robokudo.descriptors.analysis_engines.regionfilter_from_storage

Analysis engine for region-based filtering using world descriptors.

This module provides an analysis engine that demonstrates how to filter point cloud data based on predefined regions from world descriptors. It processes stored camera data and applies region-based filtering to focus on specific areas of interest.

The pipeline implements the following functionality:

  • Reading stored camera data from MongoDB

  • Image preprocessing

  • Region-based filtering using world descriptor data

  • Optional pipeline trigger for step-by-step execution

  • Optional camera viewpoint visualization

Note

This engine requires properly configured world descriptors with defined regions of interest. The regions are used to filter the point cloud data during processing.

Classes

AnalysisEngine

Analysis engine for region-based point cloud filtering.

Module Contents

class robokudo.descriptors.analysis_engines.regionfilter_from_storage.AnalysisEngine

Bases: robokudo.analysis_engine.AnalysisEngineInterface

Analysis engine for region-based point cloud filtering.

This class implements a pipeline that filters point cloud data based on predefined regions from world descriptors. It processes stored camera data and applies region filtering to focus on specific areas of interest.

The pipeline includes:

  • Collection reader for stored data access

  • Image preprocessing

  • Region-based filtering

  • Optional pipeline trigger

  • Optional viewpoint visualization

Note

The pipeline can be configured to run continuously or with step-by-step execution using the pipeline trigger. Viewpoint visualization can be enabled for debugging purposes.

name() str

Get the name of the analysis engine.

Returns:

The name identifier of this analysis engine

implementation() robokudo.pipeline.Pipeline

Create a pipeline for region-based point cloud filtering.

This method constructs a processing pipeline that applies region-based filtering to point cloud data. The regions are defined in world descriptors and used to filter the data during processing.

Pipeline execution sequence:

  1. Initialize pipeline

  2. Read stored camera data

  3. Bootstrap world descriptor entities into the shared world

  4. Preprocess image

  5. Apply region filter

  6. Optional: Visualize camera viewpoint

Returns:

The configured pipeline for region-based filtering

Note

The pipeline includes commented-out options for adding a trigger and camera viewpoint visualization, which can be useful for debugging and development.