Documentation of QGIS Python plugin brdrQ - AutoUpdateBorders - GRB Updater (Flanders-specific dataset)

Description

AutoUpdateBorders is a QGIS-processing Script to auto-update geometries that are aligned to an old GRB-referencelayer (Flanders) to a newer GRB-referencelayer. It makes predictions to find the resulting aligned geometry. If multiple predictions are found the user can choose to return ALL predictions, the BEST prediction, or the ORIGINAL geometry (for further analysis).

Parameters

Input Parameters

• Thematic Layer: A vector layer (polygon, line, or point) with EPSG:31370 or EPSG:3812 coordinates and a unique ID.

  • Default: No default value, must be provided by the user.
  • Optional: No.

• Thematic ID: Textual or numeric ID of the thematic layer used as a reference to the objects. This must be unique.

  • Default: No default value, must be provided by the user.
  • Optional: No.

• Reference Layer: Combobox to choose which GRB referencelayer will be used (on-the-fly download).

  • The on-the-fly downloads are only possible for smaller areas.

    • ADP: (on-the-fly download) - Actual administrative parcels from GRB (Grootschalig Referentie Bestand)
    • GBG: (on-the-fly download) - Actual buildings from GRB
    • KNW: (on-the-fly download) - Actual artwork from GRB
    • (Note: the on-the-fly downloads are only possible for a subset or small area of thematic objects as this results in downloading this reference-area.)

  • Default: ADP (parcels)

  • Optional: no

• Relevant Distance (meters): Positive (decimal) number in meters. This is the maximum distance by which the original boundary is maximally shifted to align with the reference layer when searching for predictions. The algorithm uses all relevant distances from 0 to ‘max’ with steps of 10cm in between.

  • Default: 3 (meters)
  • Optional: No.

• PREDICTION_STRATEGY : Strategy when multiple predictions are available:

  • ALL: All predictions will be added to the result. It is not guaranteed anymore that the ID is unique anymore in the result.

  • BEST: Only the prediction with the best prediction_score will be added to the result (ID is unique)

  • ORIGINAL: When multiple predictions are found, the original geometry will be added to the result . The ID stays unique. This strategy can be usefull in combination with the featurealigner, so the multiple predictions can be handled one-by-one.

  • Default: BEST

  • Optional: No.

ADVANCED INPUT PARAMETERS

• FULL_REFERENCE_STRATEGY: Choice that determines the score of predictions based on full overlap with reference polygons:
  • ONLY_FULL: Only predictions with a full reference overlap are shown.
  • PREFER_FULL: Predictions with a full reference overlap get a higher score, others are still shown.
  • NO_FULL: No distinction is made between predictions with or without full reference overlap.
  • Default: None
  • Optional: yes.
• ENUM_OD_STRATEGY: Strategy for handling parts of the geometry not covered by reference features (= Open Domain).
  • Default: SNAP_ALL_SIDE
  • Optional: yes.
• ENUM_SNAP_STRATEGY: Strategy for snapping to reference vertices. Options: NO_PREFERENCE, PREFER_VERTICES, ONLY_VERTICES. This parameter only applies to NetworkGeometryProcessor and SnapGeometryProcessor, typically for line and point workflows. For polygon workflows this setting has no effect.
  • Default: PREFER_VERTICES
  • Optional: yes.
• ENUM_PROCESSOR: Choice of processing algorithm. Recommended is AlignerGeometryProcessor (wrapper that chooses the fastest algorithm based on geometry types).
  • Default: AlignerGeometryProcessor
  • Optional: yes.
• THRESHOLD_OVERLAP_PERCENTAGE (0-100): Fallback overlap threshold used in edge cases where relevance cannot be determined.
  • Default: 50
  • Optional: yes.
• REVIEW_PERCENTAGE (0-100): Results with change percentage above this threshold are categorized for review.
  • Default: 10
  • Optional: yes.
• WORKING FOLDER: Folder to save the resulting geojson-files. By default empty, resulting in saving the geojson-files in a default folder on your machine.
  • Default: By default, the output will be saved in a local folder on your machine.
  • Optional: yes.
• METADATA_FIELD: Attribute field of the thematic layer containing former brdr_metadata. When provided, it is used for better prediction quality.
  • Default: None
  • Optional: yes.
• LOG_INFO (default False): When True, extra logging from brdr is written to a file (brdr_show_log_info.log) in the selected WORK_FOLDER.
  • Default: False
  • Optional: No.

Output Parameters

The script generates several output layers in the layer overview, combined into a group layer:

• brdrQ_RESULT: resulting geometries after alignment
• brdrQ_DIFF: differences (+ and -) between original and resulting geometry
• brdrQ_DIFF_PLUS: differences (+) between original and resulting geometry
• brdrQ_DIFF_MIN: differences (-) between original and resulting geometry

Example of Usage

Here is an example of how to use the script in Python:

output = processing.run(
    "brdrqprovider:brdrqautoupdateborders",
    {
        "INPUT_THEMATIC": themelayername,
        "COMBOBOX_ID_THEME": "theme_identifier",
        "ENUM_REFERENCE": 0,
        "METADATA_FIELD": "",
        "RELEVANT_DISTANCE": 5,
        "WORK_FOLDER": "",
        "ENUM_OD_STRATEGY": 2,
        "ENUM_SNAP_STRATEGY": 1,
        "ENUM_PROCESSOR": 0,
        "THRESHOLD_OVERLAP_PERCENTAGE": 50,
        "REVIEW_PERCENTAGE": 10,
        "PREDICTION_STRATEGY": 2,
        "FULL_REFERENCE_STRATEGY": 2,
        "LOG_INFO": True,
    },
)