Update ImageProcessor.py

ImageProcessor.py

@@ -7,6 +7,12 @@ import pytesseract
 import warnings
 import re
 
+def show_image(img):
+    cv2.imshow("", img)
+    cv2.waitKey(0)
+    cv2.destroyAllWindows()
+    return
+
 def pdf2png(folderpath):
     doc = fitz.open(folderpath + '/opinion.pdf')
     zoom = 1
@@ -17,12 +23,11 @@ def pdf2png(folderpath):
 
 def is_leftmost(image, x, y_top, y_bot):
     gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
-    blur = cv2.GaussianBlur(gray, (9,9), 0)
+    blur = cv2.GaussianBlur(gray, (9, 9), 0)
     thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]
-    left_portion = thresh[int((y_top+y_bot)/2), :x]
+    left_portion = thresh[int(0.2*y_top+0.8*y_bot), :x]
     return np.sum(left_portion) == 0
 
-
 def get_line_data(filename, body_bbox, page):
     image = cv2.imread(filename)
     body_rect = fitz.Rect(body_bbox)
@@ -30,7 +35,7 @@ def get_line_data(filename, body_bbox, page):
     all_lines = [(int(line['bbox'][0]), int(line['bbox'][1]), int(line['bbox'][2]), int(line['bbox'][3]), line)for block in pg_dict['blocks'] for line in block['lines']]
     line_data = []
     for (i,l) in enumerate(all_lines):
-        if not is_leftmost(image, l[0]-11, l[1], l[3]) and i > 0: # Add it
+        if not is_leftmost(image, l[0]-9, l[1], l[3]) and i > 0: # Add it
             line_data[-1] = list(line_data[-1])
             line_data[-1][-1] += " " + get_line_text(l[-1])
             line_data[-1] = tuple(line_data[-1])
@@ -140,8 +145,6 @@ def get_page_elements(filename, page):
         cv2.rectangle(image, (fn_bbox[0], fn_bbox[1]), (fn_bbox[2], fn_bbox[3]), (0, 0, 255), 2)
     if case_separator_bbox[0] is not None:
         cv2.rectangle(image, (case_separator_bbox[0], case_separator_bbox[1]), (case_separator_bbox[2], case_separator_bbox[3]), (255, 0, 255), 2)
-    # for (i, il) in enumerate(indent_lines):
-    #     cv2.circle(image, (il[1]-15, int(0.5*(il[2] + il[4]))), radius=1, color=(240, 32, 160), thickness=2)
 
     return page_bbox, header_bbox, fn_bbox, body_bbox, case_separator_bbox, line_data, image