Create json Annotation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline\n",
    "%load_ext autoreload\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from math import sqrt\n",
    "from skimage import data\n",
    "from skimage.feature import blob_dog, blob_log, blob_doh\n",
    "from skimage.color import rgb2gray, gray2rgb\n",
    "from skimage.io import imread, imshow\n",
    "import cv2\n",
    "\n",
    "import pycocotools\n",
    "\n",
    "from pycocotools import mask as mask_coco\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "\n",
    "image = imread(\"TEETH_DATASET/masks/cate1-00001.bmp\")\n",
    "image_gray = rgb2gray(image)\n",
    "image = gray2rgb(image)\n",
    "\n",
    "ret, thresh = cv2.threshold(image_gray, 0, 255, 0)\n",
    "im2, contours, hierarchy = cv2.findContours(thresh, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)\n",
    "\n",
    "for i in range(len(contours)):\n",
    "    img_cont = np.zeros(image.shape, dtype=image.dtype)\n",
    "\n",
    "    cv2.drawContours(img_cont, contours, i, (255,255,255), -1)\n",
    "\n",
    "    #imshow(img_cont)\n",
    "\n",
    "    bimask = np.zeros(img_cont.shape[0:2], dtype=np.uint8)\n",
    "\n",
    "    bimask = img_cont[:,:, 0]\n",
    "\n",
    "    #print(len(contours))\n",
    "    \n",
    "    cnt = contours[2]\n",
    "    x,y,w,h = cv2.boundingRect(cnt)\n",
    "    cv2.namedWindow('image', cv2.WINDOW_NORMAL)\n",
    "    cv2.imshow('image',bimask)\n",
    "    cv2.waitKey(0)\n",
    "    cv2.destroyAllWindows()\n",
    "    #break\n",
    "    #print(x,y,w,h)\n",
    "#M = cv2.moments(cnt)\n",
    "#print( cnt )\n",
    "\n",
    "#mask_coco.encode(img_cont)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "'''# Import libraries\n",
    "import numpy as np # linear algebra\n",
    "from PIL import Image\n",
    "\n",
    "# Fast run length encoding\n",
    "def rle (img):\n",
    "    flat_img = img.flatten()\n",
    "    flat_img = np.where(flat_img > 0.5, 1, 0).astype(np.uint8)\n",
    "\n",
    "    starts = np.array((flat_img[:-1] == 0) & (flat_img[1:] == 1))\n",
    "    ends = np.array((flat_img[:-1] == 1) & (flat_img[1:] == 0))\n",
    "    starts_ix = np.where(starts)[0] + 2\n",
    "    ends_ix = np.where(ends)[0] + 2\n",
    "    lengths = ends_ix - starts_ix\n",
    "    \n",
    "    return starts_ix, lengths\n",
    "\n",
    "mask = bimask\n",
    "mask_rle = rle(mask)\n",
    "print(mask_rle)\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['cate5-00091.png']\n",
      "0 1\n",
      "JSON created...OK in:  %s /workspace/teeth_net/temp_teeth_daquelaveznaofoimasagoravai/accuracy_score/best/annotations/instances_train2018.json\n",
      "JSON created...OK in:  %s /workspace/teeth_net/temp_teeth_daquelaveznaofoimasagoravai/accuracy_score/best/annotations/instances_test2018.json\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.5/dist-packages/sklearn/model_selection/_split.py:2026: FutureWarning: From version 0.21, test_size will always complement train_size unless both are specified.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'annotations': [{'area': 18343,\n",
       "   'bbox': [397.0, 563.0, 501.0, 168.0],\n",
       "   'category_id': 1,\n",
       "   'height': 1024,\n",
       "   'id': 0,\n",
       "   'image_id': 0,\n",
       "   'iscrowd': 1,\n",
       "   'segmentation': {'counts': [407203,\n",
       "     19,\n",
       "     999,\n",
       "     29,\n",
       "     989,\n",
       "     36,\n",
       "     982,\n",
       "     42,\n",
       "     976,\n",
       "     48,\n",
       "     969,\n",
       "     55,\n",
       "     958,\n",
       "     65,\n",
       "     953,\n",
       "     71,\n",
       "     952,\n",
       "     71,\n",
       "     949,\n",
       "     73,\n",
       "     948,\n",
       "     74,\n",
       "     950,\n",
       "     71,\n",
       "     944,\n",
       "     77,\n",
