lavaflow.utils package

lavaflow.utils.config module

Utility functions for config.

lavaflow.utils.config.compare_config(a, b, aname='a', bname='b', keys=True, types=True, values=True)[source]

Function to compare the key structure and value types of two configs.

Parameters

a (dict) – config
b (dict) – config
aname (str) – config name
bname (str) – config name
keys (bool) – check if keys are not the same at each level of the dictionary
types (bool) – check if types are not the same
values (bool) – check if values are not the same

Returns

comparison summary

Return type

d (dict)

Examples

>>> a = { 'a': 1, 'b': { 'c': 3, 'd': 4, 'e': 'x' }}
>>> b = { 'a': 3, 'b': { 'c': 3, 'd': 'y', 'g': 7 }}
>>> compare_config(a, b)
['a.a = 1 != b.a = 3', "a.b.keys() - b.b.keys() = ['e']", "b.b.keys() - a.b.keys() = ['g']", "type(a.b.d) = <class 'int'> != type(b.b.d) = <class 'str'>"]

lavaflow.utils.config.remove_inline_comments(config_yaml, prefix='#')[source]

Remove inline comments from config yaml based on comment prefix.

Parameters

config_yaml (str) – config yaml with inline comments
prefix (str) – comment prefix (default: #)

Returns

config with inline comments removed

Return type

config_yaml (str)

lavaflow.utils.core module

Utility functions.

lavaflow.utils.core.chain(x)[source]

Function to flatten a list of lists.

Parameters: x (list) – list of lists of items
Returns: list of items
Return type: items (list)

Examples

>>> x = [[1, 2], [3, 4]]
>>> chain(x)
[1, 2, 3, 4]

lavaflow.utils.core.compose(functions)[source]

Compose an iterator of functions in sequence.

Parameters: functions (iterable) – an iterator of functions
Returns: composition of functions
Return type: f (function)

Examples

>>> g = compose([lambda x: x + 1, lambda x: x * 2])
>>> g(1)
4

lavaflow.utils.core.dictify(f, x, y)[source]

Wrap a function to take a dictionary as input and to map input key(s) to output key(s).

Parameters

f (function) – function to be wrapped
x (str|list) – input key(s)
y (str|list) – output key(s)

Returns

new function

Return type

g (function)

Examples

>>> g = dictify(lambda a: a + 1, 'a', 'b')
>>> g({'a': 1})
{'a': 1, 'b': 2}

>>> g = dictify(lambda a, b: a + b, ['a', 'b'], 'c')
>>> g({'a': 1, 'b': 2})
{'a': 1, 'b': 2, 'c': 3}

>>> g = dictify(lambda a, b: (a + b, b - a), ['a', 'b'], ['c', 'd'])
>>> g({'a': 1, 'b': 2})
{'a': 1, 'b': 2, 'c': 3, 'd': 1}

lavaflow.utils.core.windowify(f, i, x, j, y, chain_args)[source]

Wrap a function to take a list of dictionaries as input and to map input key(s) to output key(s).

Parameters

f (function) – function to be wrapped
i (int|list) – input item(s)
x (str|list) – input key(s)
j (int|list) – output item(s)
y (str|list) – output key(s)
chain_args (bool) – flatten inputs

Returns

new function

Return type

g (function)

Examples

>>> g = windowify(lambda a0, a1, b0, b1: (a0 + a1, b0 + b1), [0, 2], ['a', 'b'], [1], ['c', 'd'], True)
>>> g([{'a': 1, 'b': 1}, {'a': 2, 'b': 2} , {'a': 3, 'b': 3}])
[{'a': 1, 'b': 1}, {'a': 2, 'b': 2, 'c': 4, 'd': 4}, {'a': 3, 'b': 3}]

>>> g = windowify(lambda a, b: (a[0] + a[1], b[0] + b[1]), [0, 2], ['a', 'b'], [1], ['c', 'd'], False)
>>> g([{'a': 1, 'b': 1}, {'a': 2, 'b': 2} , {'a': 3, 'b': 3}])
[{'a': 1, 'b': 1}, {'a': 2, 'b': 2, 'c': 4, 'd': 4}, {'a': 3, 'b': 3}]

lavaflow.utils.exif module

Utility functions.

lavaflow.utils.exif.get_lens_profile(tags)[source]

Function to extract lens profile from exif.

