refactoring

vidocp/figure_detection.py

@@ -2,8 +2,15 @@ import cv2
 import numpy as np
 from pdf2image import pdf2image
 
-from vidocp.utils import show_mpl, draw_rectangles, remove_included, remove_primary_text_regions, \
-    __detect_large_coherent_structures, is_large_enough, has_acceptable_format
+from vidocp.utils import (
+    show_mpl,
+    draw_rectangles,
+    remove_included,
+    remove_primary_text_regions,
+    __detect_large_coherent_structures,
+    is_large_enough,
+    has_acceptable_format,
+)
 
 
 def is_likely_figure(cont, min_area=5000, max_width_to_hight_ratio=6):

vidocp/utils.py

@@ -144,8 +144,8 @@ def vec_rect_to_xywh(rect):
     return x, y, w, h
 
 
-def remove_primary_text_regions(image):
-    """Removes regions of primary text, meaning no figure descriptions for example, but main text body paragraphs.
+def find_primary_text_regions(image):
+    """Finds regions of primary text, meaning no figure descriptions for example, but main text body paragraphs.
 
     Args:
         image: Image to remove primary text from.
@@ -174,7 +174,26 @@ def remove_primary_text_regions(image):
 
     cnts, _ = cv2.findContours(dilate, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE)
 
-    for cnt in filter(is_likely_primary_text_segments, cnts):
+    cnts = filter(is_likely_primary_text_segments, cnts)
+
+    return cnts
+
+
+def remove_primary_text_regions(image):
+    """Removes regions of primary text, meaning no figure descriptions for example, but main text body paragraphs.
+
+    Args:
+        image: Image to remove primary text from.
+
+    Returns:
+        Image with primary text removed.
+    """
+
+    image = image.copy()
+
+    cnts = find_primary_text_regions(image)
+
+    for cnt in cnts:
         x, y, w, h = cv2.boundingRect(cnt)
         cv2.rectangle(image, (x, y), (x + w, y + h), (255, 255, 255), -1)
 
@@ -206,17 +225,21 @@ def is_large_enough(cont, min_area):
     return cv2.contourArea(cont, False) > min_area
 
 
-def has_acceptable_format(cont, max_width_to_hight_ratio):
+def has_acceptable_format(cont, max_width_to_height_ratio):
     _, _, w, h = cv2.boundingRect(cont)
-    return max_width_to_hight_ratio >= w / h >= (1 / max_width_to_hight_ratio)
+    return max_width_to_height_ratio >= w / h >= (1 / max_width_to_height_ratio)
 
 
 def is_filled(hierarchy):
-    # See https://stackoverflow.com/questions/60095520/how-to-distinguish-filled-circle-contour-and-unfilled-circle-contour-in-opencv
+    """Checks whether a hierarchy is filled.
+
+    References:
+        https://stackoverflow.com/questions/60095520/how-to-distinguish-filled-circle-contour-and-unfilled-circle-contour-in-opencv
+    """
     return hierarchy[3] <= 0 and hierarchy[2] == -1
 
 
 def is_boxy(contour):
     epsilon = 0.01 * cv2.arcLength(contour, True)
     approx = cv2.approxPolyDP(contour, epsilon, True)
-    return len(approx) <= 10
+    return len(approx) <= 10