start\n",
      "shape:  (64,)\n",
      "handle norm_21 completed\n",
      "handle conv_21 start\n",
      "handle conv_21 completed\n",
      "handle norm_22 start\n",
      "shape:  (1024,)\n",
      "handle norm_22 completed\n",
      "handle conv_22 start\n",
      "handle conv_22 completed\n",
      "handle conv_23 start\n",
      "len: 2\n",
      "handle conv_23 completed\n"
     ]
    }
   ],
   "source": [
    "weight_reader.reset()\n",
    "nb_conv = 23 # 總共有23層的卷積層\n",
    "\n",
    "for i in range(1, nb_conv+1):\n",
    "    conv_layer = model.get_layer('conv_' + str(i))\n",
    "    \n",
    "    # 在conv_1~conv_22的卷積組合裡都包含了\"conv + norm\"二層, 只有conv_23是獨立一層    \n",
    "    if i < nb_conv: \n",
    "        print(\"handle norm_\" + str(i) + \" start\")        \n",
    "        norm_layer = model.get_layer('norm_' + str(i)) # 取得BatchNormalization層\n",
    "        \n",
    "        size = np.prod(norm_layer.get_weights()[0].shape) # 取得BatchNormalization層的參數量\n",
    "        print(\"shape: \", norm_layer.get_weights()[0].shape)\n",
    "        \n",
    "        beta  = weight_reader.read_bytes(size)\n",
    "        gamma = weight_reader.read_bytes(size)\n",
    "        mean  = weight_reader.read_bytes(size)\n",
    "        var   = weight_reader.read_bytes(size)\n",
    "        weights = norm_layer.set_weights([gamma, beta, mean, var])\n",
    "        print(\"handle norm_\" + str(i) + \" completed\")\n",
    "        \n",
    "    if len(conv_layer.get_weights()) > 1:\n",
    "        print(\"handle conv_\" + str(i) + \" start\")  \n",
    "        print(\"len:\",len(conv_layer.get_weights()))\n",
    "        bias   = weight_reader.read_bytes(np.prod(conv_layer.get_weights()[1].shape))\n",
    "        kernel = weight_reader.read_bytes(np.prod(conv_layer.get_weights()[0].shape))\n",
    "        kernel = kernel.reshape(list(reversed(conv_layer.get_weights()[0].shape)))\n",
    "        kernel = kernel.transpose([2,3,1,0])\n",
    "        conv_layer.set_weights([kernel, bias])\n",
    "        print(\"handle conv_\" + str(i) + \" completed\")\n",
    "    else:\n",
    "        print(\"handle conv_\" + str(i) + \" start\")        \n",
    "        kernel = weight_reader.read_bytes(np.prod(conv_layer.get_weights()[0].shape))\n",
    "        kernel = kernel.reshape(list(reversed(conv_layer.get_weights()[0].shape)))\n",
    "        kernel = kernel.transpose([2,3,1,0])\n",
    "        conv_layer.set_weights([kernel])\n",
    "        print(\"handle conv_\" + str(i) + \" completed\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### STEP 4. 設定要微調(fine-tune)的模型層級權重\n",
    "\n",
    "**Randomize weights of the last layer**\n",
    "\n",
    "由於在YOLOv2的模型中, 最後一層卷積層決定了最後的輸出, 讓我們重新來調整與訓練這一層的卷積層來讓預訓練的模型可以讓我們進行所需要的微調。\n",
    "詳細的概念與說明,見 [1.5: 使用預先訓練的卷積網絡模型](https://github.com/erhwenkuo/deep-learning-with-keras-notebooks/blob/master/1.5-use-pretrained-model-2.ipynb)。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-22T14:08:00.245248Z",
     "start_time": "2017-11-22T14:08:00.215495Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "layer   = model.layers[-4] # 找出最後一層的卷積層\n",
    "weights = layer.get_weights()\n",
    "\n",
    "new_kernel = np.random.normal(size=weights[0].shape)/(GRID_H*GRID_W)\n",
    "new_bias   = np.random.normal(size=weights[1].shape)/(GRID_H*GRID_W)\n",
    "\n",
    "layer.set_weights([new_kernel, new_bias]) # 重初始化權重"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### STEP 5. 模型訓練"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**YOLOv2訓練用的損失函數:**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-02-01T20:44:50.211553",
     "start_time": "2017-02-01T20:44:50.206006"
    }
   },
   "source": [
    "$$\\begin{multline}\n",