Parameters: tags (dict) – lens profile extracted using https://exiftool.org/
Returns: camera maker string model (str): camera model string lens (str): lens model string focal_length (double): focal length (mm) aperture (double): aperture distance (double): approximate focal distance (m)
Return type: maker (str)

lavaflow.utils.geometry module

Utility functions for geometry.

lavaflow.utils.geometry.distance(a, b)[source]

Compute the distance between two points.

Parameters

a (tuple) – point 1
b (tuple) – point 2

Returns

distance between two points

Return type

dist (float)

lavaflow.utils.geometry.line_proximity(p, a, b)[source]

Compute the shortest distance between the point p and the line passing through two points (a - b)

Parameters

p (tuple) – point to compute proximity
a (tuple) – line point 1
b (tuple) – line point 2

Returns

the shortest distance

Return type

dist (float)

lavaflow.utils.geometry.roundPoint(p)[source]

Function to round a point to the nearest integer.

Parameters: p (tuple) – point to round
Returns: point with rounding applied
Return type: p (tuple)

lavaflow.utils.io module

Utility functions.

class lavaflow.utils.io.NumpyJsonEncoder(*, skipkeys=False, ensure_ascii=True, check_circular=True, allow_nan=True, sort_keys=False, indent=None, separators=None, default=None)[source]

Bases: JSONEncoder

Extension of json.JSONEncoder that supports number arrays as lists only.

Constructor for JSONEncoder, with sensible defaults.

If skipkeys is false, then it is a TypeError to attempt encoding of keys that are not str, int, float or None. If skipkeys is True, such items are simply skipped.

If ensure_ascii is true, the output is guaranteed to be str objects with all incoming non-ASCII characters escaped. If ensure_ascii is false, the output can contain non-ASCII characters.

If check_circular is true, then lists, dicts, and custom encoded objects will be checked for circular references during encoding to prevent an infinite recursion (which would cause an OverflowError). Otherwise, no such check takes place.

If allow_nan is true, then NaN, Infinity, and -Infinity will be encoded as such. This behavior is not JSON specification compliant, but is consistent with most JavaScript based encoders and decoders. Otherwise, it will be a ValueError to encode such floats.

If sort_keys is true, then the output of dictionaries will be sorted by key; this is useful for regression tests to ensure that JSON serializations can be compared on a day-to-day basis.

If indent is a non-negative integer, then JSON array elements and object members will be pretty-printed with that indent level. An indent level of 0 will only insert newlines. None is the most compact representation.

If specified, separators should be an (item_separator, key_separator) tuple. The default is (’, ‘, ‘: ‘) if indent is None and (‘,’, ‘: ‘) otherwise. To get the most compact JSON representation, you should specify (‘,’, ‘:’) to eliminate whitespace.

If specified, default is a function that gets called for objects that can’t otherwise be serialized. It should return a JSON encodable version of the object or raise a TypeError.

default(obj)[source]

Override default method.

Parameters: obj (object) – object to be serialized
Returns: new object that is serializable (int, float, or list)
Return type: obj (object)

class lavaflow.utils.io.NumpyLineEncoder(separators=(',', ';', ':'), **kwargs)[source]

Bases: object

Class for serializing numpy arrays into a single line format for efficiency.

Note that this is intended for numeric numpy arrays with up to three axes.

Parameters

separators (list|tuple) – separators to use for each axis in order
kwargs (dict) – keyword arguments to pass to np.array2string

encode(x)[source]

Encode numpy array.

Parameters: x (np.ndarray) – numpy array
Returns: encoded numpy array
Return type: s (str)

Examples

>>> encoder = NumpyLineEncoder()
>>> x = np.array([[[1,2,3],[4,5,6],[7,8,9]],[[1,2,3],[4,5,6],[7,8,9]]])
>>> print(encoder.encode(x))
1,2,3;4,5,6;7,8,9:1,2,3;4,5,6;7,8,9

lavaflow.utils.opencv module

Utility functions for OpenCV.

lavaflow.utils.opencv.applyMask(img, mask, offset=0, radius=0, spread=0)[source]

Apply mask to image.

The mask can be dilated or eroded using cv2.morphologyEx and smoothed using cv2.GaussianBlue. A positive offset uses dilation and a negative offset uses erosion to expand or contract the mask respectively.

Parameters

img (np.ndarray) – image
mask (np.ndarray) – monocrhome mask
offset (int) – number of pixels to offset mask
radius (int) – blur radius (must be odd)
spread (float) – blur spread

Returns

image with mask applied

Return type

img (np.ndarray)

lavaflow.utils.opencv.drawPoints(vis, v, color=(255, 255, 255), points=1, line=0, loop=1, size=3, thickness=1)[source]

Draw points on an image.

