VolumeAnnotation¶
- class VolumeAnnotation[source]¶
Bases:
objectVolumeAnnotation for a single volume.
VolumeAnnotationobject is immutable.- Parameters
- volume_meta : dict
Metadata of the volume.
- objects : VolumeObjectCollection, optional
VolumeObjectCollection object.
- plane_sagittal :
Plane, optional Sagittal plane of the volume.
- plane_coronal :
Plane, optional Coronal plane of the volume.
- plane_axial :
Plane, optional Axial plane of the volume.
- tags : VolumeTagCollection, optional
VolumeTagCollection object.
- spatial_figures : List[VolumeFigure], optional
List of spatial figures associated with the volume.
- key : UUID, optional
UUID object.
- Usage example
import supervisely as sly # Simple VolumeAnnotation example path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) print(volume_ann.to_json()) # Output: { # { # "key": "56107223943346e5900fc256b8dcd7f0", # "objects": [], # "planes": [ # { "name": "sagittal", "normal": { "x": 1, "y": 0, "z": 0 }, "slices": [] }, # { "name": "coronal", "normal": { "x": 0, "y": 1, "z": 0 }, "slices": [] }, # { "name": "axial", "normal": { "x": 0, "y": 0, "z": 1 }, "slices": [] } # ], # "spatialFigures": [], # "tags": [], # "volumeMeta": { # "ACS": "RAS", # "channelsCount": 1, # "dimensionsIJK": { "x": 512, "y": 512, "z": 139 }, # "directions": [1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0], # "intensity": { "max": 3071.0, "min": -3024.0 }, # "origin": [-194.238403081894, -217.5384061336518, -347.7500000000001], # "rescaleIntercept": 0, # "rescaleSlope": 1, # "spacing": [0.7617189884185793, 0.7617189884185793, 2.5], # "windowCenter": 23.5, # "windowWidth": 6095.0 # } # } # More complex VolumeAnnotation example path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) # VolumeObjectCollection obj_class = sly.ObjClass('brain', sly.Rectangle) volume_obj = sly.VolumeObject(obj_class) objects = sly.VolumeObjectCollection([volume_obj]) # VolumeTagCollection brain_meta = sly.TagMeta('brain_tag', sly.TagValueType.ANY_STRING) from supervisely.volume_annotation.volume_tag import VolumeTag vol_tag = VolumeTag(brain_meta, value='human') from supervisely.volume_annotation.volume_tag_collection import VolumeTagCollection volume_tags = VolumeTagCollection([vol_tag]) volume_ann = sly.VolumeAnnotation(volume_meta, objects, volume_tags) print(volume_ann.to_json()) # Output: # { # "key": "4d4bb69e6fcd40e1a1cb076c07769903", # "objects": [ # { # "classTitle": "brain", # "key": "22e1082a17f74279b00eed0bfb0ba11d", # "tags": [] # } # ], # "planes": [ # { "name": "sagittal", "normal": { "x": 1, "y": 0, "z": 0 }, "slices": [] }, # { "name": "coronal", "normal": { "x": 0, "y": 1, "z": 0 }, "slices": [] }, # { "name": "axial", "normal": { "x": 0, "y": 0, "z": 1 }, "slices": [] } # ], # "spatialFigures": [], # "tags": [ # { # "key": "b9de6631d328441796119b4b0039fc61", # "name": "brain_tag", # "value": "human" # } # ], # "volumeMeta": { # "ACS": "RAS", # "channelsCount": 1, # "dimensionsIJK": { "x": 512, "y": 512, "z": 139 }, # "directions": [1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0], # "intensity": { "max": 3071.0, "min": -3024.0 }, # "origin": [-194.238403081894, -217.5384061336518, -347.7500000000001], # "rescaleIntercept": 0, # "rescaleSlope": 1, # "spacing": [0.7617189884185793, 0.7617189884185793, 2.5], # "windowCenter": 23.5, # "windowWidth": 6095.0 # } # }
Methods
Add new objects to a VolumeAnnotation object.
Add new VolumeTags to a VolumeAnnotation object.
Makes a copy of VolumeAnnotation with new fields, if fields are given, otherwise it will use fields of the original VolumeAnnotation.
Save the VolumeAnnotation to a json file.
Convert a json dict to VolumeAnnotation.
Check whether volume annotation contains objects or tags, or not.
Volume annotation key value.
Remove annotation objects from a VolumeAnnotation object.
Remove VolumeTags from a VolumeAnnotation object.
Convert the VolumeAnnotation to a json dict.
Checks if all slices in each plane contains figures.
Attributes
VolumeFigure objects.
VolumeAnnotation objects.
Axial plane of the volume.
Coronal plane of the volume.
Sagital plane of the volume.
Get a list of spatial figures.
VolumeTag objects.
Volume meta data.
- add_objects(objects)[source]¶
Add new objects to a VolumeAnnotation object.