       "     946,\n",
       "     74,\n",
       "     949,\n",
       "     72,\n",
       "     952,\n",
       "     70,\n",
       "     953,\n",
       "     67,\n",
       "     957,\n",
       "     62,\n",
       "     962,\n",
       "     57,\n",
       "     967,\n",
       "     51,\n",
       "     973,\n",
       "     44,\n",
       "     980,\n",
       "     39,\n",
       "     986,\n",
       "     35,\n",
       "     989,\n",
       "     8,\n",
       "     3,\n",
       "     21,\n",
       "     1005,\n",
       "     2,\n",
       "     6,\n",
       "     11,\n",
       "     1,\n",
       "     3,\n",
       "     16,\n",
       "     50,\n",
       "     946,\n",
       "     81,\n",
       "     943,\n",
       "     82,\n",
       "     941,\n",
       "     83,\n",
       "     941,\n",
       "     83,\n",
       "     941,\n",
       "     83,\n",
       "     941,\n",
       "     83,\n",
       "     940,\n",
       "     84,\n",
       "     940,\n",
       "     84,\n",
       "     940,\n",
       "     84,\n",
       "     939,\n",
       "     85,\n",
       "     939,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     79,\n",
       "     2,\n",
       "     5,\n",
       "     938,\n",
       "     71,\n",
       "     954,\n",
       "     61,\n",
       "     964,\n",
       "     60,\n",
       "     964,\n",
       "     59,\n",
       "     965,\n",
       "     58,\n",
       "     966,\n",
       "     56,\n",
       "     969,\n",
       "     48,\n",
       "     976,\n",
       "     37,\n",
       "     987,\n",
       "     27,\n",
       "     998,\n",
       "     22,\n",
       "     1002,\n",
       "     19,\n",
       "     1005,\n",
       "     16,\n",
       "     1009,\n",
       "     11,\n",
       "     1013,\n",
       "     8,\n",
       "     1017,\n",
       "     5,\n",
       "     1020,\n",
       "     3,\n",
       "     16451,\n",
       "     14,\n",
       "     1002,\n",
       "     23,\n",
       "     992,\n",
       "     32,\n",
       "     983,\n",
       "     41,\n",
       "     975,\n",
       "     50,\n",
       "     967,\n",
       "     57,\n",
       "     960,\n",
       "     64,\n",
       "     944,\n",
       "     80,\n",
       "     941,\n",
       "     83,\n",
       "     940,\n",
       "     84,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     937,\n",
       "     87,\n",
       "     937,\n",
       "     87,\n",
       "     937,\n",
       "     87,\n",
       "     937,\n",
       "     88,\n",
       "     936,\n",
       "     88,\n",
       "     936,\n",
       "     88,\n",
       "     936,\n",
       "     88,\n",
       "     936,\n",
       "     88,\n",
       "     936,\n",
       "     88,\n",
       "     936,\n",
       "     87,\n",
       "     937,\n",
       "     87,\n",
       "     937,\n",
       "     87,\n",
       "     937,\n",
       "     86,\n",
       "     938,\n",
       "     86,\n",
       "     938,\n",
       "     84,\n",
       "     941,\n",
       "     73,\n",
       "     951,\n",
       "     68,\n",
       "     956,\n",
       "     63,\n",
       "     961,\n",
       "     59,\n",
       "     965,\n",
       "     56,\n",
       "     969,\n",
       "     52,\n",
       "     972,\n",
       "     48,\n",
       "     976,\n",
       "     44,\n",
       "     980,\n",
       "     41,\n",
       "     983,\n",
       "     38,\n",
       "     987,\n",
       "     34,\n",
       "     990,\n",
       "     30,\n",
       "     994,\n",
       "     27,\n",
       "     997,\n",
       "     24,\n",
       "     1001,\n",
       "     20,\n",
       "     1004,\n",
       "     17,\n",
       "     1007,\n",
       "     17,\n",
       "     1007,\n",
       "     15,\n",
       "     1009,\n",
       "     14,\n",
       "     1011,\n",
       "     11,\n",
       "     1013,\n",