    "\\lambda_\\textbf{coord}\n",
    "\\sum_{i = 0}^{S^2}\n",
    "    \\sum_{j = 0}^{B}\n",
    "     L_{ij}^{\\text{obj}}\n",
    "            \\left[\n",
    "            \\left(\n",
    "                x_i - \\hat{x}_i\n",
    "            \\right)^2 +\n",
    "            \\left(\n",
    "                y_i - \\hat{y}_i\n",
    "            \\right)^2\n",
    "            \\right]\n",
    "\\\\\n",
    "+ \\lambda_\\textbf{coord} \n",
    "\\sum_{i = 0}^{S^2}\n",
    "    \\sum_{j = 0}^{B}\n",
    "         L_{ij}^{\\text{obj}}\n",
    "         \\left[\n",
    "        \\left(\n",
    "            \\sqrt{w_i} - \\sqrt{\\hat{w}_i}\n",
    "        \\right)^2 +\n",
    "        \\left(\n",
    "            \\sqrt{h_i} - \\sqrt{\\hat{h}_i}\n",
    "        \\right)^2\n",
    "        \\right]\n",
    "\\\\\n",
    "+ \\sum_{i = 0}^{S^2}\n",
    "    \\sum_{j = 0}^{B}\n",
    "        L_{ij}^{\\text{obj}}\n",
    "        \\left(\n",
    "            C_i - \\hat{C}_i\n",
    "        \\right)^2\n",
    "\\\\\n",
    "+ \\lambda_\\textrm{noobj}\n",
    "\\sum_{i = 0}^{S^2}\n",
    "    \\sum_{j = 0}^{B}\n",
    "    L_{ij}^{\\text{noobj}}\n",
    "        \\left(\n",
    "            C_i - \\hat{C}_i\n",
    "        \\right)^2\n",
    "\\\\\n",
    "+ \\sum_{i = 0}^{S^2}\n",
    "L_i^{\\text{obj}}\n",
    "    \\sum_{c \\in \\textrm{classes}}\n",
    "        \\left(\n",
    "            p_i(c) - \\hat{p}_i(c)\n",
    "        \\right)^2\n",
    "\\end{multline}$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:34:28.064549Z",
     "start_time": "2017-11-26T12:34:27.800510Z"
    },
    "code_folding": [],
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def custom_loss(y_true, y_pred):\n",
    "    mask_shape = tf.shape(y_true)[:4]\n",
    "    \n",
    "    cell_x = tf.to_float(tf.reshape(tf.tile(tf.range(GRID_W), [GRID_H]), (1, GRID_H, GRID_W, 1, 1)))\n",
    "    cell_y = tf.transpose(cell_x, (0,2,1,3,4))\n",
    "\n",
    "    cell_grid = tf.tile(tf.concat([cell_x,cell_y], -1), [BATCH_SIZE, 1, 1, 5, 1])\n",
    "    \n",
    "    coord_mask = tf.zeros(mask_shape)\n",
    "    conf_mask  = tf.zeros(mask_shape)\n",
    "    class_mask = tf.zeros(mask_shape)\n",
    "    \n",
    "    seen = tf.Variable(0.)\n",
    "    total_recall = tf.Variable(0.)\n",
    "    \n",
    "    \"\"\"\n",
    "    Adjust prediction\n",
    "    \"\"\"\n",
    "    ### adjust x and y      \n",
    "    pred_box_xy = tf.sigmoid(y_pred[..., :2]) + cell_grid\n",
    "    \n",
    "    ### adjust w and h\n",
    "    pred_box_wh = tf.exp(y_pred[..., 2:4]) * np.reshape(ANCHORS, [1,1,1,BOX,2])\n",
    "    \n",
    "    ### adjust confidence\n",
    "    pred_box_conf = tf.sigmoid(y_pred[..., 4])\n",
    "    \n",
    "    ### adjust class probabilities\n",
    "    pred_box_class = y_pred[..., 5:]\n",
    "    \n",
    "    \"\"\"\n",
    "    Adjust ground truth\n",
    "    \"\"\"\n",
    "    ### adjust x and y\n",
    "    true_box_xy = y_true[..., 0:2] # relative position to the containing cell\n",
    "    \n",
    "    ### adjust w and h\n",
    "    true_box_wh = y_true[..., 2:4] # number of cells accross, horizontally and vertically\n",
    "    \n",
    "    ### adjust confidence\n",
    "    true_wh_half = true_box_wh / 2.\n",
    "    true_mins    = true_box_xy - true_wh_half\n",
    "    true_maxes   = true_box_xy + true_wh_half\n",
    "    \n",
    "    pred_wh_half = pred_box_wh / 2.\n",
    "    pred_mins    = pred_box_xy - pred_wh_half\n",
    "    pred_maxes   = pred_box_xy + pred_wh_half       \n",
    "    \n",
    "    intersect_mins  = tf.maximum(pred_mins,  true_mins)\n",
    "    intersect_maxes = tf.minimum(pred_maxes, true_maxes)\n",
