{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "Lateral line detection in the video -- definitivo" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "import cv2\n", "from pathlib import Path\n", "import numpy as np\n", "import math\n", "import pandas as pd" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Load video" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Opened: True, FPS: 59.94005994005994, Total Frames: 227.0\n" ] } ], "source": [ "video_number = \"3\"\n", "# Define the relative path to the video file\n", "notebook_dir = Path().resolve()\n", "project_root = notebook_dir.parent.parent\n", "video_path = (\n", " project_root\n", " / \"data\"\n", " / f\"recording_{video_number}\"\n", " / f\"Recording_{video_number}.mp4\"\n", ")\n", "video_path = str(video_path)\n", "\n", "# Load the video\n", "cap = cv2.VideoCapture(video_path)\n", "\n", "# Check\n", "print(\n", " f\"Opened: {cap.isOpened()}, FPS: {cap.get(cv2.CAP_PROP_FPS)}, Total Frames: {cap.get(cv2.CAP_PROP_FRAME_COUNT)}\"\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Import horizontal lines" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "# Define the path to the CSV file\n", "input_data_path = (\n", " project_root\n", " / \"notebook\"\n", " / \"lane_detection\"\n", " / \"intermediate_data\"\n", " / \"lane_lines\"\n", " / f\"horizontal_lines_{video_number}.csv\"\n", ")\n", "\n", "# Load the CSV file into a DataFrame\n", "horizontal_lines = pd.read_csv(input_data_path)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Define Functions" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "# Funzione per estendere una linea\n", "def extend_line(x1, y1, x2, y2, length=1000):\n", " # Calcola la lunghezza originale della linea\n", " dx, dy = x2 - x1, y2 - y1\n", " norm = np.sqrt(dx**2 + dy**2) # Distanza euclidea tra i due punti\n", "\n", " # Evita divisioni per zero\n", " if norm == 0:\n", " return x1, y1, x2, y2\n", "\n", " # Calcola i punti estesi\n", " x1_ext = int(x1 - length * (dx / norm))\n", " y1_ext = int(y1 - length * (dy / norm))\n", " x2_ext = int(x2 + length * (dx / norm))\n", " y2_ext = int(y2 + length * (dy / norm))\n", "\n", " return x1_ext, y1_ext, x2_ext, y2_ext" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Drow one line on the frame" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "\"\"\"Disegna la linea sul frame\"\"\"\n", "\n", "\n", "def write_line_on_frame(frame, line):\n", " # Create a copy of the original frame to draw the first line\n", " modified_frame = np.copy(frame)\n", "\n", " if line is not None:\n", " x1, y1, x2, y2 = line\n", "\n", " # Allunga la linea di 1000 pixel da entrambe le estremità\n", " x1_ext, y1_ext, x2_ext, y2_ext = extend_line(x1, y1, x2, y2, length=1000)\n", "\n", " # Disegna la linea estesa\n", " cv2.line(modified_frame, (x1_ext, y1_ext), (x2_ext, y2_ext), (0, 255, 0), 2)\n", "\n", " # x1, y1, x2, y2 = map(int, line)\n", " # cv2.line(modified_frame, (x1, y1), (x2, y2), (0, 255, 0), 2)\n", "\n", " # return the modified frame\n", " return modified_frame" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Drow all the lines on the frame" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "def write_lines_on_frame(frame, lines):\n", " for i in range(len(lines)):\n", " frame = write_line_on_frame(frame, lines[i])\n", " frame = write_line_on_frame(frame, [960, 0, 960, 1300])\n", " return frame" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Transform the line from 'polar coordinates' (hough transform coordinates) to cartesian" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "def polar_to_cartesian(line):\n", " rho, theta = line\n", " a = np.cos(theta)\n", " b = np.sin(theta)\n", " x0 = a * rho\n", " y0 = b * rho\n", " x1 = x0 + 1000 * (-b)\n", " y1 = y0 + 1000 * (a)\n", " x2 = x0 - 1000 * (-b)\n", " y2 = y0 - 1000 * (a)\n", " return np.array([[x1, y1, x2, y2]])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Select the central point of the line on the frame" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "def select_central_point(line, frame):\n", " rho, theta = line\n", " a = np.cos(theta)\n", " b = np.sin(theta)\n", "\n", " # Get the x-size of the image\n", " x_size = frame.shape[1]\n", "\n", " # Calculate the x-coordinate that is half of the x-size\n", " x_half = x_size // 2\n", " print(\"b:\", b)\n", " # Calculate the corresponding y-coordinate using the line equation\n", " y_half = int((rho - x_half * a) / b)\n", "\n", " return x_half, y_half" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Get the edges from the colored image" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "def get_edges(frame):\n", " # Define the range for light brown color in HSV\n", " lower_brown = np.array([00, 30, 100])\n", " upper_brown = np.array([20, 200, 255])\n", "\n", " # Define the range for rose color in HSV\n", " lower_rose = np.array([150, 30, 200])\n", " upper_rose = np.array([180, 200, 255])\n", "\n", " # Convert the image to HSV color space\n", " hsv_image = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)\n", "\n", " # Create masks for brown and rose colors\n", " mask_brown = cv2.inRange(hsv_image, lower_brown, upper_brown)\n", " mask_rose = cv2.inRange(hsv_image, lower_rose, upper_rose)\n", "\n", " # Combine the masks\n", " combined_mask = cv2.bitwise_or(mask_brown, mask_rose)\n", "\n", " # apply brown and rose mask\n", " extracted_image = cv2.bitwise_and(frame, frame, mask=combined_mask)\n", "\n", " # blur the image\n", " blurred_image = cv2.GaussianBlur(extracted_image, (15, 15), 0)\n", "\n", " # Convert the bottom image to grayscale\n", " gray_image = cv2.cvtColor(blurred_image, cv2.COLOR_BGR2GRAY)\n", "\n", " # Compute Otsu's threshold\n", " otsu_thresh, _ = cv2.threshold(\n", " gray_image, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU\n", " )\n", "\n", " # Set lower and upper thresholds relative to Otsu's threshold\n", " lower = 0.5 * otsu_thresh\n", " upper = 1.5 * otsu_thresh\n", "\n", " # get edges\n", " edges = cv2.Canny(gray_image, lower, upper)\n", "\n", " return edges" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Get the lines from the edges" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [], "source": [ "def get_lines(edges):\n", " # Apply Probabilistic Hough Line Transform (allow to set minLineLength and maxLineGap)\n", " min_line_length = 50\n", " max_line_gap = 5\n", " lines_p = cv2.HoughLinesP(\n", " edges,\n", " 1,\n", " np.pi / 180,\n", " 100,\n", " minLineLength=min_line_length,\n", " maxLineGap=max_line_gap,\n", " )\n", "\n", " return lines_p" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Compute the slope of the line" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "# Function to calculate the angle of a line\n", "def calculate_angle(x1, y1, x2, y2):\n", " return math.degrees(math.atan2(y2 - y1, x2 - x1))" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Transform the line from 2 points to homogeneous coordinates" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "def cartesian_to_homogeneous(line):\n", " hom_line = np.cross([line[0], line[1], 1], [line[2], line[3], 1])\n", " hom_line = hom_line / hom_line[2]\n", "\n", " return hom_line" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Compute the intersection between the lines and the horizontal line" ] }, { "cell_type": "code", "execution_count": 14, "metadata": {}, "outputs": [], "source": [ "def compute_intersection(lines, horizontal_line):\n", " intersection_points_x = []\n", " # compute the intersection of each line with the horizontal line\n", " for line in