       "     8,\n",
       "     1016,\n",
       "     7,\n",
       "     1017,\n",
       "     7,\n",
       "     1020,\n",
       "     2,\n",
       "     8260,\n",
       "     4,\n",
       "     1016,\n",
       "     8,\n",
       "     1011,\n",
       "     14,\n",
       "     1007,\n",
       "     17,\n",
       "     1002,\n",
       "     23,\n",
       "     996,\n",
       "     28,\n",
       "     991,\n",
       "     33,\n",
       "     987,\n",
       "     38,\n",
       "     983,\n",
       "     42,\n",
       "     979,\n",
       "     45,\n",
       "     976,\n",
       "     48,\n",
       "     973,\n",
       "     52,\n",
       "     969,\n",
       "     55,\n",
       "     965,\n",
       "     59,\n",
       "     959,\n",
       "     66,\n",
       "     955,\n",
       "     69,\n",
       "     948,\n",
       "     76,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     78,\n",
       "     946,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     75,\n",
       "     949,\n",
       "     70,\n",
       "     954,\n",
       "     67,\n",
       "     957,\n",
       "     65,\n",
       "     959,\n",
       "     63,\n",
       "     961,\n",
       "     62,\n",
       "     962,\n",
       "     61,\n",
       "     964,\n",
       "     58,\n",
       "     966,\n",
       "     56,\n",
       "     968,\n",
       "     54,\n",
       "     970,\n",
       "     52,\n",
       "     973,\n",
       "     48,\n",
       "     977,\n",
       "     45,\n",
       "     979,\n",
       "     42,\n",
       "     982,\n",
       "     40,\n",
       "     984,\n",
       "     37,\n",
       "     988,\n",
       "     33,\n",
       "     991,\n",
       "     31,\n",
       "     993,\n",
       "     28,\n",
       "     996,\n",
       "     25,\n",
       "     1000,\n",
       "     21,\n",
       "     1003,\n",
       "     18,\n",
       "     1007,\n",
       "     15,\n",
       "     1009,\n",
       "     13,\n",
       "     1011,\n",
       "     12,\n",
       "     1013,\n",
       "     9,\n",
       "     1015,\n",
       "     8,\n",
       "     1017,\n",
       "     5,\n",
       "     1019,\n",
       "     4,\n",
       "     1084,\n",
       "     8,\n",
       "     1010,\n",
       "     14,\n",
       "     1005,\n",
       "     20,\n",
       "     998,\n",
       "     27,\n",
       "     992,\n",
       "     32,\n",
       "     988,\n",
       "     36,\n",
       "     986,\n",
       "     38,\n",
       "     982,\n",
       "     43,\n",
       "     977,\n",
       "     47,\n",
       "     973,\n",
       "     51,\n",
       "     969,\n",
       "     55,\n",
       "     963,\n",
       "     62,\n",
       "     957,\n",
       "     67,\n",
       "     948,\n",
       "     4,\n",
       "     2,\n",
       "     70,\n",
       "     948,\n",
       "     76,\n",
       "     948,\n",
       "     76,\n",
       "     948,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     946,\n",
       "     78,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     947,\n",
       "     77,\n",
       "     948,\n",
       "     75,\n",
       "     949,\n",
       "     74,\n",
       "     950,\n",
       "     72,\n",
       "     952,\n",
       "     69,\n",
       "     955,\n",
       "     67,\n",
       "     958,\n",
       "     63,\n",
       "     961,\n",
       "     60,\n",
       "     964,\n",
       "     56,\n",
       "     968,\n",
       "     53,\n",
       "     972,\n",
       "     49,\n",
       "     975,\n",
       "     45,\n",
       "     979,\n",
       "     42,\n",
       "     982,\n",
       "     38,\n",
       "     987,\n",
       "     34,\n",
       "     990,\n",
       "     31,\n",
       "     993,\n",
       "     27,\n",
       "     997,\n",
       "     24,\n",
       "     1001,\n",
       "     20,\n",
