def test_FindRegions():
image = skimage.data.camera()
with Pipeline() as pipeline:
mask = ThresholdConst(image, 255)
result = FindRegions(mask, image, 0, 100, padding=10)
pipeline.run()
def test_ExtractROI():
with Pipeline() as pipeline:
image = Unpack([skimage.data.camera()])
mask = ThresholdConst(image, 255)
regions = FindRegions(mask, image)
ExtractROI(image, regions)
ExtractROI(image, regions, 0.5)
pipeline.run()
def test_DrawContoursOnParent():
image = skimage.data.camera()
with Pipeline() as pipeline:
mask = ThresholdConst(image, 255)
regions = FindRegions(mask, image, 0, 100, padding=10)
output_ref = image
result = DrawContoursOnParent(image, mask, output_ref, regions, (0, 255, 0))
stream = pipeline.transform_stream()
pipeline.run()
def test_CalculateZooProcessFeatures(prefix):
with Pipeline() as p:
i = Unpack(range(10))
mask = BinaryBlobs()
image = NoiseImage(mask.shape)
regionprops = FindRegions(mask, image)
CalculateZooProcessFeatures(regionprops, prefix=prefix)
p.run()
def test_FindRegions(warn_empty, recwarn):
images = [skimage.data.camera(), np.zeros((10, 10), np.uint8) + 255]
with Pipeline() as pipeline:
image = Unpack(images)
mask = ThresholdConst(image, 255)
result = FindRegions(mask,
image,
0,
100,
padding=10,
warn_empty=warn_empty)
pipeline.run()
if warn_empty:
w = recwarn.pop(UserWarning)
assert re.search(r"^(Image|foo) did not contain any objects.$",
str(w.message))
img_dilate = Call(cv2.dilate, img_canny, kernel, iterations=2)
#Close
kernel = Call(cv2.getStructuringElement, cv2.MORPH_ELLIPSE, (5, 5))
img_close = Call(cv2.morphologyEx, img_dilate, cv2.MORPH_CLOSE, kernel, iterations=1)
#Erode
kernel = Call(cv2.getStructuringElement, cv2.MORPH_ELLIPSE, (15, 15))
mask = Call(cv2.erode, img_close, kernel, iterations=2)
frame_fn = Format(os.path.join(ANNOTATED, "{name}.jpg"), name=name)
ImageWriter(frame_fn, mask)
# Find objects
regionprops = FindRegions(
mask, img_gray, min_area=1000, padding=10, warn_empty=name
)
# For an object, extract a vignette/ROI from the image
roi_orig = ExtractROI(img, regionprops, bg_color=255)
# For an object, extract a vignette/ROI from the image
roi_mask = ExtractROI(mask, regionprops, bg_color=255)
# Generate an object identifier
i = Enumerate()
#Call(print,i)
object_id = Format("{name}_{i:d}", name=name, i=i)
#Call(print,object_id)
# Remove path and extension from the filename
source_basename = Call(lambda x: os.path.splitext(os.path.basename(x))[0], path)
with ParallelPipeline():
# The following operations are distributed among multiple
# worker processes to speed up the calculations.
# Read the image
image = ImageReader(path)
# Do some thresholding
mask = image < 128
# Find regions in the image
region = FindRegions(mask, image)
# Extract just the object
roi_image = region.intensity_image
# An object is identified by its label
roi_label = region.label
# Calculate a filename for the ROI image:
# "RUNNING_NUMBER-SOURCE_BASENAME-ROI_LABEL"
roi_name = Format(
"{:d}-{}-{:d}.jpg", running_number, source_basename, roi_label
)
meta = CalculateZooProcessFeatures(region, prefix="object_")
# End of parallel execution
meta,
drop_duplicates_subset="sample_id",
)
img = Call(skimage.io.imread, abs_path)
StreamBuffer(maxsize=2)
TQDM(rel_path)
img = Rescale(img, in_range=(9252, 65278), dtype=np.uint8)
mask = ThresholdConst(img, 245) # 245(ubyte) / 62965(uint16)
mask = Call(skimage.segmentation.clear_border, mask)
regionprops = FindRegions(mask, img, 100, padding=10)
# TODO: Draw object info
# Extract a vignette from the image
vignette = ExtractROI(img, regionprops)
# # Extract features from vignette
# model = resnet18(pretrained=True)
# model = torch.nn.Sequential(OrderedDict(
# list(model.named_children())[:-2]))
# features = PyTorch(lambda x: model(x).cpu().numpy())(vignette)
i = Enumerate()
object_id = Format(