lines:\n", " hom_line = cartesian_to_homogeneous(line[0])\n", " hom_hor_line = cartesian_to_homogeneous(horizontal_line[0])\n", "\n", " int_point = np.cross(hom_line, hom_hor_line)\n", " int_point = int_point / int_point[2]\n", "\n", " intersection_points_x.append(int_point[0])\n", "\n", " return intersection_points_x" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Select the closest left and right line" ] }, { "cell_type": "code", "execution_count": 15, "metadata": {}, "outputs": [], "source": [ "def select_closest_lines(lines, horizontal_line, center):\n", " # compute the intersection of the lines with the horizontal lines\n", " intersections_points_x = compute_intersection(lines, horizontal_line)\n", " left_lines = []\n", " right_lines = []\n", " left_distances = []\n", " right_distances = []\n", " for i in range(len(lines)):\n", " # if the intersection in at the left of the center\n", " if intersections_points_x[i] < center:\n", " left_lines.append(lines[i])\n", " left_distances.append(abs(center - intersections_points_x[i]))\n", "\n", " else: # if the intersection is at the right of the center\n", " right_lines.append(lines[i])\n", " right_distances.append(abs(center - intersections_points_x[i]))\n", "\n", " # compute the indeces of the minimum distance point\n", " min_left_index = (\n", " left_distances.index(min(left_distances)) if left_distances else None\n", " )\n", " min_right_index = (\n", " right_distances.index(min(right_distances)) if right_distances else None\n", " )\n", "\n", " # if exists, return the lines closest to the point\n", " if min_left_index is None:\n", " if min_right_index is None:\n", " return None, None\n", " return None, right_lines[min_right_index]\n", "\n", " if min_right_index is None:\n", " return left_lines[min_left_index], None\n", "\n", " return left_lines[min_left_index], right_lines[min_right_index]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Select the non horizontal lines that have both points over the horizontal line - divide left and rigth" ] }, { "cell_type": "code", "execution_count": 16, "metadata": {}, "outputs": [], "source": [ "def filter_lines(lines_p, horizontal_line, image_center, tolerance_angle=20):\n", " # Calculate the homogeneous coordinates of the horizontal line\n", " x1, y1, x2, y2 = horizontal_line[0]\n", " horizontal_line_homogeneous = np.cross([x1, y1, 1], [x2, y2, 1])\n", " horizontal_line_homogeneous = (\n", " horizontal_line_homogeneous / horizontal_line_homogeneous[0]\n", " )\n", "\n", " # Filter out lines that are 'quite horizontal' with a tolerance of 20 degrees\n", " filtered_lines = []\n", " if lines_p is not None:\n", " for line in lines_p:\n", " x1, y1, x2, y2 = line[0]\n", " angle = calculate_angle(x1, y1, x2, y2)\n", " if abs(angle) > tolerance_angle:\n", " y_max = max(y1, y2)\n", " x_max = x1 if y_max == y1 else x2\n", " # Filter the lines that have both endpoints over the horizontal line\n", " if (\n", " x_max\n", " + y_max * horizontal_line_homogeneous[1]\n", " + horizontal_line_homogeneous[2]\n", " > 0\n", " ): # se a*x + b*y + c > 0 allora il punto è sopra la linea\n", " filtered_lines.append(line)\n", "\n", " if len(filtered_lines) == 0:\n", " return None, None\n", "\n", " # divide the lines in left and right and select the closests\n", " left_line, right_line = select_closest_lines(\n", " filtered_lines, horizontal_line, image_center\n", " )\n", "\n", " return left_line, right_line" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Compute the lateral lines for a single frame" ] }, { "cell_type": "code", "execution_count": 17, "metadata": {}, "outputs": [], "source": [ "def compute_lines(frame, horizontal_line):\n", " # define the null line\n", " null_line = np.array([[0, 0, 0, 0]])\n", " # create an empty list\n", " all_lines = []\n", " # convert the lines in 2 points\n", " horizontal_line_carteisan = polar_to_cartesian(horizontal_line)\n", " # add the horizontal line\n", " all_lines.append(horizontal_line_carteisan[0])\n", " # define the central point on the horizontal line\n", " central_point = select_central_point(horizontal_line, frame)\n", " # compute the edges\n", " edges = get_edges(frame)\n", " # get the lateral lines\n", " lines_p = get_lines(edges)\n", "\n", " # select the closest left an right line\n", " left_line, right_line = filter_lines(\n", " lines_p, horizontal_line_carteisan, central_point[0]\n", " )\n", "\n", " # add the lines to the list\n", " if left_line is None:\n", " all_lines.append(null_line[0])\n", " else:\n", " all_lines.append(left_line[0])\n", "\n", " if right_line is None:\n", " all_lines.append(null_line[0])\n", " else:\n", " all_lines.append(right_line[0])\n", "\n", " return all_lines" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Main" ] }, { "cell_type": "code", "execution_count": 18, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "227\n", "b: 0.9999352459591057\n", "b: 0.9999099545922074\n", "b: 0.9998804512545729\n", "b: 0.9998568255432034\n", "b: 0.9998527664601821\n", "b: 0.9998557691252282\n", "b: 0.9998683611476573\n", "b: 0.9998814976014934\n", "b: 0.9998895606632995\n", "b: 0.9998901556943395\n", "b: 0.9998845641079749\n", "b: 0.9998857123452798\n", "b: 0.9998806225972304\n", "b: 0.9998867611329645\n", "b: 0.9998879503607253\n", "b: 0.9998952363060616\n", "b: 0.9998991135387997\n", "b: 0.9999086076991831\n", "b: 0.9999200332461285\n", "b: 0.9999260946342385\n", "b: 0.9999273358715206\n", "b: 0.999921528944179\n", "b: 0.9999299333408088\n", "b: 0.9999251199349254\n", "b: 0.9999201574878317\n", "b: 0.9999150491064517\n", "b: 0.9999147820782102\n", "b: 0.9998983765995302\n", "b: 0.9998862050383474\n", "b: 0.9998794609305839\n", "b: 0.9998872536241195\n", "b: 0.9998950506171649\n", "b: 0.999870874890744\n", "b: 0.999813672975844\n", "b: 0.9997454637462588\n", "b: 0.9996768703977311\n", "b: 0.9995996053199921\n", "b: 0.9995793589761323\n", "b: 0.9996055846700124\n", "b: 0.9996419424845241\n", "b: 0.9996399516907685\n", "b: 0.9996242923879346\n", "b: 0.9996079467077668\n", "b: 0.9996037570882312\n", "b: 0.9995876527024623\n", "b: 0.9995831399110006\n", "b: 0.9996011225574956\n", "b: 0.9996207642419855\n", "b: 0.9996277534883864\n", "b: 0.9996277928764999\n", "b: 0.9995993475450655\n", "b: 0.9995615318853331\n", "b: 0.9995198918049472\n", "b: 0.9994788635416031\n", "b: 0.999443094468464\n", "b: 0.9994394577584143\n", "b: 0.9994339733919415\n", "b: 0.9994620863988346\n", "b: 0.9995211737036884\n", "b: 0.9995791326005927\n", "b: 0.9996354358447255\n", "b: 0.9996895537251627\n", "b: 0.9997166013184763\n", "b: 0.9997192915753913\n", "b: 0.9997212992231203\n", "b: 0.9996724005506846\n", "b: 0.9996186396544936\n", "b: 0.9996616821455988\n", "b: 0.9997019917386248\n", "b: 0.9996560497144404\n", "b: 0.9996619742831359\n", "b: 0.999748749250127\n", "b: 0.9997532162537726\n", "b: 0.999757615316332\n", "b: 0.9998294989581157\n", "b: 0.9998593853759654\n", "b: 0.9998535170243139\n", "b: 0.9998646097065224\n", "b: 0.9998674472116144\n", "b: 0.9998715785661836\n", "b: 0.9999014478180217\n", "b: 0.9998956405185663\n", "b: 0.9998690203715975\n", "b: 0.9998506225362515\n", "b: 0.9998206515487136\n", "b: 0.9997793286600268\n", "b: 0.9997909111328099\n", "b: 0.999800187805\n", "b: 0.9998108529750774\n", "b: 0.9998257871031965\n", "b: 0.99985956187491\n", "b: 0.9998661117381741\n", "b: 0.9998760794258541\n", "b: 0.9998802401987993\n", "b: 0.9998876287708831\n", "b: 0.9998726059346832\n", "b: 0.9998779638768238\n", "b: 0.9998831145253774\n", "b: 0.999889908687669\n", "b: 0.9998983180427963\n", "b: 0.99990658332165\n", "b: 