    "    intersect_wh    = tf.maximum(intersect_maxes - intersect_mins, 0.)\n",
    "    intersect_areas = intersect_wh[..., 0] * intersect_wh[..., 1]\n",
    "    \n",
    "    true_areas = true_box_wh[..., 0] * true_box_wh[..., 1]\n",
    "    pred_areas = pred_box_wh[..., 0] * pred_box_wh[..., 1]\n",
    "\n",
    "    union_areas = pred_areas + true_areas - intersect_areas\n",
    "    iou_scores  = tf.truediv(intersect_areas, union_areas)\n",
    "    \n",
    "    true_box_conf = iou_scores * y_true[..., 4]\n",
    "    \n",
    "    ### adjust class probabilities\n",
    "    true_box_class = tf.argmax(y_true[..., 5:], -1)\n",
    "    \n",
    "    \"\"\"\n",
    "    Determine the masks\n",
    "    \"\"\"\n",
    "    ### coordinate mask: simply the position of the ground truth boxes (the predictors)\n",
    "    coord_mask = tf.expand_dims(y_true[..., 4], axis=-1) * COORD_SCALE\n",
    "    \n",
    "    ### confidence mask: penelize predictors + penalize boxes with low IOU\n",
    "    # penalize the confidence of the boxes, which have IOU with some ground truth box < 0.6\n",
    "    true_xy = true_boxes[..., 0:2]\n",
    "    true_wh = true_boxes[..., 2:4]\n",
    "    \n",
    "    true_wh_half = true_wh / 2.\n",
    "    true_mins    = true_xy - true_wh_half\n",
    "    true_maxes   = true_xy + true_wh_half\n",
    "    \n",
    "    pred_xy = tf.expand_dims(pred_box_xy, 4)\n",
    "    pred_wh = tf.expand_dims(pred_box_wh, 4)\n",
    "    \n",
    "    pred_wh_half = pred_wh / 2.\n",
    "    pred_mins    = pred_xy - pred_wh_half\n",
    "    pred_maxes   = pred_xy + pred_wh_half    \n",
    "    \n",
    "    intersect_mins  = tf.maximum(pred_mins,  true_mins)\n",
    "    intersect_maxes = tf.minimum(pred_maxes, true_maxes)\n",
    "    intersect_wh    = tf.maximum(intersect_maxes - intersect_mins, 0.)\n",
    "    intersect_areas = intersect_wh[..., 0] * intersect_wh[..., 1]\n",
    "    \n",
    "    true_areas = true_wh[..., 0] * true_wh[..., 1]\n",
    "    pred_areas = pred_wh[..., 0] * pred_wh[..., 1]\n",
    "\n",
    "    union_areas = pred_areas + true_areas - intersect_areas\n",
    "    iou_scores  = tf.truediv(intersect_areas, union_areas)\n",
    "\n",
    "    best_ious = tf.reduce_max(iou_scores, axis=4)\n",
    "    conf_mask = conf_mask + tf.to_float(best_ious < 0.6) * (1 - y_true[..., 4]) * NO_OBJECT_SCALE\n",
    "    \n",
    "    # penalize the confidence of the boxes, which are reponsible for corresponding ground truth box\n",
    "    conf_mask = conf_mask + y_true[..., 4] * OBJECT_SCALE\n",
    "    \n",
    "    ### class mask: simply the position of the ground truth boxes (the predictors)\n",
    "    class_mask = y_true[..., 4] * tf.gather(CLASS_WEIGHTS, true_box_class) * CLASS_SCALE       \n",
    "    \n",
    "    \"\"\"\n",
    "    Warm-up training\n",
    "    \"\"\"\n",
    "    no_boxes_mask = tf.to_float(coord_mask < COORD_SCALE/2.)\n",
    "    seen = tf.assign_add(seen, 1.)\n",
    "    \n",
    "    true_box_xy, true_box_wh, coord_mask = tf.cond(tf.less(seen, WARM_UP_BATCHES), \n",
    "                          lambda: [true_box_xy + (0.5 + cell_grid) * no_boxes_mask, \n",
    "                                   true_box_wh + tf.ones_like(true_box_wh) * np.reshape(ANCHORS, [1,1,1,BOX,2]) * no_boxes_mask, \n",
    "                                   tf.ones_like(coord_mask)],\n",
    "                          lambda: [true_box_xy, \n",
    "                                   true_box_wh,\n",
    "                                   coord_mask])\n",
    "    \n",
    "    \"\"\"\n",
    "    Finalize the loss\n",
    "    \"\"\"\n",
    "    nb_coord_box = tf.reduce_sum(tf.to_float(coord_mask > 0.0))\n",
