mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-26 20:14:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore(ocr): revert layout updates

Browse Source 
Signed-off-by: Clément Doumouro <clement.doumouro@gmail.com>
This commit is contained in:
Clément Doumouro 2025-04-22 17:35:43 +02:00
parent 1181338737
commit bf00fa1a9f
4 changed files with 42 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
docling/models/tesseract_ocr_cli_model.py
View File

@ -266,11 +266,15 @@ class TesseractOcrCliModel(BaseOcrModel):
text = row["text"] text = row["text"]
conf = row["conf"] conf = row["conf"]
l, t = float(row["left"]), float(row["top"]) left, top = float(row["left"]), float(row["top"])
r = l + float(row["width"]) right = left + float(row["width"])
b = t + row["height"] bottom = top + row["height"]
bbox = BoundingBox( bbox = BoundingBox(
l=l, t=t, r=r, b=b, coord_origin=CoordOrigin.TOPLEFT l=left,
t=top,
r=right,
b=bottom,
coord_origin=CoordOrigin.TOPLEFT,
) )
rect = tesseract_box_to_bounding_rectangle( rect = tesseract_box_to_bounding_rectangle(
bbox, bbox,

12
docling/models/tesseract_ocr_model.py
View File

@ -194,11 +194,15 @@ class TesseractOcrModel(BaseOcrModel):
# Extract text within the bounding box # Extract text within the bounding box
text = local_reader.GetUTF8Text().strip() text = local_reader.GetUTF8Text().strip()
confidence = local_reader.MeanTextConf() confidence = local_reader.MeanTextConf()
l, t = box["x"], box["y"] left, top = box["x"], box["y"]
r = l + box["w"] right = left + box["w"]
b = t + box["h"] bottom = top + box["h"]
bbox = BoundingBox( bbox = BoundingBox(
l=l, t=t, r=r, b=b, coord_origin=CoordOrigin.TOPLEFT l=left,
t=top,
r=right,
b=bottom,
coord_origin=CoordOrigin.TOPLEFT,
) )
rect = tesseract_box_to_bounding_rectangle( rect = tesseract_box_to_bounding_rectangle(
bbox, bbox,

2
docling/utils/ocr_utils.py
View File

@ -54,7 +54,7 @@ def tesseract_box_to_bounding_rectangle(
coord_origin=CoordOrigin.TOPLEFT, coord_origin=CoordOrigin.TOPLEFT,
) )
if original_offset is not None: if original_offset is not None:
if not original_offset.coord_origin is CoordOrigin.TOPLEFT: if original_offset.coord_origin is not CoordOrigin.TOPLEFT:
msg = f"expected coordinate origin to be {CoordOrigin.TOPLEFT.value}" msg = f"expected coordinate origin to be {CoordOrigin.TOPLEFT.value}"
raise ValueError(msg) raise ValueError(msg)
if original_offset is not None: if original_offset is not None:

50
docling/utils/orientation.py
View File

@ -13,44 +13,44 @@ def rotate_bounding_box(
# Bounding rectangle start with r_0 at the bottom left whatever the # Bounding rectangle start with r_0 at the bottom left whatever the
# coordinate system. Then other corners are found rotating counterclockwise # coordinate system. Then other corners are found rotating counterclockwise
bbox = bbox.to_top_left_origin(im_size[1]) bbox = bbox.to_top_left_origin(im_size[1])
l, t, w, h = bbox.l, bbox.t, bbox.width, bbox.height left, top, width, height = bbox.l, bbox.t, bbox.width, bbox.height
im_h, im_w = im_size im_h, im_w = im_size
angle = angle % 360 angle = angle % 360
if angle == 0: if angle == 0:
r_x0 = l r_x0 = left
r_y0 = t + h r_y0 = top + height
r_x1 = r_x0 + w r_x1 = r_x0 + width
r_y1 = r_y0 r_y1 = r_y0
r_x2 = r_x0 + w r_x2 = r_x0 + width
r_y2 = r_y0 - h r_y2 = r_y0 - height
r_x3 = r_x0 r_x3 = r_x0
r_y3 = r_y0 - h r_y3 = r_y0 - height
elif angle == 90: elif angle == 90:
r_x0 = im_w - (t + h) r_x0 = im_w - (top + height)
r_y0 = l r_y0 = left
r_x1 = r_x0 r_x1 = r_x0
r_y1 = r_y0 + w r_y1 = r_y0 + width
r_x2 = r_x0 + h r_x2 = r_x0 + height
r_y2 = r_y0 + w r_y2 = r_y0 + width
r_x3 = r_x0 r_x3 = r_x0
r_y3 = r_y0 + w r_y3 = r_y0 + width
elif angle == 180: elif angle == 180:
r_x0 = im_h - l r_x0 = im_h - left
r_y0 = im_w - (t + h) r_y0 = im_w - (top + height)
r_x1 = r_x0 - w r_x1 = r_x0 - width
r_y1 = r_y0 r_y1 = r_y0
r_x2 = r_x0 - w r_x2 = r_x0 - width
r_y2 = r_y0 + h r_y2 = r_y0 + height
r_x3 = r_x0 r_x3 = r_x0
r_y3 = r_y0 + h r_y3 = r_y0 + height
elif angle == 270: elif angle == 270:
r_x0 = t + h r_x0 = top + height
r_y0 = im_h - l r_y0 = im_h - left
r_x1 = r_x0 r_x1 = r_x0
r_y1 = r_y0 - w r_y1 = r_y0 - width
r_x2 = r_x0 - h r_x2 = r_x0 - height
r_y2 = r_y0 - w r_y2 = r_y0 - width
r_x3 = r_x0 - h r_x3 = r_x0 - height
r_y3 = r_y0 r_y3 = r_y0
else: else:
msg = ( msg = (