0.9999083805278187\n", "b: 0.999902246989958\n", "b: 0.9999007192076472\n", "b: 0.9998950277045546\n", "b: 0.9999026401653248\n", "b: 0.9999025787849267\n", "b: 0.9999103797759475\n", "b: 0.9999014640420433\n", "b: 0.9998944469628173\n", "b: 0.9998759814604811\n", "b: 0.9998520196215154\n", "b: 0.9998259347681495\n", "b: 0.9998432913847581\n", "b: 0.9998597135993074\n", "b: 0.999874154576434\n", "b: 0.9998876929206812\n", "b: 0.999891975271201\n", "b: 0.999868257934949\n", "b: 0.9998475113032204\n", "b: 0.999831493818174\n", "b: 0.9998332300104823\n", "b: 0.999845479127809\n", "b: 0.9998644832660794\n", "b: 0.9998775212001805\n", "b: 0.9998919169402337\n", "b: 0.9998891472984569\n", "b: 0.9998830644190126\n", "b: 0.9998747382066531\n", "b: 0.9998661074487519\n", "b: 0.999859861161036\n", "b: 0.9998561744490447\n", "b: 0.9998561215574864\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "b: 0.9998560951007662\n", "Adjusted video saved to C:\\Users\\miche\\OneDrive\\Documenti\\GitHub\\bowling-analysis\\data\\recording_3\\Lines_video_non_processed.mp4\n" ] } ], "source": [ "# Reset the video to the beginning\n", "cap.set(cv2.CAP_PROP_POS_FRAMES, 0)\n", "\n", "# Define the codec and create a VideoWriter object to save the modified frames\n", "output_path = (\n", " project_root\n", " / \"data\"\n", " / f\"recording_{video_number}\"\n", " / \"Lines_video_non_processed.mp4\"\n", ")\n", "fourcc = cv2.VideoWriter_fourcc(*\"mp4v\") # Use 'mp4v' codec for MP4 format\n", "fps = int(cap.get(cv2.CAP_PROP_FPS))\n", "frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))\n", "frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))\n", "out = cv2.VideoWriter(str(output_path), fourcc, fps, (frame_width, frame_height))\n", "\n", "# Loop through each frame in the video\n", "frame_index = 0\n", "lines_array = []\n", "print(len(horizontal_lines))\n", "while frame_index < len(horizontal_lines):\n", " ret, video_frame = cap.read()\n", " if not ret:\n", " print(\"End of video or failed to read the frame at iteration\", frame_index)\n", " break\n", "\n", " # Compute the three lines in the frame\n", " all_lines = compute_lines(video_frame, horizontal_lines.iloc[frame_index])\n", "\n", " if all_lines is None:\n", " print(\"No lines found in frame\", frame_index)\n", " modified_frame = video_frame\n", "\n", " else:\n", " # draw the lines on the frame\n", " modified_frame = write_lines_on_frame(video_frame, all_lines)\n", " # add lines to the output array\n", " lines_array.append(all_lines)\n", "\n", " # Write the modified frame to the output video\n", " out.write(modified_frame)\n", "\n", " # Increment the frame index\n", " frame_index += 1\n", "\n", "# Release the video capture and writer objects\n", "# cap.release()\n", "out.release()\n", "\n", "print(f\"Adjusted video saved to {output_path}\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Save lines in csv file" ] }, { "cell_type": "code", "execution_count": 19, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Lines array saved to C:\\Users\\miche\\OneDrive\\Documenti\\GitHub\\bowling-analysis\\notebook\\lane_detection\\intermediate_data\\lane_lines\\three_lines_3.csv\n" ] } ], "source": [ "# Define the output path for the lines CSV file\n", "output_lines_path = (\n", " project_root\n", " / \"notebook\"\n", " / \"lane_detection\"\n", " / \"intermediate_data\"\n", " / \"lane_lines\"\n", " / f\"three_lines_{video_number}.csv\"\n", ")\n", "\n", "# Convert lines_array to a DataFrame\n", "lines_df = pd.DataFrame(lines_array)\n", "\n", "# Save the DataFrame to a CSV file\n", "lines_df.to_csv(output_lines_path, index=False)\n", "\n", "print(f\"Lines array saved to {output_lines_path}\")" ] } ], "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": "3.11.9" } }, "nbformat": 4, "nbformat_minor": 2 }