robokudo.descriptors.analysis_engines.demo

Analysis engine demonstrating basic tabletop segmentation.

This module provides a basic analysis engine that demonstrates tabletop segmentation using a Kinect camera. It implements a straightforward pipeline for processing point cloud data to identify objects on a table surface.

The pipeline implements the following functionality: - Reading data from a Kinect camera (without transform lookup) - Image preprocessing - Point cloud cropping - Plane detection (table surface) - Point cloud cluster extraction (objects)

Note

This is a basic demonstration pipeline that can be used as a starting point for more complex object detection and segmentation tasks.

Classes

AnalysisEngine

Analysis engine for basic tabletop segmentation.

Module Contents

class robokudo.descriptors.analysis_engines.demo.AnalysisEngine

Bases: robokudo.analysis_engine.AnalysisEngineInterface

Analysis engine for basic tabletop segmentation.

This class implements a simple pipeline for tabletop segmentation using a Kinect camera. It processes point cloud data to identify and segment objects on a table surface.

The pipeline includes: - Collection reader for Kinect camera data - Image preprocessing - Point cloud cropping - Plane detection - Point cloud cluster extraction

Note

The pipeline uses the Kinect configuration without transform lookup for simplicity. For more advanced applications, consider using the version with transform lookup enabled.

name() → str

Get the name of the analysis engine.

Returns:

The name identifier of this analysis engine

implementation() → robokudo.pipeline.Pipeline

Create a basic pipeline for tabletop segmentation.

This method constructs a processing pipeline that performs tabletop segmentation using point cloud data from a Kinect camera. The pipeline processes the data through several stages to identify objects on a table surface.

The pipeline execution sequence is: 1. Initialize pipeline 2. Read frame from Kinect 3. Preprocess image 4. Crop point cloud to region of interest 5. Detect table plane 6. Extract object clusters

Returns:

The configured pipeline for tabletop segmentation

Note

The pipeline includes commented-out options for adding triggers and slow processing simulation, which can be useful for debugging.