       "     1004,\n",
       "     15,\n",
       "     1009,\n",
       "     15,\n",
       "     1009,\n",
       "     14,\n",
       "     36,\n",
       "     3,\n",
       "     972,\n",
       "     11,\n",
       "     34,\n",
       "     10,\n",
       "     969,\n",
       "     9,\n",
       "     31,\n",
       "     16,\n",
       "     968,\n",
       "     8,\n",
       "     28,\n",
       "     21,\n",
       "     968,\n",
       "     5,\n",
       "     26,\n",
       "     25,\n",
       "     969,\n",
       "     2,\n",
       "     25,\n",
       "     29,\n",
       "     992,\n",
       "     32,\n",
       "     988,\n",
       "     36,\n",
       "     985,\n",
       "     40,\n",
       "     981,\n",
       "     43,\n",
       "     977,\n",
       "     47,\n",
       "     974,\n",
       "     51,\n",
       "     962,\n",
       "     3,\n",
       "     5,\n",
       "     54,\n",
       "     961,\n",
       "     64,\n",
       "     959,\n",
       "     65,\n",
       "     959,\n",
       "     65,\n",
       "     958,\n",
       "     66,\n",
       "     958,\n",
       "     67,\n",
       "     957,\n",
       "     67,\n",
       "     957,\n",
       "     67,\n",
       "     957,\n",
       "     67,\n",
       "     957,\n",
       "     67,\n",
       "     958,\n",
       "     67,\n",
       "     957,\n",
       "     66,\n",
       "     958,\n",
       "     66,\n",
       "     958,\n",
       "     64,\n",
       "     960,\n",
       "     62,\n",
       "     963,\n",
       "     59,\n",
       "     965,\n",
       "     57,\n",
       "     967,\n",
       "     55,\n",
       "     969,\n",
       "     53,\n",
       "     972,\n",
       "     50,\n",
       "     974,\n",
       "     48,\n",
       "     976,\n",
       "     46,\n",
       "     978,\n",
       "     43,\n",
       "     982,\n",
       "     41,\n",
       "     983,\n",
       "     39,\n",
       "     985,\n",
       "     36,\n",
       "     988,\n",
       "     33,\n",
       "     992,\n",
       "     28,\n",
       "     996,\n",
       "     24,\n",
       "     1001,\n",
       "     18,\n",
       "     1006,\n",
       "     12,\n",
       "     1013,\n",
       "     10,\n",
       "     1016,\n",
       "     6,\n",
       "     129983,\n",
       "     6,\n",
       "     1015,\n",
       "     11,\n",
       "     1012,\n",
       "     13,\n",
       "     1009,\n",
       "     17,\n",
       "     1006,\n",
       "     19,\n",
       "     1003,\n",
       "     22,\n",
       "     1001,\n",
       "     24,\n",
       "     998,\n",
       "     27,\n",
       "     995,\n",
       "     30,\n",
       "     993,\n",
       "     32,\n",
       "     991,\n",
       "     33,\n",
       "     989,\n",
       "     36,\n",
       "     987,\n",
       "     38,\n",
       "     985,\n",
       "     40,\n",
       "     983,\n",
       "     42,\n",
       "     981,\n",
       "     43,\n",
       "     980,\n",
       "     45,\n",
       "     978,\n",
       "     46,\n",
       "     977,\n",
       "     48,\n",
       "     975,\n",
       "     49,\n",
       "     974,\n",
       "     51,\n",
       "     973,\n",
       "     51,\n",
       "     973,\n",
       "     52,\n",
       "     972,\n",
       "     53,\n",
       "     971,\n",
       "     53,\n",
       "     971,\n",
       "     54,\n",
       "     970,\n",
       "     54,\n",
       "     969,\n",
       "     55,\n",
       "     969,\n",
       "     56,\n",
       "     968,\n",
       "     56,\n",
       "     968,\n",
       "     57,\n",
       "     965,\n",
       "     59,\n",
       "     962,\n",
       "     63,\n",
       "     958,\n",
       "     67,\n",
       "     954,\n",
       "     70,\n",
       "     954,\n",
       "     71,\n",
       "     952,\n",
       "     73,\n",
       "     951,\n",