    "    nb_conf_box  = tf.reduce_sum(tf.to_float(conf_mask  > 0.0))\n",
    "    nb_class_box = tf.reduce_sum(tf.to_float(class_mask > 0.0))\n",
    "    \n",
    "    loss_xy    = tf.reduce_sum(tf.square(true_box_xy-pred_box_xy)     * coord_mask) / (nb_coord_box + 1e-6) / 2.\n",
    "    loss_wh    = tf.reduce_sum(tf.square(true_box_wh-pred_box_wh)     * coord_mask) / (nb_coord_box + 1e-6) / 2.\n",
    "    loss_conf  = tf.reduce_sum(tf.square(true_box_conf-pred_box_conf) * conf_mask)  / (nb_conf_box  + 1e-6) / 2.\n",
    "    loss_class = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=true_box_class, logits=pred_box_class)\n",
    "    loss_class = tf.reduce_sum(loss_class * class_mask) / (nb_class_box + 1e-6)\n",
    "    \n",
    "    loss = loss_xy + loss_wh + loss_conf + loss_class\n",
    "    \n",
    "    nb_true_box = tf.reduce_sum(y_true[..., 4])\n",
    "    nb_pred_box = tf.reduce_sum(tf.to_float(true_box_conf > 0.5) * tf.to_float(pred_box_conf > 0.3))\n",
    "\n",
    "    \"\"\"\n",
    "    Debugging code\n",
    "    \"\"\"    \n",
    "    current_recall = nb_pred_box/(nb_true_box + 1e-6)\n",
    "    total_recall = tf.assign_add(total_recall, current_recall) \n",
    "\n",
    "    loss = tf.Print(loss, [tf.zeros((1))], message='Dummy Line \\t', summarize=1000)\n",
    "    loss = tf.Print(loss, [loss_xy], message='Loss XY \\t', summarize=1000)\n",
    "    loss = tf.Print(loss, [loss_wh], message='Loss WH \\t', summarize=1000)\n",
    "    loss = tf.Print(loss, [loss_conf], message='Loss Conf \\t', summarize=1000)\n",
    "    loss = tf.Print(loss, [loss_class], message='Loss Class \\t', summarize=1000)\n",
    "    loss = tf.Print(loss, [loss], message='Total Loss \\t', summarize=1000)\n",
    "    loss = tf.Print(loss, [current_recall], message='Current Recall \\t', summarize=1000)\n",
    "    loss = tf.Print(loss, [total_recall/seen], message='Average Recall \\t', summarize=1000)\n",
    "    \n",
    "    return loss"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**用來產生Keras訓練模型的BatchGenerator的設定:**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:38:44.283547Z",
     "start_time": "2017-11-26T12:38:44.277155Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "generator_config = {\n",
    "    'IMAGE_H'         : IMAGE_H, # YOLOv2網絡輸入的image_h\n",
    "    'IMAGE_W'         : IMAGE_W, # YOLOv2網絡輸入的image_w\n",
    "    'GRID_H'          : GRID_H,  # 直向網格的拆分數量\n",
    "    'GRID_W'          : GRID_W,  # 橫向網格的拆分數量\n",
    "    'BOX'             : BOX,     # 每個單一網格要預測的邊界框數量\n",
    "    'LABELS'          : LABELS,  # 要預測的圖像種類列表\n",
    "    'CLASS'           : len(LABELS), # 要預測的圖像種類數\n",
    "    'ANCHORS'         : ANCHORS, # 每個單一網格要預測的邊界框時用的錨點\n",
    "    'BATCH_SIZE'      : BATCH_SIZE, # 訓練時的批量數\n",
    "    'TRUE_BOX_BUFFER' : 50, # 一個訓練圖像最大數量的邊界框數\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 解析圖像標註檔\n",
    "由於這個資料集的標註檔並不是採用PASCAL VOC格式而是自行定義的格式：\n",
    "```\n",
    "% bbGt version=3\n",
    "leftHand_driver 87 295 57 67 0 0 0 0 0 0 0\n",
    "rightHand_driver 223 283 62 64 0 0 0 0 0 0 0\n",
    "leftHand_passenger 483 356 91 71 0 0 0 0 0 0 0\n",
    "rightHand_passenger 548 328 86 70 0 0 0 0 0 0 0\n",
    "```\n",
    "因此要對標註檔的解析進行客製化。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from tqdm import tqdm\n",
    "from PIL import Image\n",
    "def parse_hands_annotation(ann_dir, img_dir, labels=[]):\n",
    "    \"\"\"解析圖像標註檔\n",
    "\n",
    "    根據手部標註檔存放的目錄路徑迭代地解析每一個標註檔，\n",
    "    將每個圖像的檔名(filename)、圖像的寬(width)、高(height)、圖像的類別(name)以\n",
    "    及物體的邊界框的坐標(xmin,ymin,xmax,ymax)擷取出來。以下是圖像標註檔的範例:\n",
    "    \n",
    "    % bbGt version=3\n",
    "    leftHand_driver 87 295 57 67 0 0 0 0 0 0 0\n",
    "    rightHand_driver 223 283 62 64 0 0 0 0 0 0 0\n",
