Merge remote-tracking branch 'bertsky/ocrd-utils-scale'

ocrd/ocrd/workspace.py

@@ -1051,3 +1051,22 @@ def _rotate(log, name, skew, segment, segment_image, segment_coords, segment_xyw
             log, name, segment, segment_image, segment_coords,
             op='recropped', **kwargs)
     return segment_image, segment_coords, segment_xywh
+
+def _scale(log, name, factor, segment_image, segment_coords, segment_xywh, **kwargs):
+    # Resize linearly
+    segment_coords['transform'] = scale_coordinates(
+        segment_coords['transform'], [factor, factor])
+    segment_coords['scale'] = segment_coords.setdefault('scale', 1.0) * factor
+    segment_xywh['w'] *= factor
+    segment_xywh['h'] *= factor
+    # resize, if (still) necessary
+    if not 'scaled' in segment_coords['features']:
+        log.info("Scaling %s by %.2f", name, factor)
+        segment_coords['features'] += ',scaled'
+        # FIXME: validate factor against PAGE-XML attributes
+        # FIXME: factor should become less precise due to rounding
+        segment_image = segment_image.resize((int(segment_image.width * factor),
+                                              int(segment_image.height * factor)),
+                                             # slowest, but highest quality:
+                                             Image.BICUBIC)
+    return segment_image, segment_coords, segment_xywh

ocrd_utils/ocrd_utils/image.py

@@ -27,6 +27,7 @@
     'polygon_mask',
     'rotate_coordinates',
     'shift_coordinates',
+    'scale_coordinates',
     'transform_coordinates',
     'transpose_coordinates',
     'xywh_from_bbox',
@@ -300,6 +301,23 @@ def shift_coordinates(transform, offset):
     shift[1, 2] = offset[1]
     return np.dot(shift, transform)
 
+def scale_coordinates(transform, factors):
+    """Compose an affine coordinate transformation with a proportional scaling.
+    Given a numpy array ``transform`` of an existing transformation
+    matrix in homogeneous (3d) coordinates, and a numpy array
+    ``factors`` of the scaling factors, calculate the affine
+    coordinate transform corresponding to the composition of both
+    transformations.
+    
+    Return a numpy array of the resulting affine transformation matrix.
+    """
+    LOG = getLogger('ocrd_utils.coords.scale_coordinates')
+    LOG.debug('scaling coordinates by %s', str(factors))
+    scale = np.eye(3)
+    scale[0, 0] = factors[0]
+    scale[1, 1] = factors[1]
+    return np.dot(scale, transform)
+
 def transform_coordinates(polygon, transform=None):
     """Apply an affine transformation to a set of points.
     Augment the 2d numpy array of points ``polygon`` with a an extra