       "     73,\n",
       "     950,\n",
       "     75,\n",
       "     949,\n",
       "     76,\n",
       "     947,\n",
       "     78,\n",
       "     946,\n",
       "     78,\n",
       "     946,\n",
       "     79,\n",
       "     945,\n",
       "     79,\n",
       "     945,\n",
       "     80,\n",
       "     944,\n",
       "     81,\n",
       "     943,\n",
       "     81,\n",
       "     943,\n",
       "     81,\n",
       "     944,\n",
       "     81,\n",
       "     943,\n",
       "     82,\n",
       "     943,\n",
       "     81,\n",
       "     943,\n",
       "     82,\n",
       "     943,\n",
       "     81,\n",
       "     944,\n",
       "     82,\n",
       "     943,\n",
       "     81,\n",
       "     944,\n",
       "     81,\n",
       "     944,\n",
       "     80,\n",
       "     945,\n",
       "     80,\n",
       "     946,\n",
       "     79,\n",
       "     947,\n",
       "     77,\n",
       "     949,\n",
       "     76,\n",
       "     950,\n",
       "     74,\n",
       "     952,\n",
       "     72,\n",
       "     954,\n",
       "     70,\n",
       "     956,\n",
       "     68,\n",
       "     957,\n",
       "     67,\n",
       "     959,\n",
       "     65,\n",
       "     961,\n",
       "     63,\n",
       "     962,\n",
       "     61,\n",
       "     966,\n",
       "     58,\n",
       "     967,\n",
       "     57,\n",
       "     969,\n",
       "     54,\n",
       "     971,\n",
       "     52,\n",
       "     973,\n",
       "     36,\n",
       "     1,\n",
       "     13,\n",
       "     975,\n",
       "     36,\n",
       "     2,\n",
       "     9,\n",
       "     977,\n",
       "     36,\n",
       "     3,\n",
       "     5,\n",
       "     981,\n",
       "     36,\n",
       "     989,\n",
       "     35,\n",
       "     990,\n",
       "     35,\n",
       "     989,\n",
       "     35,\n",
       "     990,\n",
       "     34,\n",
       "     991,\n",
       "     33,\n",
       "     992,\n",
       "     32,\n",
       "     992,\n",
       "     33,\n",
       "     993,\n",
       "     31,\n",
       "     994,\n",
       "     30,\n",
       "     995,\n",
       "     29,\n",
       "     997,\n",
       "     27,\n",
       "     999,\n",
       "     24,\n",
       "     1001,\n",
       "     23,\n",
       "     1003,\n",
       "     21,\n",
       "     1005,\n",
       "     18,\n",
       "     1008,\n",
       "     14,\n",
       "     1012,\n",
       "     11,\n",
       "     1018,\n",
       "     1,\n",
       "     129419],\n",
       "    'size': [1024, 1024]},\n",
       "   'width': 1024}],\n",
       " 'categories': [{'id': 1, 'name': 'tooth', 'supercategory': 'teeth'}],\n",
       " 'images': [{'coco_url': '',\n",
       "   'date_captured': '2018-06-08 19:11:55.169890',\n",
       "   'file_name': 'cate5-00091.jpg',\n",
       "   'flickr_url': '',\n",
       "   'height': 1024,\n",
       "   'id': 0,\n",
       "   'license': 1,\n",
       "   'width': 1024}],\n",
       " 'info': {'contributor': 'IvisionLab',\n",
       "  'date_created': '2018-06-08 18:36:22.309424',\n",
       "  'description': 'TEETH DATASET',\n",
       "  'url': 'https://github.com/gil.../',\n",
       "  'version': '0.1.0',\n",
       "  'year': 2018},\n",
       " 'licenses': [{'id': 1,\n",
       "   'name': 'Attribution-NonCommercial-ShareAlike License',\n",
       "   'url': 'http://creativecommons.org/licenses/by-nc-sa/2.0/'}]}"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%autoreload 2\n",
    "import numpy as np\n",
    "import random\n",
    "import os\n",
    "import fnmatch\n",
    "import datetime\n",
    "from sklearn import model_selection\n",
    "from PIL import Image\n",
    "import pycococreatortool_teeth as cococreator_teeth\n",
    "\n",
    "# dataset_path = '/workspace/teeth_net/classified_panoramic_X-ray_images_dataset/mask'\n",
