Added Histogram Equalization Layer

src/vision_stack/__init__.py

@@ -14,3 +14,4 @@
 from .layers.RGBtoBGRLayer import RGBtoBGRLayer
 from .layers.SobelLayer import SobelLayer
 from .layers.UnderwaterEnhancementLayer import UnderWaterImageEnhancementLayer
+from .layers.HistogramEqualizationLayer import HistogramEqualizationLayer

src/vision_stack/layers/HistogramEqualizationLayer.py

@@ -0,0 +1,34 @@
+import cv2
+
+from .Layer import PreprocessLayer
+
+
+class HistogramEqualizationLayer(PreprocessLayer):
+
+    def __init__(self, grid_size=8) -> None:
+        """
+        Passes an adaptive cv2 histogram equalization over the image.
+
+        Paramters:
+            gridsize: Size of grid for histogram equalization. Input image will be divided into 
+            equally sized rectangular tiles. gridsize defines the number of tiles in row and column.
+        """
+        super().__init__("histogramequalization")
+        self.grid_size = grid_size
+
+    def process(self, image):
+        if len(image.shape) == 2: #Basic Histogram Equalization for grayscale image
+            clahe = cv2.createCLAHE(clipLimit=2, tileGridSize=(self.grid_size,self.grid_size))
+            equalized_image = clahe.apply(image)
+            return (equalized_image, None)
+        else: # Histogram Equalization for colored images
+            lab_image = cv2.cvtColor(image, cv2.COLOR_BGR2LAB)
+            l, a, b = cv2.split(lab_image)
+
+            clahe = cv2.createCLAHE(clipLimit=2, tileGridSize=(self.grid_size,self.grid_size))
+            equalized_l = clahe.apply(l)
+
+            equalized_lab_image = cv2.merge((equalized_l, a, b))
+            equalized_image = cv2.cvtColor(equalized_lab_image, cv2.COLOR_LAB2BGR)
+
+            return(equalized_image, None)