    "    leftHand_passenger 483 356 91 71 0 0 0 0 0 0 0\n",
    "    rightHand_passenger 548 328 86 70 0 0 0 0 0 0 0\n",
    "    \n",
    "    擷取目標: [hands_class x y w h ...]\n",
    "    hands_class: leftHand_driver/leftHand_passenger: left_hand, rightHand_driver/rightHand_passenger: right_hand, \n",
    "\n",
    "    參數:\n",
    "        ann_dir: 圖像標註檔存放的目錄路徑\n",
    "        img_dir: 圖像檔存放的目錄路徑\n",
    "        labels: 圖像資料集的物體類別列表\n",
    "\n",
    "    回傳:\n",
    "        all_imgs: 一個列表物件, 每一個物件都包括了要訓練用的重要資訊。例如:\n",
    "                    {\n",
    "                        'filename': '/tmp/img/img001.jpg',\n",
    "                        'width': 128,\n",
    "                        'height': 128,\n",
    "                        'object':[\n",
    "                            {'name':'person',xmin:0, ymin:0, xmax:28, ymax:28},\n",
    "                            {'name':'person',xmin:45, ymin:45, xmax:60, ymax:60}\n",
    "                        ]\n",
    "                    }\n",
    "        seen_labels: 一個字典物件(k:圖像類別, v:出現的次數)用來檢視每一個圖像類別出現的次數\n",
    "    \"\"\"\n",
    "    print(\"start parsing annotation..\")\n",
    "    \n",
    "    # 產生一個標註圖資料標註的mapping\n",
    "    hands_label_map = {'leftHand_driver':'left_hand', 'leftHand_passenger':'left_hand',\n",
    "                      'rightHand_driver':'right_hand', 'rightHand_passenger':'right_hand'}\n",
    "    all_imgs = []\n",
    "    seen_labels = {}\n",
    "    \n",
    "    # 迭代每個標註檔\n",
    "    for ann in tqdm(sorted(os.listdir(ann_dir))):\n",
    "        img = {'object':[]}\n",
    "        # 處理圖檔檔案路徑\n",
    "        img_filename = ann[0:len(ann)-3]+\"png\"\n",
    "                \n",
    "        # 圖檔檔案路徑\n",
    "        img['filename'] = os.path.join(img_dir, img_filename)\n",
    "        \n",
    "        im = Image.open(img['filename'])\n",
    "        img_width, img_height = im.size\n",
    "        \n",
    "        # 圖檔大小\n",
    "        img['width'] = img_width\n",
    "        img['height'] = img_height\n",
    "        \n",
    "        line = 0 # 行數\n",
    "        with open(os.path.join(ann_dir, ann), 'r') as fann:\n",
    "            # 一行一行讀進來處理\n",
    "            for cnt, line in enumerate(fann):\n",
    "                # 忽略第一行的資料\n",
    "                if cnt == 0:\n",
    "                    continue\n",
    "                # 建立物件來保留bbox\n",
    "                obj = {}\n",
    "                \n",
    "                tokens = line.split()\n",
    "                label = hands_label_map[tokens[0]]\n",
    "                bbox_x = int(tokens[1])\n",
    "                bbox_y = int(tokens[2])\n",
    "                bbox_w = int(tokens[3])\n",
    "                bbox_h = int(tokens[4])\n",
    "                #print(\"Line {}: {},{},{},{},{}\".format(cnt, label, bbox_x, bbox_y, bbox_w, bbox_h))\n",
    "                \n",
    "                obj['name'] = label\n",
    "                \n",
    "                if obj['name'] in seen_labels:\n",
    "                    seen_labels[obj['name']] += 1\n",
    "                else:\n",
    "                    seen_labels[obj['name']] = 1\n",
    "                \n",
    "                obj['xmin'] = bbox_x\n",
    "                obj['ymin'] = bbox_y\n",
    "                obj['xmax'] = bbox_x + bbox_w\n",
    "                obj['ymax'] = bbox_y + bbox_h\n",
    "                \n",
    "                #檢看是是否有物體的標籤是沒有在傳入的物體類別(labels)中\n",
    "                if len(labels) > 0 and obj['name'] not in labels:\n",
    "                    continue\n",
    "                else:\n",
    "                    img['object'] += [obj]\n",
    "            \n",
