creek

  1"""
  2Color
  3Original file: https://colab.research.google.com/drive/1o7cMSAesuuaxFQsteQL4ViwL_qNeTWbm
  4"""
  5
  6import cv2
  7import numpy as np
  8import matplotlib.pyplot as plt
  9import matplotlib.colors as mcolors
 10from sklearn.cluster import KMeans
 11from google.colab import files
 12from PIL import Image
 13import io
 14
 15
 16def get_dominant_colors(image, k=5):
 17    """
 18    Extracts dominant colors using KMeans.
 19    Expects a PIL Image or Numpy array (RGB).
 20    """
 21    if isinstance(image, Image.Image):
 22        image = np.array(image)
 23
 24    img_small = cv2.resize(image, (150, 150), interpolation=cv2.INTER_AREA)
 25
 26    pixels = img_small.reshape((-1, 3))
 27
 28    kmeans = KMeans(n_clusters=k, n_init="auto", random_state=42)
 29    kmeans.fit(pixels)
 30
 31    colors = kmeans.cluster_centers_.astype(int)
 32
 33    return sorted(colors.tolist(), key=lambda x: sum(x))
 34
 35
 36def classify_mood(rgb_colors):
 37    """
 38    Classifies mood based on HSV values using Matplotlib colors.
 39    """
 40    norm_colors = np.array(rgb_colors) / 255.0
 41
 42    hsv_stats = mcolors.rgb_to_hsv(norm_colors)
 43
 44    avg_sat = np.mean(hsv_stats[:, 1])
 45    avg_val = np.mean(hsv_stats[:, 2])
 46
 47    if avg_sat < 0.15 and avg_val > 0.65:
 48        return "Minimalist"
 49    if avg_val < 0.35:
 50        return "Dark/Moody"
 51    if avg_sat < 0.45 and avg_val > 0.75:
 52        return "Pastel"
 53    if avg_sat > 0.65 and avg_val > 0.5:
 54        return "Neon"
 55
 56    earthy_votes = sum(1 for h, s, v in hsv_stats if (0.02 <= h <= 0.42) and s < 0.8)
 57    if earthy_votes >= 3:
 58        return "Earthy"
 59
 60    warm_votes = sum(1 for h, s, v in hsv_stats if h < 0.17 or h > 0.83)
 61    return "Warm" if warm_votes >= 3 else "Cool"
 62
 63
 64def rgb_to_hex(rgb):
 65    return "#{:02x}{:02x}{:02x}".format(rgb[0], rgb[1], rgb[2])
 66
 67
 68# Runner
 69
 70
 71def analyze_uploaded_image():
 72    print("Please upload an image file...")
 73    uploaded = files.upload()
 74
 75    for fn in uploaded.keys():
 76        print(f"\nProcessing {fn}...")
 77
 78        image_data = uploaded[fn]
 79        image = Image.open(io.BytesIO(image_data)).convert("RGB")
 80
 81        colors = get_dominant_colors(image)
 82        if colors is None:
 83            print("Error processing image.")
 84            continue
 85
 86        mood = classify_mood(colors)
 87        hex_colors = [rgb_to_hex(c) for c in colors]
 88
 89        print(f"Detected Mood: {mood}")
 90        print("Extracted Palette:")
 91        print("-" * 30)
 92        for i, h in enumerate(hex_colors):
 93            print(f"Color {i+1}: {h}")
 94        print("-" * 30)
 95
 96        fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(10, 4))
 97
 98        ax1.imshow(image)
 99        ax1.axis("off")
100        ax1.set_title(f"Mood: {mood}", fontweight="bold")
101
102        y_pos = np.arange(len(colors))
103        ax2.barh(y_pos, [1] * len(colors), color=hex_colors)
104        ax2.invert_yaxis()
105        ax2.axis("off")
106
107        for i, h in enumerate(hex_colors):
108            brightness = sum(colors[i]) / 3
109            text_col = "black" if brightness > 128 else "white"
110            ax2.text(
111                0.5,
112                i,
113                f"{h}",
114                ha="center",
115                va="center",
116                color=text_col,
117                fontweight="bold",
118                fontsize=12,
119            )
120
121        plt.tight_layout()
122        plt.show()
123
124
125if __name__ == "__main__":
126    analyze_uploaded_image()