X or O Detector

📁 Upload Image:

✏️ Draw Here:

Result will appear here...

🧠 How the Detection Algorithm Works

🔍 X Detection Strategy:

• Scans diagonal lines from top-left to bottom-right and top-right to bottom-left
• Checks for crossing patterns in the center area (where diagonals intersect)
• Weights center crossings more heavily as they're characteristic of X shapes

⭕ O Detection Strategy:

• Samples pixels in circular patterns at multiple radii (15, 20, 25, 30 pixels)
• Checks for hollow center - a key characteristic of O shapes
• Awards bonus points when center area is mostly empty (60%+ hollow)

⚖️ Decision Making Process:

• Calculates separate scores for X and O patterns
• Requires 20% score difference for high confidence classification
• Uses fallback heuristics when scores are close (hollow center, total dark pixels)
• Provides confidence levels: High, Medium, or Low based on score differences

💡 Note: This algorithm is completely deterministic and uses geometric pattern recognition without any AI or machine learning.

🐍 Python Implementation

import numpy as np
from PIL import Image
import math

def analyze_image(image_path):
    # Load and convert image to grayscale
    img = Image.open(image_path).convert('L')
    img = img.resize((100, 100))  # Resize to 100x100
    data = np.array(img)
    
    w, h = data.shape
    center_x, center_y = w // 2, h // 2
    
    # Enhanced diagonal pattern detection (X pattern)
    diag_score = 0
    for i in range(15, 85, 3):
        # Top-left to bottom-right diagonal
        x1 = int((i / 100) * w)
        y1 = int((i / 100) * h)
        
        # Top-right to bottom-left diagonal
        x2 = int(((100 - i) / 100) * w)
        y2 = int((i / 100) * h)
        
        if is_dark(data[y1, x1]):
            diag_score += 1
        if is_dark(data[y2, x2]):
            diag_score += 1
    
    # Check for crossing pattern in center area
    center_crossing = 0
    for offset in range(-8, 9, 2):
        x1, y1 = center_x + offset, center_y + offset
        x2, y2 = center_x - offset, center_y + offset
        
        if 0 <= x1 < w and 0 <= y1 < h and is_dark(data[y1, x1]):
            center_crossing += 1
        if 0 <= x2 < w and 0 <= y2 < h and is_dark(data[y2, x2]):
            center_crossing += 1
    
    # Enhanced circular pattern detection (O pattern)
    circle_score = 0
    radiuses = [15, 20, 25, 30]
    
    for radius in radiuses:
        for angle in range(0, 360, 15):
            x = int(center_x + radius * math.cos(math.radians(angle)))
            y = int(center_y + radius * math.sin(math.radians(angle)))
            
            if 0 <= x < w and 0 <= y < h and is_dark(data[y, x]):
                circle_score += 1
    
    # Check for hollow center (characteristic of O)
    center_hollow = 0
    center_total = 0
    for x in range(center_x - 10, center_x + 11, 2):
        for y in range(center_y - 10, center_y + 11, 2):
            if 0 <= x < w and 0 <= y < h:
                center_total += 1
                if not is_dark(data[y, x]):
                    center_hollow += 1
    
    # Enhanced scoring system
    x_score = diag_score + center_crossing * 2
    o_score = circle_score + (20 if center_hollow > center_total * 0.6 else 0)
    
    # Decision making
    if x_score > o_score * 1.2:
        return "X", "High"
    elif o_score > x_score * 1.2:
        return "O", "High"
    else:
        return ("X" if x_score > o_score else "O"), "Low"

def is_dark(pixel_value):
    return pixel_value < 100  # Threshold for dark pixels

🎯 X or O Detector

🧠 How the Detection Algorithm Works

🔍 X Detection Strategy:

⭕ O Detection Strategy:

⚖️ Decision Making Process:

🐍 Python Implementation