    "            # 把img物件加進要回傳的列表中\n",
    "            if len(['object']) > 0:\n",
    "                all_imgs += [img]\n",
    "    \n",
    "    print(\"Parsing annotation completed!\")\n",
    "    print(\"Total: {} images processed.\".format(len(all_imgs)))\n",
    "    return all_imgs, seen_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:38:51.836129Z",
     "start_time": "2017-11-26T12:38:51.766843Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "start parsing annotation..\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████| 5500/5500 [00:15<00:00, 360.49it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parsing annotation completed!\n",
      "Total: 5500 images processed.\n",
      "start parsing annotation..\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|█████████████████████████████████████████████████████████████████████████████| 5500/5500 [00:15<00:00, 363.95it/s]\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Parsing annotation completed!\n",
      "Total: 5500 images processed.\n"
     ]
    }
   ],
   "source": [
    "# 進行圖像標註檔的解析 (在Racoon資料集的標註採用的是PASCAL VOC的XML格式)\n",
    "train_imgs, seen_train_labels = parse_hands_annotation(train_annot_folder, train_image_folder, labels=LABELS)\n",
    "\n",
    "# 建立一個訓練用的資料產生器\n",
    "train_batch = BatchGenerator(train_imgs, generator_config, norm=normalize)\n",
    "\n",
    "# 進行圖像標註檔的解析 (在Racoon資料集的標註採用的是PASCAL VOC的XML格式)\n",
    "valid_imgs, seen_valid_labels = parse_hands_annotation(valid_annot_folder, valid_image_folder, labels=LABELS)\n",
    "\n",
    "# 建立一個驗證用的資料產生器\n",
    "valid_batch = BatchGenerator(valid_imgs, generator_config, norm=normalize, jitter=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**設置一些回調函式並開始訓練**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-26T12:38:15.714460Z",
     "start_time": "2017-11-26T12:38:15.708674Z"
    },
    "code_folding": [],
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 如果超過3次的循環在loss的收歛上沒有改善就停止訓練 \n",
    "early_stop = EarlyStopping(monitor='val_loss', \n",
    "                           min_delta=0.001, \n",
    "                           patience=3, \n",
    "                           mode='min', \n",
    "                           verbose=1)\n",
    "\n",
    "# 每次的訓練循都去比較模型的loss是否有改善, 有就把模型的權重儲存下來\n",
    "checkpoint = ModelCheckpoint('weights_hands.h5', \n",
    "                             monitor='val_loss', \n",
    "                             verbose=1, \n",
    "                             save_best_only=True, \n",
    "                             mode='min', \n",
    "                             period=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**開始訓練**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "ExecuteTime": {
     "start_time": "2017-11-26T20:38:54.037Z"
    },
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 00001: val_loss improved from inf to 0.11776, saving model to weights_hands.h5\n",
      "Epoch 00002: val_loss improved from 0.11776 to 0.05966, saving model to weights_hands.h5\n",
      "Epoch 00003: val_loss improved from 0.05966 to 0.05382, saving model to weights_hands.h5\n",
      "Epoch 00004: val_loss improved from 0.05382 to 0.04547, saving model to weights_hands.h5\n",
      "Epoch 00005: val_loss improved from 0.04547 to 0.03713, saving model to weights_hands.h5\n",
      "Epoch 00006: val_loss did not improve\n",
      "Epoch 00007: val_loss improved from 0.03713 to 0.03611, saving model to weights_hands.h5\n",
      "Epoch 00008: val_loss improved from 0.03611 to 0.03438, saving model to weights_hands.h5\n",
      "Epoch 00009: val_loss improved from 0.03438 to 0.02191, saving model to weights_hands.h5\n",
      "Epoch 00010: val_loss did not improve\n",
