Skip to content
Snippets Groups Projects
ball_analysis.ipynb 80.3 KiB
Newer Older
gautham's avatar
gautham committed
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# read the pickle file\n",
    "with open('../tracker_stubs/ball_detections.pkl', 'rb') as f:\n",
    "    ball_positions = pickle.load(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "ball_positions = [x.get(1,[]) for x in ball_positions]\n",
    "# convert the list into pandas dataframe\n",
    "df_ball_positions = pd.DataFrame(ball_positions,columns=['x1','y1','x2','y2'])\n",
    "\n",
    "# interpolate the missing values\n",
    "df_ball_positions = df_ball_positions.interpolate()\n",
    "df_ball_positions = df_ball_positions.bfill()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_ball_positions['mid_y'] = (df_ball_positions['y1'] + df_ball_positions['y2'])/2\n",
    "df_ball_positions['mid_y_rolling_mean'] = df_ball_positions['mid_y'].rolling(window=5, min_periods=1, center=False).mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x20f1bba1d10>]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot mid_y_rolling_mean\n",
    "plt.plot(df_ball_positions['mid_y_rolling_mean'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_ball_positions['delta_y'] = df_ball_positions['mid_y_rolling_mean'].diff()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x20f444c1390>]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot mid_y_rolling_mean\n",
    "plt.plot(df_ball_positions['delta_y'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_ball_positions['ball_hit']=0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[11, 58, 95, 131, 182]\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Ensure 'ball_hit' column exists and initialize it with 0\n",
    "df_ball_positions['ball_hit'] = 0\n",
    "\n",
    "# Define the minimum frames required for a hit\n",
    "minimum_change_frames_for_hit = 25\n",
    "\n",
    "# Loop through each frame to detect hits\n",
    "for i in range(1, len(df_ball_positions) - int(minimum_change_frames_for_hit * 1.2)):\n",
    "    negative_position_change = (\n",
    "        df_ball_positions['delta_y'].iloc[i] > 0 and df_ball_positions['delta_y'].iloc[i + 1] < 0\n",
    "    )\n",
    "    positive_position_change = (\n",
    "        df_ball_positions['delta_y'].iloc[i] < 0 and df_ball_positions['delta_y'].iloc[i + 1] > 0\n",
    "    )\n",
    "\n",
    "    if negative_position_change or positive_position_change:\n",
    "        change_count = 0\n",
    "\n",
    "        # Check subsequent frames for consistent position changes\n",
    "        for change_frame in range(i + 1, i + int(minimum_change_frames_for_hit * 1.2) + 1):\n",
    "            negative_position_change_following_frame = (\n",
    "                df_ball_positions['delta_y'].iloc[i] > 0 and df_ball_positions['delta_y'].iloc[change_frame] < 0\n",
    "            )\n",
    "            positive_position_change_following_frame = (\n",
    "                df_ball_positions['delta_y'].iloc[i] < 0 and df_ball_positions['delta_y'].iloc[change_frame] > 0\n",
    "            )\n",
    "\n",
    "            if negative_position_change and negative_position_change_following_frame:\n",
    "                change_count += 1\n",
    "            elif positive_position_change and positive_position_change_following_frame:\n",
    "                change_count += 1\n",
    "\n",
    "        # Mark the ball hit if enough changes are detected\n",
    "        if change_count > minimum_change_frames_for_hit - 1:\n",
    "            df_ball_positions.at[i, 'ball_hit'] = 1  # Use .at for safer assignment\n",
    "\n",
    "# Get the frame numbers where ball hits were detected\n",
    "frame_nums_with_ball_hits = df_ball_positions[df_ball_positions['ball_hit'] == 1].index.tolist()\n",
    "\n",
    "# Print or return the result if needed\n",
    "print(frame_nums_with_ball_hits)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>x1</th>\n",
       "      <th>y1</th>\n",
       "      <th>x2</th>\n",
       "      <th>y2</th>\n",
       "      <th>mid_y</th>\n",
       "      <th>mid_y_rolling_mean</th>\n",
       "      <th>delta_y</th>\n",
       "      <th>ball_hit</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>776.865967</td>\n",
       "      <td>717.330017</td>\n",
       "      <td>796.806519</td>\n",
       "      <td>738.393188</td>\n",
       "      <td>727.861603</td>\n",
       "      <td>735.918115</td>\n",
       "      <td>6.523407</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>925.881409</td>\n",
       "      <td>240.971042</td>\n",
       "      <td>939.039478</td>\n",
       "      <td>253.989072</td>\n",
       "      <td>247.480057</td>\n",
       "      <td>243.406097</td>\n",
       "      <td>-1.957851</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>624.777161</td>\n",
       "      <td>748.891968</td>\n",
       "      <td>642.157257</td>\n",
       "      <td>766.698242</td>\n",
       "      <td>757.795105</td>\n",
       "      <td>775.403400</td>\n",
       "      <td>0.871759</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>716.963562</td>\n",
       "      <td>229.095024</td>\n",
       "      <td>729.239868</td>\n",
       "      <td>242.786232</td>\n",
       "      <td>235.940628</td>\n",
       "      <td>235.241684</td>\n",
       "      <td>-0.557164</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>1294.891235</td>\n",
       "      <td>739.127197</td>\n",
       "      <td>1314.160156</td>\n",
       "      <td>760.564819</td>\n",
       "      <td>749.846008</td>\n",
       "      <td>738.733578</td>\n",
       "      <td>5.602832</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              x1          y1           x2          y2       mid_y  \\\n",
       "11    776.865967  717.330017   796.806519  738.393188  727.861603   \n",
       "58    925.881409  240.971042   939.039478  253.989072  247.480057   \n",
       "95    624.777161  748.891968   642.157257  766.698242  757.795105   \n",
       "131   716.963562  229.095024   729.239868  242.786232  235.940628   \n",
       "182  1294.891235  739.127197  1314.160156  760.564819  749.846008   \n",
       "\n",
       "     mid_y_rolling_mean   delta_y  ball_hit  \n",
       "11           735.918115  6.523407         1  \n",
       "58           243.406097 -1.957851         1  \n",
       "95           775.403400  0.871759         1  \n",
       "131          235.241684 -0.557164         1  \n",
       "182          738.733578  5.602832         1  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_ball_positions[df_ball_positions['ball_hit']==1]"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "myenv",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}