Myosotis-Researches 0.1.35__tar.gz → 0.1.37__tar.gz

{myosotis_researches-0.1.35 → myosotis_researches-0.1.37/Myosotis_Researches.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: Myosotis-Researches
-Version: 0.1.35
+Version: 0.1.37
 Summary: A repository for storing my progress of researches.
 Home-page: https://github.com/Zeyu-Xie/Myosotis-Researches
 Author: Zeyu Xie

{myosotis_researches-0.1.35/Myosotis_Researches.egg-info → myosotis_researches-0.1.37}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: Myosotis-Researches
-Version: 0.1.35
+Version: 0.1.37
 Summary: A repository for storing my progress of researches.
 Home-page: https://github.com/Zeyu-Xie/Myosotis-Researches
 Author: Zeyu Xie

{myosotis_researches-0.1.35 → myosotis_researches-0.1.37}/myosotis_researches/CcGAN/utils/similarity.py

@@ -59,19 +59,25 @@ def PSNR(images):
             PSNR_sum += _PSNR(images[i], images[j])
     return PSNR_sum / (n * (n - 1) / 2)
 
+
 # Cross-correlation
 
+
 def _cross_correlation(image_1, image_2):
-    array_1 = np.array(image_1)
-    array_2 = np.array(image_2)
-    channels_1 = cv2.split(array_1)
-    channels_2 = cv2.split(array_2)
-    scores = []
-    for c1, c2 in zip(channels_1, channels_2):
-        result = cv2.matchTemplate(c1, c2, method=cv2.TM_CCOEFF_NORMED)
-        _, max_val, _, _ = cv2.minMaxLoc(result)
-        scores.append(max_val)
-    return sum(scores) / len(scores)
+    array_1 = np.array(image_1).astype(np.float32) / 255
+    array_2 = np.array(image_2).astype(np.float32) / 255
+    correlations = []
+    for c in range(3):
+        channel1 = array_1[..., c]
+        channel2 = array_2[..., c]
+        channel1 -= channel1.mean()
+        channel2 -= channel2.mean()
+        numerator = np.sum(channel1 * channel2)
+        denominator = np.sqrt(np.sum(channel1**2) * np.sum(channel2**2))
+        corr = numerator / denominator if denominator != 0 else 0
+        correlations.append(corr)
+    return np.mean(correlations)
+
 
 def cross_correlation(images):
     cross_correlation_sum = 0
@@ -81,8 +87,43 @@ def cross_correlation(images):
             cross_correlation_sum += _cross_correlation(images[i], images[j])
     return cross_correlation_sum / (n * (n - 1) / 2)
 
+
+# Histogram intersection
+def _rgb_histogram(image, bins=32):
+    """
+    image: numpy array of shape (H, W, 3), RGB format
+    returns: concatenated histogram for R, G, B channels
+    """
+    hist_r, _ = np.histogram(image[:, :, 0], bins=bins, range=(0, 255), density=True)
+    hist_g, _ = np.histogram(image[:, :, 1], bins=bins, range=(0, 255), density=True)
+    hist_b, _ = np.histogram(image[:, :, 2], bins=bins, range=(0, 255), density=True)
+    hist = np.concatenate([hist_r, hist_g, hist_b])
+    return hist
+
+
+def _histogram_intersection(image_1, image_2):
+    hist_1 = _rgb_histogram(image_1)
+    hist_2 = _rgb_histogram(image_2)
+    return np.sum(np.minimum(hist_1, hist_2))
+
+
+def histogram_intersection(images):
+    histogram_intersection_sum = 0
+    n = len(images)
+    for i in range(n):
+        for j in range(i + 1, n):
+            histogram_intersection_sum += _histogram_intersection(images[i], images[j])
+    return histogram_intersection_sum / (n * (n - 1) / 2)
+
+
 # Similarity
-_method_dict = {"MSE": MSE, "SSIM": SSIM, "PSNR": PSNR, "cross-correlation": cross_correlation}
+_method_dict = {
+    "MSE": MSE,
+    "SSIM": SSIM,
+    "PSNR": PSNR,
+    "cross-correlation": cross_correlation,
+    "histogram-intersection": histogram_intersection,
+}
 
 
 def similarity(images, method):

{myosotis_researches-0.1.35 → myosotis_researches-0.1.37}/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
 
 setup(
     name="Myosotis-Researches",
-    version="0.1.35",
+    version="0.1.37",
     description="A repository for storing my progress of researches.",
     long_description=open("README.md").read(),
     long_description_content_type="text/markdown",