- Parameters
- objects : List[VolumeObject] or VolumeObjectCollection
New volume objects.
- Returns
A VolumeAnnotation object containing the original and new volume objects.
- Return type
- Usage example
import os from dotenv import load_dotenv import supervisely as sly path = "/vol_01.nrrd" _, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) obj_class_heart = sly.ObjClass('heart', sly.Mask3D) volume_obj_heart = sly.VolumeObject(obj_class_heart) volume_ann = volume_ann.add_objects([volume_obj_heart])
- add_tags(tags)[source]¶
Add new VolumeTags to a VolumeAnnotation object.
- Parameters
- tags : List[VolumeTag] or VolumeTagCollection
New VolumeTags.
- Returns
A VolumeAnnotation object containing the original and new volume tags.
- Return type
- Usage example
import supervisely as sly path = "/vol_01.nrrd" _, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) brain_meta = sly.TagMeta('brain_tag', sly.TagValueType.ANY_STRING) vol_tag = sly.VolumeTag(brain_meta, value='human') volume_ann = volume_ann.add_tags([vol_tag])
-
clone(volume_meta=
None, objects=None, plane_sagittal=None, plane_coronal=None, plane_axial=None, tags=None, spatial_figures=None)[source]¶ Makes a copy of VolumeAnnotation with new fields, if fields are given, otherwise it will use fields of the original VolumeAnnotation.
- Parameters
- volume_meta : dict
Metadata of the volume.
- objects : VolumeObjectCollection, optional
VolumeObjectCollection object.
- plane_sagittal :
Plane, optional Sagittal plane of the volume.
- plane_coronal :
Plane, optional Coronal plane of the volume.
- plane_axial :
Plane, optional Axial plane of the volume.
- tags : VolumeTagCollection, optional
VolumeTagCollection object.
- spatial_figures : List[VolumeFigure], optional
List of spatial figures associated with the volume.
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) obj_class_heart = sly.ObjClass('heart', sly.Rectangle) video_obj_heart = sly.VolumeObject(obj_class_heart) new_objects = sly.VolumeObjectCollection([volume_obj_heart]) new_volume_ann = volume_ann.clone(objects=new_objects)
-
dump_json(path, key_id_map=
None)[source]¶ Save the VolumeAnnotation to a json file.
- Parameters
- Returns
None
- Return type
NoneType- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) volume_ann.dump_json("/home/admin/work/volumes/vol_01.json")
-
classmethod from_json(data, project_meta, key_id_map=
None)[source]¶ Convert a json dict to VolumeAnnotation.
- Parameters
- data : dict
Volume annotation in json format as a dict.
- project_meta : ProjectMeta
Input
ProjectMeta.- key_id_map : KeyIdMap, optional
KeyIdMap object.
- Returns
VolumeAnnotation object
- Return type
- Usage example
import supervisely as sly from supervisely.video_annotation.key_id_map import KeyIdMap meta = sly.ProjectMeta() key_id_map = KeyIdMap() ann_json = { "key": "56107223943346e5900fc256b8dcd7f0", "objects": [], "planes": [ { "name": "sagittal", "normal": { "x": 1, "y": 0, "z": 0 }, "slices": [] }, { "name": "coronal", "normal": { "x": 0, "y": 1, "z": 0 }, "slices": [] }, { "name": "axial", "normal": { "x": 0, "y": 0, "z": 1 }, "slices": [] } ], "spatialFigures": [], "tags": [], "volumeMeta": { "ACS": "RAS", "channelsCount": 1, "dimensionsIJK": { "x": 512, "y": 512, "z": 139 }, "directions": [1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0], "intensity": { "max": 3071.0, "min": -3024.0 }, "origin": [-194.238403081894, -217.5384061336518, -347.7500000000001], "rescaleIntercept": 0, "rescaleSlope": 1, "spacing": [0.7617189884185793, 0.7617189884185793, 2.5], "windowCenter": 23.5, "windowWidth": 6095.0 } } ann = sly.VolumeAnnotation.from_json(ann_json, meta, key_id_map)
- is_empty()[source]¶
Check whether volume annotation contains objects or tags, or not.
- Returns
True if volume annotation is empty, False otherwise.
- Return type
- Usage exmaple
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) is_empty = vol_ann.is_empty()
- key()[source]¶
Volume annotation key value.
- Returns
Key value of VolumeAnnotation object.
- Return type
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) key = vol_ann.key()
- remove_objects(keys)[source]¶
Remove annotation objects from a VolumeAnnotation object.
- Parameters
- Returns
A VolumeAnnotation object containing the original volume objects without the removed objects.
- Return type
- Usage example
import supervisely as sly path = "/vol_01.nrrd" _, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) obj_class_heart = sly.ObjClass('heart', sly.Mask3D) volume_obj_heart = sly.VolumeObject(obj_class_heart) obj_class_tumor = sly.ObjClass('tumor', sly.Mask3D) volume_obj_tumor = sly.VolumeObject(obj_class_tumor) volume_ann = volume_ann.add_objects([volume_obj_heart, volume_obj_tumor]) volume_ann = volume_ann.remove_objects(volume_obj_heart.key())
- remove_tags(keys)[source]¶
Remove VolumeTags from a VolumeAnnotation object.
