remnant_2_video_analyzer.py

# before running:
#
# 1. Install dependencies:
#   cd remnant-2-video-analyzer
#   pip install opencv-python pytesseract
# 2. Install Tesseract (https://github.com/tesseract-ocr/tesseract)
# 3. Run python script:
#   python main.py

import cv2
import pytesseract
import sys

# Path to the Tesseract executable (modify if necessary)
pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'  # Adjust for your installation

bosses = [
    # bosses
    'Abomination',
    'Bloat King',
    'Bruin, Blade of the King',
    'Cancer',
    'Cinderclad Forge',
    'Gwendil: The Unburnt',
    'Kaeula\'s Shadow',
    'Legion',
    'Magister Dullain',
    'Mother Mind',
    'N\'Erudian Reaper',
    'Primogenitor',
    'Shrewd',
    'Amalgam Kel\'Roth',
    'The Astropath',
    'Custodian\'s Eye',
    'The Huntress',
    'The Red Prince',
    'The Stonewarden',
    'Sunken Witch',
    'Venom',
    # world bosses
    'Alepsis-Taura',
    'Annihilation',
    'Corrupted Ravager',
    'Corruptor',
    'Imposter King',
    'Labyrinth Sentinel',
    'Lydusa',
    'Sha\'Hala: Spectral Guardian of N\'Erud',
    'Sha\'Hala: Guardian of N\'Erud',
    'Tal\'Ratha',
    'Tal\'Ratha (Metaphysical)',
    'The Nightweaver',
    'The One True King',
    # aberrations
    'Arcanum Diviner',
    'Barghest the Vile',
    'Blightspawn',
    'Firth: The Oathkeeper',
    'Gorge',
    'Grimshot',
    'Highborn Stalker',
    'Maleficent Glimmer',
    'Mournshot',
    'Ripsaw',
    'Scorchslinger',
    'Seer of the Veil',
    'The Executioner',
    'Tortured Flame',
    'Vilethorn Brawler',
    'Vilethorn Snapper',
    'Fetid Corpse',
    'Restless Spirit',
    'W.D. 109',
    'The Progeny',
    'E.D. Alpha',
    'Defiler',
    'Rot',
    'Wither',
    'Mantagora',
    'The Weald Stalker',
    'The Gnarled Archer',
    'Fester',
    'Atrophy',
    'Rot Stalker',
    'Plaguebringer',
    'Talonscythe',
    'Bastion',
    'Bane'
]

class Result:
    def __init__(self, boss_name, start_time):
        self.boss_name = boss_name
        self.start_time = start_time

def detect_text_in_video(video_path):
    results = []

    # Open the video file
    cap = cv2.VideoCapture(video_path)
    total_frames = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))

    if not cap.isOpened():
        print("Error: Could not open video.")
        return

    frame_number = 0

    while True:
        # Read a frame from the video
        cap.set(cv2.CAP_PROP_POS_FRAMES, frame_number)
        ret, frame = cap.read()
        
        if not ret:
            break  # Exit if no frames are left

        x, y, width, height = 1100, 0, 1650, 250
        roi = frame[y:y + height, x:x + width]

        frame_number += 60
        
        # Convert the frame to gray scale (optional, but can help with OCR)
        gray_frame = cv2.cvtColor(roi, cv2.COLOR_BGR2GRAY)

        # Use Tesseract to do OCR on the frame
        text = pytesseract.image_to_string(gray_frame)

        # Check for the boss name in the frame
        for boss in bosses:
            if boss in text:
                result_bosses = [result.boss_name for result in results]
                if boss not in result_bosses:
                    seconds = int(frame_number / 60)
                    time = seconds_to_hh_mm_ss(seconds)
                    results.append(Result(boss, time))

        progress_bar(frame_number, total_frames)

    # Release the video capture object
    cap.release()
    cv2.destroyAllWindows()

    print()
    print()
    for result in results:
        formatted_string = "{:<30} | {:<10}".format(result.boss_name, result.time)
        print(formatted_string)

def progress_bar(iteration, total, length = 50):
    percent = iteration / total
    percent = 1 if percent > 1 else percent

    progress = int(length * iteration / total)
    bar = '=' * progress + '-' * (length - progress)
    percent_progress = 100 * iteration / total

    sys.stdout.write(f'\r[{bar}] {percent_progress:.2f}%')
    sys.stdout.flush()

def seconds_to_hh_mm_ss(total_seconds):
    hours = total_seconds // 3600
    minutes = (total_seconds % 3600) // 60
    seconds = total_seconds % 60
    return f"{hours:02}:{minutes:02}:{seconds:02}"

# Example usage - looking for a file alongside the script
video_file = 'ivha.mp4'
detect_text_in_video(video_file)