def test_valid_boundary():
x = [0, 0, 1, 0, 1, 1, 0, 1]
assert not utils.valid_boundary(x, True)
assert not utils.valid_boundary([0])
assert utils.valid_boundary(x, False)
x = [0, 0, 1, 0, 1, 1, 0, 1, 1]
assert utils.valid_boundary(x, True)
def boundary_iou(src, target):
"""Calculate the IOU between two boundaries.
Args:
src (list): Source boundary.
target (list): Target boundary.
Returns:
iou (float): The iou between two boundaries.
"""
assert utils.valid_boundary(src, False)
assert utils.valid_boundary(target, False)
src_poly = points2polygon(src)
target_poly = points2polygon(target)
return poly_iou(src_poly, target_poly)
def boundary_iou(src, target, zero_division=0):
"""Calculate the IOU between two boundaries.
Args:
src (list): Source boundary.
target (list): Target boundary.
zero_division (int|float): The return value when invalid
boundary exists.
Returns:
iou (float): The iou between two boundaries.
"""
assert utils.valid_boundary(src, False)
assert utils.valid_boundary(target, False)
src_poly = points2polygon(src)
target_poly = points2polygon(target)
return poly_iou(src_poly, target_poly, zero_division=zero_division)
def imshow_pred_boundary(img,
boundaries_with_scores,
labels,
score_thr=0,
boundary_color='blue',
text_color='blue',
thickness=1,
font_scale=0.5,
show=True,
win_name='',
wait_time=0,
out_file=None,
show_score=False):
"""Draw boundaries and class labels (with scores) on an image.
Args:
img (str or ndarray): The image to be displayed.
boundaries_with_scores (list[list[float]]): Boundaries with scores.
labels (list[int]): Labels of boundaries.
score_thr (float): Minimum score of boundaries to be shown.
boundary_color (str or tuple or :obj:`Color`): Color of boundaries.
text_color (str or tuple or :obj:`Color`): Color of texts.
thickness (int): Thickness of lines.
font_scale (float): Font scales of texts.
show (bool): Whether to show the image.
win_name (str): The window name.
wait_time (int): Value of waitKey param.
out_file (str or None): The filename of the output.
show_score (bool): Whether to show text instance score.
"""
assert isinstance(img, (str, np.ndarray))
assert utils.is_2dlist(boundaries_with_scores)
assert utils.is_type_list(labels, int)
assert utils.equal_len(boundaries_with_scores, labels)
if len(boundaries_with_scores) == 0:
warnings.warn('0 text found in ' + out_file)
return
utils.valid_boundary(boundaries_with_scores[0])
img = mmcv.imread(img)
scores = np.array([b[-1] for b in boundaries_with_scores])
inds = scores > score_thr
boundaries = [boundaries_with_scores[i][:-1] for i in np.where(inds)[0]]
scores = [scores[i] for i in np.where(inds)[0]]
labels = [labels[i] for i in np.where(inds)[0]]
boundary_color = mmcv.color_val(boundary_color)
text_color = mmcv.color_val(text_color)
font_scale = 0.5
for boundary, score, label in zip(boundaries, scores, labels):
boundary_int = np.array(boundary).astype(np.int32)
cv2.polylines(img, [boundary_int.reshape(-1, 1, 2)],
True,
color=boundary_color,
thickness=thickness)
if show_score:
label_text = f'{score:.02f}'
cv2.putText(img, label_text,
(boundary_int[0], boundary_int[1] - 2),
cv2.FONT_HERSHEY_COMPLEX, font_scale, text_color)
if show:
mmcv.imshow(img, win_name, wait_time)
if out_file is not None:
mmcv.imwrite(img, out_file)
return img
def eval_hmean(results,
img_infos,
ann_infos,
metrics={'hmean-iou'},
score_thr=0.3,
rank_list=None,
logger=None,
**kwargs):
"""Evaluation in hmean metric.
Args:
results (list[dict]): Each dict corresponds to one image,
containing the following keys: boundary_result
img_infos (list[dict]): Each dict corresponds to one image,
containing the following keys: filename, height, width
ann_infos (list[dict]): Each dict corresponds to one image,
containing the following keys: masks, masks_ignore
score_thr (float): Score threshold of prediction map.
metrics (set{str}): Hmean metric set, should be one or all of
{'hmean-iou', 'hmean-ic13'}
Returns:
dict[str: float]
"""
assert utils.is_type_list(results, dict)
assert utils.is_type_list(img_infos, dict)
assert utils.is_type_list(ann_infos, dict)
assert len(results) == len(img_infos) == len(ann_infos)
assert isinstance(metrics, set)
gts, gts_ignore = get_gt_masks(ann_infos)
preds = []
pred_scores = []
for result in results:
_, texts, scores = extract_boundary(result)
if len(texts) > 0:
assert utils.valid_boundary(texts[0], False)
valid_texts, valid_text_scores = filter_2dlist_result(
texts, scores, score_thr)
preds.append(valid_texts)
pred_scores.append(valid_text_scores)
eval_results = {}
for metric in metrics:
msg = f'Evaluating {metric}...'