- Parameters
- Returns
A VolumeAnnotation object containing the original VolumeTags without the removed VolumeTags.
- Return type
- Usage example
import supervisely as sly path = "/vol_01.nrrd" _, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) brain_meta = sly.TagMeta('brain_tag', sly.TagValueType.ANY_STRING) vol_tag = sly.VolumeTag(brain_meta, value='human') volume_ann = volume_ann.add_tags([vol_tag]) volume_ann = volume_ann.remove_tags(vol_tag.key())
-
to_json(key_id_map=
None)[source]¶ Convert the VolumeAnnotation to a json dict.
- Parameters
- key_id_map : KeyIdMap, optional
KeyIdMap object.
- Returns
Volume annotation in json format as a dict.
- Return type
- Usage example
import supervisely as sly from supervisely.video_annotation.key_id_map import KeyIdMap path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) volume_ann = sly.VolumeAnnotation(volume_meta) print(volume_ann.to_json()) # Output: { # { # "key": "56107223943346e5900fc256b8dcd7f0", # "objects": [], # "planes": [ # { "name": "sagittal", "normal": { "x": 1, "y": 0, "z": 0 }, "slices": [] }, # { "name": "coronal", "normal": { "x": 0, "y": 1, "z": 0 }, "slices": [] }, # { "name": "axial", "normal": { "x": 0, "y": 0, "z": 1 }, "slices": [] } # ], # "spatialFigures": [], # "tags": [], # "volumeMeta": { # "ACS": "RAS", # "channelsCount": 1, # "dimensionsIJK": { "x": 512, "y": 512, "z": 139 }, # "directions": [1.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 1.0], # "intensity": { "max": 3071.0, "min": -3024.0 }, # "origin": [-194.238403081894, -217.5384061336518, -347.7500000000001], # "rescaleIntercept": 0, # "rescaleSlope": 1, # "spacing": [0.7617189884185793, 0.7617189884185793, 2.5], # "windowCenter": 23.5, # "windowWidth": 6095.0 # } # }
- validate_figures_bounds()[source]¶
Checks if all slices in each plane contains figures.
- Raises
OutOfImageBoundsException, if figure is out of slices images bounds- Returns
None
- Return type
NoneType- Usage Example
import supervisely as sly plane_axial = sly.Plane(sly.Plane.AXIAL, frames, volume_meta=volume_meta) volume_ann = sly.VolumeAnnotation(volume_meta, objects, plane_axial=plane_axial) volume_ann.validate_figures_bounds()
- property figures¶
VolumeFigure objects.
- Returns
List of VolumeFigure objects from VolumeAnnotation object.
- Return type
List[VolumeFigure]
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) figures = vol_ann.figures
- property objects¶
VolumeAnnotation objects.
- Returns
VolumeObjectCollection object
- Return type
- Usage example
import supervisely as sly path = "/Users/Downloads/volumes/Demo volumes_ds1_CTChest.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) # VolumeObjectCollection obj_class_heart = sly.ObjClass('heart', sly.Rectangle) volume_obj_heart = sly.VolumeObject(obj_class_heart) objects = sly.VolumeObjectCollection([volume_obj_heart]) volume_ann = sly.VolumeAnnotation(volume_meta, objects) print(volume_ann.objects.to_json()) # Output: [ # { # "key": "2b5d70baa5a74d06a525b950b5f2b756", # "classTitle": "heart", # "tags": [] # } # ]
- property plane_axial¶
Axial plane of the volume.
- Returns
Axial plane of the volume.
- Return type
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) plane_axial = vol_ann.plane_axial
- property plane_coronal¶
Coronal plane of the volume.
- Returns
Coronal plane of the volume.
- Return type
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) plane_coronal = vol_ann.plane_coronal
- property plane_sagittal¶
Sagital plane of the volume.
- Returns
Sagittal plane of the volume.
- Return type
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) plane_sagittal = vol_ann.plane_sagittal
- property spatial_figures¶
Get a list of spatial figures.
- Returns
List of spatial figures from VolumeAnnotation object.
- Return type
List[VolumeFigure]
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) spatial_figures = vol_ann.spatial_figures
- property tags¶
VolumeTag objects.
- Returns
VolumeTagCollection
- Return type
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) tags = vol_ann.tags
- property volume_meta¶
Volume meta data.
- Returns
Sagittal plane of the volume.
- Return type
- Usage example
import supervisely as sly path = "/home/admin/work/volumes/vol_01.nrrd" volume, volume_meta = sly.volume.read_nrrd_serie_volume_np(path) vol_ann = sly.VolumeAnnotation(volume_meta) volume_meta = vol_ann.volume_meta