firetac commited on
Commit
a642897
·
verified ·
1 Parent(s): f709e40

Delete HandwritingOCR/testParser.py

Browse files
Files changed (1) hide show
  1. HandwritingOCR/testParser.py +0 -45
HandwritingOCR/testParser.py DELETED
@@ -1,45 +0,0 @@
1
- import layoutparser as lp
2
- import cv2
3
- import matplotlib.pyplot as plt
4
-
5
-
6
- image = cv2.imread("Projects/HandwritingOCR/captured_images/pasted_image.jpg")
7
- #image = image[..., ::-1]
8
-
9
- model = lp.Detectron2LayoutModel(
10
- "lp://PubLayNet/faster_rcnn_R_50_FPN_3x/config",
11
- extra_config=["MODEL.ROI_HEADS.SCORE_THRESH_TEST", 0.8],
12
- label_map={0: "Text", 1: "Title", 2: "List", 3:"Table", 4:"Figure"},
13
- )
14
-
15
-
16
- # Phát hiện bố cục trên ảnh đã được cải thiện độ phân giải
17
- layout = model.detect(image)
18
-
19
- # Vẽ các hộp phát hiện lên ảnh
20
- lp.draw_box(image, layout, box_width=3)
21
-
22
- print(type(layout))
23
- print(layout[0])
24
- text_blocks = lp.Layout([b for b in layout if b.type=='Text'])
25
- figure_blocks = lp.Layout([b for b in layout if b.type=='Figure'])
26
-
27
- text_blocks = lp.Layout([b for b in text_blocks \
28
- if not any(b.is_in(b_fig) for b_fig in figure_blocks)])
29
- h, w = image.shape[:2]
30
-
31
- left_interval = lp.Interval(0, w/2*1.05, axis='x').put_on_canvas(image)
32
-
33
- left_blocks = text_blocks.filter_by(left_interval, center=True)
34
- left_blocks.sort(key = lambda b:b.coordinates[1])
35
-
36
- right_blocks = [b for b in text_blocks if b not in left_blocks]
37
- right_blocks.sort(key = lambda b:b.coordinates[1])
38
-
39
- # And finally combine the two list and add the index
40
- # according to the order
41
- text_blocks = lp.Layout([b.set(id = idx) for idx, b in enumerate(left_blocks + right_blocks)])
42
-
43
- lp.draw_box(image, text_blocks,
44
- box_width=3,
45
- show_element_id=True)