Auto-sort images by dominant color

Auto-sorting images by dominant color involves analyzing each image to determine its primary color and then organizing the images based on those color values. Here’s a step-by-step outline of how to do this programmatically using Python:

Requirements

Install the necessary libraries:

bash
CopyEdit
pip install opencv-python pillow numpy scikit-learn

Python Script to Auto-Sort Images by Dominant Color

python
CopyEdit
import os
import shutil
import numpy as np
from PIL import Image
from sklearn.cluster import KMeans

# Configuration
input_folder = 'images'  # Folder with unsorted images
output_folder = 'sorted_images'  # Destination folder

def get_dominant_color(image_path, k=4, resize=150):
    image = Image.open(image_path)
    image = image.resize((resize, resize))  # Resize for faster processing
    data = np.array(image).reshape(-1, 3)  # Flatten image to RGB pixels

    kmeans = KMeans(n_clusters=k)
    kmeans.fit(data)
    colors = kmeans.cluster_centers_
    labels = kmeans.labels_
    counts = np.bincount(labels)
    
    dominant = colors[np.argmax(counts)]
    return tuple(map(int, dominant))

def color_to_hex(color):
    return "#{:02x}{:02x}{:02x}".format(*color)

def sort_images_by_color(input_dir, output_dir):
    if not os.path.exists(output_dir):
        os.makedirs(output_dir)

    for filename in os.listdir(input_dir):
        if filename.lower().endswith(('.jpg', '.jpeg', '.png', '.bmp', '.gif')):
            file_path = os.path.join(input_dir, filename)
            try:
                dominant_color = get_dominant_color(file_path)
                color_hex = color_to_hex(dominant_color)
                color_folder = os.path.join(output_dir, color_hex)

                if not os.path.exists(color_folder):
                    os.makedirs(color_folder)

                shutil.copy(file_path, os.path.join(color_folder, filename))
                print(f"Sorted {filename} -> {color_hex}")
            except Exception as e:
                print(f"Failed to process {filename}: {e}")

# Run
sort_images_by_color(input_folder, output_folder)

What This Script Does

• Analyzes each image for dominant color using KMeans clustering.

• Converts that color to a hexadecimal string.

• Creates folders named by color (like #aabbcc) and copies the images there.

Tips

• You can group colors further by rounding or mapping to standard color names.

• Add cv2 for more preprocessing (e.g., color space conversion to HSV if needed).

• If working with a large dataset, consider optimizing the resizing and clustering for speed.

Let me know if you want this adapted into a GUI app or CLI tool.