def get_words_in_bounding_boxes(extracted_bboxes, gt_bboxes, chars, page_num):
if page_num in extracted_bboxes.keys():
extracted_bbox = extracted_bboxes[page_num]
else:
extracted_bbox = []
if page_num in gt_bboxes.keys():
gt_bbox = gt_bboxes[page_num]
else:
gt_bbox = []
extracted_chars = []
gt_chars = []
for i, c in enumerate(chars):
try:
if any([
isContained((c.y0, c.x0, c.y1, c.x1), bbox)
for bbox in extracted_bbox
]):
extracted_chars += [i]
if any([
isContained((c.y0, c.x0, c.y1, c.x1), bbox)
for bbox in gt_bbox
]):
gt_chars += [i]
except AttributeError:
pass
return extracted_chars, gt_chars
def compute_overlap_matrix(pdf_bboxes, iou_thresh):
nb_tables = len(pdf_bboxes)
overlap = np.zeros((nb_tables, nb_tables))
for i, bb1 in enumerate(pdf_bboxes):
for j, bb2 in enumerate(pdf_bboxes):
if i != j and bb1[0] == bb2[0] and doOverlap(bb1[-4:], bb2[-4:]):
iou = compute_iou(bb1[-4:], bb2[-4:])
if iou > iou_thresh or isContained(
bb1[-4:], bb2[-4:]) or isContained(bb2[-4:], bb1[-4:]):
overlap[i, j] = 1.
return overlap
def get_lines_within_bbox(bbox, segments):
lines_within_bbox = []
for line in segments:
bbox_line = (int(line.y0), int(line.x0), int(line.y1), int(line.x1))
if isContained(bbox_line, bbox[-4:]):
lines_within_bbox += [line]
return lines_within_bbox
def get_mentions_within_bbox(bbox, mentions):
mentions_within_bbox = []
for mention in mentions:
bbox_mention = (int(mention.y0), int(mention.x0), int(mention.y1),
int(mention.x1))
if isContained(bbox_mention, bbox[-4:]):
mentions_within_bbox += [mention]
return mentions_within_bbox