      "Epoch 00011: val_loss improved from 0.02191 to 0.01899, saving model to weights_hands.h5\n",
      "Epoch 00012: val_loss did not improve\n",
      "Epoch 00013: val_loss did not improve\n",
      "Epoch 00014: val_loss did not improve\n",
      "Epoch 00014: early stopping\n"
     ]
    }
   ],
   "source": [
    "optimizer = Adam(lr=0.5e-4, beta_1=0.9, beta_2=0.999, epsilon=1e-08, decay=0.0)\n",
    "#optimizer = SGD(lr=1e-4, decay=0.0005, momentum=0.9)\n",
    "#optimizer = RMSprop(lr=1e-4, rho=0.9, epsilon=1e-08, decay=0.0)\n",
    "\n",
    "model.compile(loss=custom_loss, optimizer=optimizer)\n",
    "\n",
    "history = model.fit_generator(generator = train_batch, \n",
    "                    steps_per_epoch  = len(train_batch), \n",
    "                    epochs           = 20, # 應該增加更多次的訓練循環 \n",
    "                    verbose          = 0,\n",
    "                    validation_data  = valid_batch,\n",
    "                    validation_steps = len(valid_batch),\n",
    "                    callbacks        = [early_stop, checkpoint], \n",
    "                    max_queue_size   = 3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### STEP 6. 圖像的物體偵測"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 載入訓練好的模型權重\n",
    "model.load_weights(\"weights_hands.h5\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-11-22T14:07:49.271978Z",
     "start_time": "2017-11-22T14:07:49.268999Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 產生一個Dummy的標籤輸入\n",
    "\n",
    "# 在訓練階段放的是真實的邊界框與圖像類別訊息\n",
    "# 但在預測階段還是需要有一個Dummy的輸入, 因為定義在網絡的結構中有兩個輸入： \n",
    "#   1.圖像的輸人 \n",
    "#   2.圖像邊界框/錨點/信心分數的輸入\n",
    "dummy_array = np.zeros((1,1,1,1,TRUE_BOX_BUFFER,4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlkAAAFdCAYAAAAwgXjMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRYsWLVq02Af8f4znglOUi+bMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1e82f6c6b00>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 選一張圖像\n",
    "img_filepath = train_imgs[np.random.randint(len(train_imgs))]['filename']\n",
    "\n",
    "# 使用OpenCV讀入圖像\n",
    "image = cv2.imread(img_filepath) # 載入圖像\n",
    "\n",
    "plt.figure(figsize=(10,10))\n",
    "\n",
    "# 進行圖像輸入的前處理\n",
    "input_image = cv2.resize(image, (416, 416)) # 修改輸入圖像大小來符合模型的要求\n",
    "input_image = input_image / 255. # 進行圖像歸一處理\n",
    "input_image = np.expand_dims(input_image, 0) # 增加 batch dimension\n",
    "\n",
    "# 進行圖像偵測\n",
    "netout = model.predict([input_image, dummy_array])\n",
    "\n",
    "# 解析網絡的輸出來取得最後偵測出來的邊界框(bounding boxes)列表\n",
    "boxes = decode_netout(netout[0], \n",
    "                      obj_threshold=OBJ_THRESHOLD,\n",
    "                      nms_threshold=NMS_THRESHOLD,\n",
    "                      anchors=ANCHORS, \n",
    "                      nb_class=CLASS)\n",
    "\n",
    "# \"draw_bgr_image_boxes\"\n",
    "# 一個簡單把邊界框與預測結果打印到原始圖像(BGR)上的工具函式\n",
    "# 參數: image 是image的numpy ndarray [h, w, channels(BGR)]\n",
    "#       boxes 是偵測的結果\n",
    "#       labels 是模型訓練的圖像類別列表\n",
    "# 回傳： image 是image的numpy ndarray [h, w, channels(RGB)]\n",
    "image = draw_bgr_image_boxes(image, boxes, labels=LABELS)\n",
    "\n",
    "# 把最後的結果秀出來\n",
    "plt.imshow(image)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## STEP 6. 影像的物體偵測"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-10-06T13:28:28.029334Z",
     "start_time": "2017-10-06T13:28:28.024662Z"
    },
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 載入訓練好的模型權重\n",
    "model.load_weights(\"weights_hands.h5\")\n",
    "\n",
    "# 產生一個Dummy的標籤輸入\n",
    "\n",
    "# 在訓練階段放的是真實的邊界框與圖像類別訊息\n",
    "# 但在預測階段還是需要有一個Dummy的輸入, 因為定義在網絡的結構中有兩個輸入： \n",
    "#   1.圖像的輸人 \n",
