Update app.py

app.py

@@ -6,25 +6,25 @@ from openai import OpenAI
 
 # Image Preprocessing Function
 def preprocess_image(image_path):
-    # Read the image in grayscale
-    image = cv2.imread(image_path, cv2.IMREAD_GRAYSCALE)
-    
-    # Resize the image if it is too large or too small; maintain the aspect ratio
-    # Skip resizing if the image is already at a desirable size
-    
-    # Apply a mild Gaussian blur to reduce noise which can help in edge detection
-    # Use a small kernel size to ensure details are not lost
-    blurred_image = cv2.GaussianBlur(image, (3, 3), 0)
-    
-    # Apply adaptive thresholding to create a binary image
-    # This will help in making the text/numbers crisp
-    crisp_image = cv2.adaptiveThreshold(
-        blurred_image, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
-        cv2.THRESH_BINARY, 11, 2
-    )
+    # Convert the file to an OpenCV image
+    file_bytes = np.asarray(bytearray(image_file.read()), dtype=np.uint8)
+    image = cv2.imdecode(file_bytes, cv2.IMREAD_COLOR)
+
+    # Resize the image, if necessary
+    # image = cv2.resize(image, (desired_width, desired_height))
+
+    # Convert to grayscale
+    gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+
+    # Apply Gaussian blur to reduce noise and improve OCR accuracy
+    blurred_image = cv2.GaussianBlur(gray_image, (5, 5), 0)
+
+    # Adaptive Histogram Equalization
+    clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8))
+    equalized_image = clahe.apply(blurred_image)
     
     # Return the preprocessed image
-    return crisp_image
+    return image_path
 
 # Function to encode the image to base64
 def encode_image(image_file):