Merge pull request #394 from jlaura/master (942f2e83) · Commits · aflab / astrogeology / Autocnet

autocnet/matcher/subpixel.py

+31 −13

Original line number	Diff line number	Diff line
		@@ -304,8 +304,10 @@ def iterative_phase(sx, sy, dx, dy, s_img, d_img, size=251, reduction=11, conver
		A plio geodata object from which the template is extracted
		d_img : object
		A plio geodata object from which the search is extracted
		size : int
		One half of the total size of the template, so a 251 default results in a 502 pixel search space
		size : int, tuple
		One half of the total size of the template, so a 251 default results in a 502 pixel search space.
		If an int, the template is square. If a tuple, in the form (x,y), is passed an
		irregularly shaped template can be used.
		reduction : int
		With each recursive call to this func, the size is reduced by this amount
		convergence_threshold : float
		@@ -329,26 +331,29 @@ def iterative_phase(sx, sy, dx, dy, s_img, d_img, size=251, reduction=11, conver
		# get initial destination location
		dsample = dx
		dline = dy
		if isinstance(size, int):
		size = (size, size)
		while True:
		s_template, _, _ = clip_roi(s_img, sx, sy,
		size_x=size, size_y=size)
		size_x=size[0], size_y=size[1])
		d_search, dxr, dyr = clip_roi(d_img, dx, dy,
		size_x=size, size_y=size)
		size_x=size[0], size_y=size[1])

		if (s_template is None) or (d_search is None):
		return None, None, None
		if s_template.shape != d_search.shape:
		s_size = s_template.shape
		d_size = d_search.shape
		updated_size = int(min(s_size + d_size) / 2)
		updated_size_x = int(min(s_size[1] + d_size[1])) # Why is this /2?
		updated_size_y = int(min(s_size[0] + d_size[0]))
		# Since the image is smaller than the requested size, set the size to
		# the current maximum image size and reduce from there on potential
		# future iterations.
		size = updated_size
		size = (updated_size_x, updated_size_y)
		s_template, _, _ = clip_roi(s_template, sx, sy,
		size_x=updated_size, size_y=updated_size)
		size_x=size[0], size_y=size[1])
		d_search, dxr, dyr = clip_roi(d_search, dx, dy,
		size_x=updated_size, size_y=updated_size)
		size_x=size[0], size_y=size[1])
		if (s_template is None) or (d_search is None):
		return None, None, None

		@@ -362,9 +367,10 @@ def iterative_phase(sx, sy, dx, dy, s_img, d_img, size=251, reduction=11, conver
		dy += (shift_y + dyr)

		# Break if the solution has converged
		size -= reduction
		size = (size[0] - reduction, size[1] - reduction)

		dist = np.linalg.norm([dsample-dx, dline-dy])
		if size <1:
		if min(size) < 1:
		return None, None, None
		if abs(shift_x) <= convergence_threshold and\
		abs(shift_y) <= convergence_threshold and\
		@@ -450,6 +456,12 @@ def subpixel_register_point(pointid, iterative_phase_kwargs={}, subpixel_templat
		measure.sample = new_template_x
		measure.line = new_template_y
		measure.weight = cost

		# In case this is a second run, set the ignore to False if this
		# measures passed. Also, set the source measure back to ignore=False
		measure.ignore = False
		source.ignore = False

		session.commit()
		session.close()

		@@ -494,6 +506,7 @@ def cluster_subpixel_register_points(iterative_phase_kwargs={'size': 251},
		subpixel_template_kwargs={'image_size':(251,251)},
		cost_func=lambda x,y: 1/x*2 y,
		threshold=0.005,
		filters = {},
		walltime='00:10:00',
		chunksize=1000,
		exclude=None):
		@@ -521,7 +534,9 @@ def cluster_subpixel_register_points(iterative_phase_kwargs={'size': 251},
		threshold : numeric
		measures with a cost <= the threshold are marked as ignore=True in
		the database.

		filters : dict
		with keys equal to attributes of the Points mapping and values
		equal to some criteria.
		exclude : str
		string containing the name(s) of any slurm nodes to exclude when
		completing a cluster job. (e.g.: 'gpu1' or 'gpu1,neb12')
		@@ -534,9 +549,12 @@ def cluster_subpixel_register_points(iterative_phase_kwargs={'size': 251},
		# Push the job messages onto the queue
		queuename = config['redis']['processing_queue']


		session = Session()
		for i, point in enumerate(session.query(Points)):
		query = session.query(Points)
		for attr, value in filters.items():
		query = query.filter(getattr(Points, attr)==value)
		res = query.all()
		for i, point in enumerate(res):
		msg = {'id' : point.id,
		'iterative_phase_kwargs' : iterative_phase_kwargs,
		'subpixel_template_kwargs' : subpixel_template_kwargs,