    "#   2.圖像邊界框/錨點/信心分數的輸入\n",
    "dummy_array = np.zeros((1,1,1,1,TRUE_BOX_BUFFER,4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2017-10-06T13:39:09.640646Z",
     "start_time": "2017-10-06T13:31:44.627609Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|████████████████████████████████████████████████████████████████████████████| 17191/17191 [07:58<00:00, 35.91it/s]\n"
     ]
    }
   ],
   "source": [
    "# 資料集目錄\n",
    "VIDEO_DATA_PATH = os.path.join(DATA_SET_PATH, \"video\")\n",
    "\n",
    "# 選擇要進行浣熊影像偵測的影像檔\n",
    "# 在這個測試我從YOUTUBE下載了: https://www.youtube.com/watch?v=c0IykwK6zkY\n",
    "video_inp =  os.path.join(VIDEO_DATA_PATH, \"cardriving.mp4\")\n",
    "\n",
    "# 偵測結果的輸出影像檔\n",
    "video_out =  os.path.join(VIDEO_DATA_PATH, \"cardriving-out.mp4\")\n",
    "\n",
    "# 透過OpenCv擷取影像\n",
    "video_reader = cv2.VideoCapture(video_inp)\n",
    "\n",
    "# 取得影像的基本資訊\n",
    "nb_frames = int(video_reader.get(cv2.CAP_PROP_FRAME_COUNT)) # 總共有多少frames\n",
    "frame_h = int(video_reader.get(cv2.CAP_PROP_FRAME_HEIGHT))  # 每個frame的高\n",
    "frame_w = int(video_reader.get(cv2.CAP_PROP_FRAME_WIDTH))   # 每個frame的寬\n",
    "\n",
    "# 設定影像的輸出\n",
    "video_writer = cv2.VideoWriter(video_out,\n",
    "                               cv2.VideoWriter_fourcc(*'XVID'), \n",
    "                               50.0, \n",
    "                               (frame_w, frame_h))\n",
    "\n",
    "# 迭代每一個frame來進行圖像偵測\n",
    "for i in tqdm(range(nb_frames)):\n",
    "    ret, image = video_reader.read() # 讀取一個frame\n",
    "    \n",
    "    input_image = cv2.resize(image, (416, 416)) # 修改輸入圖像大小來符合模型的要求\n",
    "    input_image = input_image / 255. # 進行圖像歸一處理\n",
    "    input_image = np.expand_dims(input_image, 0) # 增加 batch dimension\n",
    "\n",
    "    # 進行圖像偵測\n",
    "    netout = model.predict([input_image, dummy_array])\n",
    "\n",
    "    # 解析網絡的輸出來取得最後偵測出來的邊界框(bounding boxes)列表\n",
    "    boxes = decode_netout(netout[0], \n",
    "                          obj_threshold=OBJ_THRESHOLD,\n",
    "                          nms_threshold=NMS_THRESHOLD,\n",
    "                          anchors=ANCHORS, \n",
    "                          nb_class=CLASS)\n",
    "    \n",
    "    # \"draw_bgr_image_boxes\"\n",
    "    # 一個簡單把邊界框與預測結果打印到原始圖像(BGR)上的工具函式\n",
    "    # 參數: image 是image的numpy ndarray [h, w, channels(BGR)]\n",
    "    #       boxes 是偵測的結果\n",
    "    #       labels 是模型訓練的圖像類別列表\n",
    "    # 回傳： image 是image的numpy ndarray [h, w, channels(RGB)]\n",
    "    image = draw_bgr_image_boxes(image, boxes, labels=LABELS)\n",
    "\n",
    "    # 透過OpenCV把影像輸出出來\n",
    "    video_writer.write(np.uint8(image[:,:,::-1])) # 轉換 RGB -> BGR來讓Open CV寫Video\n",
    "    \n",
    "video_reader.release() # 釋放資源\n",
    "video_writer.release() # 釋放資源"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "參考:\n",
    "* [YOLO官網](https://pjreddie.com/darknet/yolo/)\n",
    "* [llSourcell/YOLO_Object_Detection](https://github.com/llSourcell/YOLO_Object_Detection)\n",
    "* [experiencor/basic-yolo-keras](https://github.com/experiencor/basic-yolo-keras)\n",
    "* [VIVA Hand Detection Challenge](http://cvrr.ucsd.edu/vivachallenge/index.php/hands/hand-detection/)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "hide_input": false,
  "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": "3.5.4"
  },
  "toc": {
   "nav_menu": {
    "height": "381px",
    "width": "251px"
   },
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "toc_cell": false,
   "toc_position": {
    "height": "758px",
    "left": "0px",
    "right": "1096px",
    "top": "73px",
    "width": "253px"
   },
   "toc_section_display": "block",
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}