Parameters

vis (np.ndarray) – visualization image
v (np.ndarray) – array of points, shaped (N, 2)
color (tuple) – color
points (int) – flag to draw points
line (int) – flag to connect points with lines
loop (int) – flag to connect end point with start point (ignored if line = 0)
size (int) – point size
thickness (int) – line thickness

Returns

visualization image

Return type

vis (np.ndarray)

lavaflow.utils.opencv.drawSegments(vis, s, color=(255, 255, 255), points=0, line=1, size=3, thickness=1)[source]

Draw line segments on an image.

Parameters

vis (np.ndarray) – visualization image
s (np.ndarray) – array of line segments, shaped (N, 2, 2)
color (tuple) – color
points (int) – flag to draw points
line (int) – flag to connect points with lines
size (int) – point size
thickness (int) – line thickness

Returns

visualization image

Return type

vis (np.ndarray)

lavaflow.utils.opencv.drawTriangles(vis, tri, color=(255, 255, 255), segment_color=(0, 0, 255), points=1, edges=1, segments=0, size=3, thickness=1)[source]

Draw triangulation on an image.

Parameters

vis (np.ndarray) – visualization image
tri (dict) – output from triangle.triangulate
color (tuple) – color
segment_color (tuple) – color for segment constraints
points (int) – flag to draw triangle points
edges (int) – flag to draw triangle edges
segments (int) – flag to draw segment constraints
size (int) – point size
thickness (int) – line thickness

Returns

visualization image

Return type

vis (np.ndarray)

lavaflow.utils.opencv.drawVelocityGrid(vis, x, y, u, v, color=(255, 255, 255), spacing=1, magnification=1, points=0, arrows=1, size=3, thickness=1, tipLength=0.05)[source]

Draw (N, M) velocity grid on an image.

Parameters

vis (np.ndarray) – visualization image
x (np.ndarray) – grid x coordinates, shaped (N, M)
y (np.ndarray) – grid y coordinates, shaped (N, M)
u (np.ndarray) – velocity horizontal components, shaped (N, M)
v (np.ndarray) – velocity vertical components, shaped (N, M)
spacing (double) – minimum spacing
magnification (float) – magnification to apply to velocity
color (tuple) – color
points (int) – flag to draw points
arrows (int) – flag to draw arrows
size (int) – point size
thickness (int) – line thickness
tipLength (int) – arrow tip length

Returns

visualization image

Return type

vis (np.ndarray)

lavaflow.utils.opencv.drawVelocityLine(vis, x, y, u, v, color=(255, 255, 255), spacing=1, magnification=1, points=0, arrows=1, size=3, thickness=1, tipLength=0.05)[source]

Draw (N, M) velocity grid on an image.

Parameters

vis (np.ndarray) – visualization image
x (np.ndarray) – line x coordinates, shaped (N)
y (np.ndarray) – line y coordinates, shaped (N)
u (np.ndarray) – velocity horizontal components, shaped (N)
v (np.ndarray) – velocity vertical components, shaped (N)
spacing (double) – minimum spacing
magnification (float) – magnification to apply to velocity
color (tuple) – color
points (int) – flag to draw points
arrows (int) – flag to draw arrows
size (int) – point size
thickness (int) – line thickness
tipLength (int) – arrow tip length

Returns

visualization image

Return type

vis (np.ndarray)

lavaflow.utils.opencv.ensureColor(img)[source]

Ensure image is BGR not GRAY.

Parameters: img (np.ndarray) – image
Returns: image converted from GRAY to BGR
Return type: img (np.ndarray)

lavaflow.utils.opencv.imgrid(array, shape=None, max_size=(2160, 3840))[source]

Function for composing multiple images into a grid.

Parameters

array (list) – list of images to compose
shape (tuple) – grid shape (rows, columns)
max_size (tuple) – resolution used to scale image grid

Returns

image grid

Return type

grid (np.ndarray)

lavaflow.utils.opencv.immask(img, mask, color=(255, 255, 0), alpha=0.5)[source]

Function to apply a transparent mask overlay to an image

Parameters

img (np.ndarray) – image
mask (np.ndarray) – mask to overlay
color (tuple) – mask color
alpha (float) – mask transparency

Returns

img with mask applied

Return type

background (np.ndarray)

lavaflow.utils.opencv.imshow(img)[source]

Wrapper for OpenCV imshow.

Parameters: img (np.ndarray) – image