def process_image(image,
filters,
ROI=None,
return_predictions=False,
combine_results=False,
sort_filters=True):
"""Processes one image with process-function and returns the resulting value.
"""
return qualipy.process(image, filters, ROI, return_predictions,
combine_results, sort_filters)
def test_works_with_magic_thresholds():
assert qualipy.process(TEST_IMG, [
Framed() == 1,
Pattern() > 0.3,
HDR() >= 0.5,
UnconventionalSize() <= 16 / 9.,
Exposure() != 1,
Highlights() < 0.1
]) == {
TEST_IMG: False
}
def test_fails_for_invalid_images():
with pytest.raises(TypeError):
assert qualipy.process(0, [Framed()])
def test_fails_for_invalid_amount_of_ROIs():
with pytest.raises(ValueError):
assert qualipy.process([TEST_IMG, TEST_IMG2], [Framed()],
[(0, 0, 100, 100)])
def test_fails_for_invalid_length_ROI():
with pytest.raises(TypeError):
assert qualipy.process(TEST_IMG, [Framed()], (10, 10, 100))
def test_works_correctly_for_multiple_ROIs():
assert qualipy.process([TEST_IMG, TEST_IMG2], [Framed()],
[(0, 0, 100, 100), None])
def test_works_correctly_for_valid_ROI():
assert qualipy.process(TEST_IMG, [Framed()], (0, 0, 100, 100))
def test_processes_single_image_correctly():
assert qualipy.process(TEST_IMG, [Framed()])
def test_returns_boolean_for_each_filter_when_not_combining_results():
assert qualipy.process(TEST_IMG, [Framed()], None, False,
False)[TEST_IMG]['framed'] == 0.
def test_returns_float_when_correct_parameter_is_set():
assert qualipy.process(TEST_IMG, [Framed()], None,
True)[TEST_IMG]['framed'] == 0.
assert qualipy.process(TEST_IMG, [Framed()], None, True,
False)[TEST_IMG]['framed'] == 0.
def test_returns_False_when_some_filters_return_positive():
assert qualipy.process(TEST_IMG2, [Framed(), Pattern()]) == {
TEST_IMG2: False
}
def test_returns_True_when_all_filters_return_negative():
assert qualipy.process(TEST_IMG, [Framed(), Pattern()])
def test_images_exist_in_resulting_dict():
res = qualipy.process([TEST_IMG, TEST_IMG2], [])
assert TEST_IMG in res and TEST_IMG2 in res
def test_processes_list_of_images_correctly():
assert len(qualipy.process([TEST_IMG], [Framed()], None, True)) == 1
def test_returns_true_for_no_filters():
assert qualipy.process(TEST_IMG, []) == {TEST_IMG: True}
def get_prediction(path):
return not qualipy.process(path, [filter])
def test_ROI_can_be_None():
assert qualipy.process(TEST_IMG, [Framed()], None)