#!/usr/bin/env python3 """ Visualize and manually decode printer tracking dots Helps identify the 15x8 grid pattern """ from PIL import Image, ImageDraw, ImageFont import numpy as np import os def visualize_dot_pattern(image_path="yd_Zard_uncompressed_dots.png"): """Create a visualization to help manually decode the pattern""" print(f"Analyzing {image_path}...") # Load the dots image img = Image.open(image_path) img_array = np.array(img) print(f"Image size: {img_array.shape}") # Convert to grayscale if needed if len(img_array.shape) == 3: gray = np.mean(img_array, axis=2) else: gray = img_array # Find potential dots (bright pixels) threshold = np.percentile(gray, 99.5) dots_mask = gray > threshold # Get dot coordinates y_coords, x_coords = np.where(dots_mask) print(f"Found {len(x_coords)} potential dot pixels") if len(x_coords) == 0: print("No dots found! Trying lower threshold...") threshold = np.percentile(gray, 95) dots_mask = gray > threshold y_coords, x_coords = np.where(dots_mask) print(f"Found {len(x_coords)} potential dots at 95th percentile") if len(x_coords) > 0: # Try to find the grid pattern # Printer dots are typically arranged in a regular 15x8 grid # Focus on a small region to find the pattern crop_size = 500 crop = gray[:crop_size, :crop_size] crop_threshold = np.percentile(crop, 98) crop_dots = crop > crop_threshold crop_y, crop_x = np.where(crop_dots) print(f"\nIn top-left {crop_size}x{crop_size} region: {len(crop_x)} dots") if len(crop_x) > 10: # Sort and find spacing sorted_x = np.sort(crop_x) sorted_y = np.sort(crop_y) # Find differences x_diffs = np.diff(sorted_x) y_diffs = np.diff(sorted_y) # Filter reasonable spacing (5-200 pixels) x_valid = x_diffs[(x_diffs > 5) & (x_diffs < 200)] y_valid = y_diffs[(y_diffs > 5) & (y_diffs < 200)] if len(x_valid) > 0: # Find most common spacing using histogram hist, bins = np.histogram(x_valid, bins=30) most_common_x_spacing = bins[np.argmax(hist)] print(f"Estimated X spacing: {most_common_x_spacing:.1f} pixels") if len(y_valid) > 0: hist, bins = np.histogram(y_valid, bins=30) most_common_y_spacing = bins[np.argmax(hist)] print(f"Estimated Y spacing: {most_common_y_spacing:.1f} pixels") # Create enhanced visualization viz_scale = 5 viz = Image.fromarray(crop).convert('RGB') viz = viz.resize((crop_size * viz_scale, crop_size * viz_scale), Image.NEAREST) draw = ImageDraw.Draw(viz) # Mark all dots in red for x, y in zip(crop_x, crop_y): x_scaled = x * viz_scale y_scaled = y * viz_scale draw.ellipse([x_scaled-8, y_scaled-8, x_scaled+8, y_scaled+8], fill='red', outline='yellow', width=2) viz.save("dot_pattern_visualization.png") print(f"\nSaved visualization to: dot_pattern_visualization.png") print("Look for a 15x8 grid pattern of red dots") # Try to extract a 15x8 grid if we found spacing if len(x_valid) > 0 and len(y_valid) > 0: print("\n" + "="*70) print("Attempting to extract 15x8 grid...") print("="*70) # Use median spacing x_spacing = np.median(x_valid) y_spacing = np.median(y_valid) # Find the top-left corner (minimum x and y with dots) start_x = crop_x.min() start_y = crop_y.min() print(f"Grid starting at: ({start_x}, {start_y})") print(f"Spacing: X={x_spacing:.1f}, Y={y_spacing:.1f}") # Try to extract a 15x8 grid grid = np.zeros((8, 15), dtype=int) for row in range(8): for col in range(15): expected_x = start_x + col * x_spacing expected_y = start_y + row * y_spacing # Check if there's a dot near this position tolerance = x_spacing / 3 nearby = ((crop_x > expected_x - tolerance) & (crop_x < expected_x + tolerance) & (crop_y > expected_y - tolerance) & (crop_y < expected_y + tolerance)) if np.any(nearby): grid[row, col] = 1 print("\nExtracted grid (1=dot, 0=no dot):") print("Rows (top to bottom), Columns (left to right)") print(" " + "".join(f"{i:2}" for i in range(1, 16))) for i, row in enumerate(grid): print(f"R{i}: " + "".join(f" {val}" for val in row)) # Decode the pattern print("\n" + "="*70) print("DECODING:") print("="*70) # Each column encodes a number in binary (ignoring first row) decoded = [] for col_num in range(1, 15): # Columns 1-14 (0-indexed: 0-13) col = grid[1:, col_num] # Skip first row (row 0) # Read as binary (top to bottom = MSB to LSB) value = sum(bit * (2 ** (6 - idx)) for idx, bit in enumerate(col)) decoded.append(value) # Identify what this column represents description = "" if col_num == 1: description = f"Minutes: {value:02d}" elif col_num == 4: description = f"Hour: {value:02d}" elif col_num == 5: description = f"Day: {value:02d}" elif col_num == 6: description = f"Month: {value:02d}" elif col_num == 7: description = f"Year: 20{value:02d}" elif 10 <= col_num <= 13: description = f"Serial digit {col_num-9}: {value}" if description: print(f"Column {col_num:2d}: {value:3d} ({value:07b}) - {description}") # Format the result if decoded[6] > 0: # Column 7 (year) should have a value year = f"20{decoded[6]:02d}" month = f"{decoded[5]:02d}" day = f"{decoded[4]:02d}" hour = f"{decoded[3]:02d}" minutes = f"{decoded[0]:02d}" serial = f"{decoded[9]}{decoded[10]}{decoded[11]}{decoded[12]}" print("\n" + "="*70) print("FINAL RESULT:") print("="*70) print(f"Date/Time: {year}_{month}_{day}_{hour}:{minutes}") print(f"Serial: {serial}") print(f"\nFlag format: uoftctf{{{year}_{month}_{day}_{hour}:{minutes}_{serial}}}") print("="*70) if __name__ == "__main__": import sys # Try the dots image first if os.path.exists("yd_Zard_uncompressed_dots.png"): visualize_dot_pattern("yd_Zard_uncompressed_dots.png") else: print("yd_Zard_uncompressed_dots.png not found!") print("Run: deda_extract_yd Zard_uncompressed.png -d 300 -e --debug")