if logger is None:
msg = '\n' + msg
print_log(msg, logger=logger)
best_result = dict(hmean=-1)
for iter in range(3, 10):
thr = iter * 0.1
if thr < score_thr:
continue
top_preds = select_top_boundary(preds, pred_scores, thr)
if metric == 'hmean-iou':
result, img_result = hmean_iou.eval_hmean_iou(
top_preds, gts, gts_ignore)
elif metric == 'hmean-ic13':
result, img_result = hmean_ic13.eval_hmean_ic13(
top_preds, gts, gts_ignore)
else:
raise NotImplementedError
if rank_list is not None:
output_ranklist(img_result, img_infos, rank_list)
print_log('thr {0:.2f}, recall: {1[recall]:.3f}, '
'precision: {1[precision]:.3f}, '
'hmean: {1[hmean]:.3f}'.format(thr, result),
logger=logger)
if result['hmean'] > best_result['hmean']:
best_result = result
eval_results[metric + ':recall'] = best_result['recall']
eval_results[metric + ':precision'] = best_result['precision']
eval_results[metric + ':hmean'] = best_result['hmean']
return eval_results
def eval_hmean(results,
img_infos,
ann_infos,
metrics={'hmean-iou'},
score_thr=None,
min_score_thr=0.3,
max_score_thr=0.9,
step=0.1,
rank_list=None,
logger=None,
**kwargs):
"""Evaluation in hmean metric. It conducts grid search over a range of
boundary score thresholds and reports the best result.
Args:
results (list[dict]): Each dict corresponds to one image,
containing the following keys: boundary_result
img_infos (list[dict]): Each dict corresponds to one image,
containing the following keys: filename, height, width
ann_infos (list[dict]): Each dict corresponds to one image,
containing the following keys: masks, masks_ignore
score_thr (float): Deprecated. Please use min_score_thr instead.
min_score_thr (float): Minimum score threshold of prediction map.
max_score_thr (float): Maximum score threshold of prediction map.
step (float): The spacing between score thresholds.
metrics (set{str}): Hmean metric set, should be one or all of
{'hmean-iou', 'hmean-ic13'}
Returns:
dict[str: float]
"""
assert utils.is_type_list(results, dict)
assert utils.is_type_list(img_infos, dict)
assert utils.is_type_list(ann_infos, dict)
if score_thr:
warnings.warn('score_thr is deprecated. Please use min_score_thr '
'instead.')
min_score_thr = score_thr
assert 0 <= min_score_thr <= max_score_thr <= 1
assert 0 <= step <= 1
assert len(results) == len(img_infos) == len(ann_infos)
assert isinstance(metrics, set)
min_score_thr = float(min_score_thr)
max_score_thr = float(max_score_thr)
step = float(step)
gts, gts_ignore = get_gt_masks(ann_infos)
preds = []
pred_scores = []
for result in results:
_, texts, scores = extract_boundary(result)
if len(texts) > 0:
assert utils.valid_boundary(texts[0], False)
valid_texts, valid_text_scores = filter_2dlist_result(
texts, scores, min_score_thr)
preds.append(valid_texts)
pred_scores.append(valid_text_scores)
eval_results = {}
for metric in metrics:
msg = f'Evaluating {metric}...'
if logger is None:
msg = '\n' + msg
print_log(msg, logger=logger)
best_result = dict(hmean=-1)
for thr in np.arange(min_score_thr, min(max_score_thr + step, 1.0),
step):
top_preds = select_top_boundary(preds, pred_scores, thr)
if metric == 'hmean-iou':
result, img_result = hmean_iou.eval_hmean_iou(
top_preds, gts, gts_ignore)
elif metric == 'hmean-ic13':
result, img_result = hmean_ic13.eval_hmean_ic13(
top_preds, gts, gts_ignore)
else:
raise NotImplementedError
if rank_list is not None:
output_ranklist(img_result, img_infos, rank_list)
print_log(
'thr {0:.2f}, recall: {1[recall]:.3f}, '
'precision: {1[precision]:.3f}, '
'hmean: {1[hmean]:.3f}'.format(thr, result),
logger=logger)
if result['hmean'] > best_result['hmean']:
best_result = result
eval_results[metric + ':recall'] = best_result['recall']
eval_results[metric + ':precision'] = best_result['precision']
eval_results[metric + ':hmean'] = best_result['hmean']
return eval_results