    "# dataset_path_save = '/workspace/teeth_net/classified_panoramic_X-ray_images_dataset'\n",
    "\n",
    "#accuracy_score\n",
    "#f1_score\n",
    "#precision_score\n",
    "#recall_score\n",
    "#specificity_score\n",
    "\n",
    "dataset_path = '/workspace/teeth_net/temp_teeth_daquelaveznaofoimasagoravai/accuracy_score/best/expected'\n",
    "dataset_path_save = '/workspace/teeth_net/temp_teeth_daquelaveznaofoimasagoravai/accuracy_score/best/'\n",
    "\n",
    "\n",
    "images_path = os.listdir(dataset_path)\n",
    "#images_path = fnmatch.filter(images_path, '*.png')\n",
    "train_data, val_data = model_selection.train_test_split(images_path, train_size=0, shuffle=True)\n",
    "arr_dict_images_train = []\n",
    "arr_dict_images_val = []\n",
    "arr_dict_annotation_train = []\n",
    "arr_dict_annotation_val = []\n",
    "annotation_id = 0\n",
    "\n",
    "path_dir_train = '/workspace/teeth_net/temp_teeth_daquelaveznaofoimasagoravai/accuracy_score/best/expected'#\"/workspace/teeth_net/TEETH_DATASET/masks\"\n",
    "paths_files_train = [f.split(\".\")[0] for f in os.listdir(path_dir_train)]\n",
    "\n",
    "\n",
    "#images_path = [f for f in images_path if f.split('.')[0] not in paths_files_train]\n",
    "print(images_path)\n",
    "for img_id, img_path in enumerate(images_path):\n",
    "    img_fullpath = os.path.join(dataset_path, img_path)\n",
    "    binary_mask = np.asarray(Image.open(img_fullpath).convert('1')).astype(np.uint8)\n",
    "    \n",
    "    # Create dict images_info\n",
    "    dict_images = cococreator_teeth.create_image_info(\n",
    "        image_id = img_id,\n",
    "        file_name = img_path.replace('.png', '.jpg'), \n",
    "        image_size = binary_mask.shape,\n",
    "        date_captured = datetime.datetime.utcnow().isoformat(' '))\n",
    "    \n",
    "    if img_path in train_data: arr_dict_images_train.append(dict_images)\n",
    "    elif img_path in val_data: arr_dict_images_val.append(dict_images)\n",
    "    \n",
    "    # Create dict annotations_info for each image\n",
    "    annotations = cococreator_teeth.get_teeth_masks_test(img_fullpath)\n",
    "    for ann in annotations:\n",
    "        dict_annotation_info = cococreator_teeth.create_annotation_info(\n",
    "                annotation_id = annotation_id, \n",
    "                image_id = img_id, \n",
    "                binary_mask = ann,\n",
    "                class_id = 1,\n",
    "                is_crowd = 1,\n",
    "                image_size = binary_mask.shape,\n",
    "                tolerance = 2, \n",
    "                bounding_box = None)\n",
    "        annotation_id +=1 \n",
    "        if dict_annotation_info is None: continue\n",
    "        if img_path in train_data: arr_dict_annotation_train.append(dict_annotation_info)\n",
    "        elif img_path in val_data: arr_dict_annotation_val.append(dict_annotation_info)\n",
    "        \n",
    "print(len(arr_dict_images_train), len(arr_dict_images_val))\n",
    "# Build JSON for subsets\n",
    "#dataset_path = os.path.join(dataset_path, os.pardir) # Go Back 1 folder\n",
    "#print(arr_dict_images_train)\n",
    "cococreator_teeth.build_coco_format(images=arr_dict_images_train, \n",
    "                                    annotations=arr_dict_annotation_train, \n",
    "                                    dataset_path=dataset_path_save, \n",
    "                                    subset_dir='train2018')\n",
    "cococreator_teeth.build_coco_format(images=arr_dict_images_val, \n",
    "                                    annotations=arr_dict_annotation_val, \n",
    "                                    dataset_path=dataset_path_save, \n",
    "                                    subset_dir='test2018')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "arr_dict_annotation_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}