## Point Cloud Auto-Registration ##

### Function Overview ###

Point Cloud Auto-Registration registers a source point cloud to a reference point cloud automatically. The tool first extracts registration features from the two point clouds, estimates an initial transformation, and can optionally refine the result with ICP optimization.

### Usage ###

Click *Data Management > Data Registration > Point Cloud Auto-Registration* to open the tool.

<div align=center>
<img src="../../../ScreenShot/en/DataManagement/Transform/AutoPointCloudRegistration.png" alt="Automatic Point Cloud Registration" width="480" style='margin-left:24%;'/>
</div>

### Parameter Settings ###

* **Reference Point Cloud File**: Reference point cloud used as the registration target.

* **Registration Point Cloud File**: Point cloud to be transformed and registered to the reference point cloud.

#### Preprocessing Parameters ####

* **Downsampling Voxel Size**: Voxel size used for point cloud downsampling before feature extraction.

* **Maximum Number of Key Points**: Maximum number of key points extracted from the point cloud.

#### ICP Optimization ####

* **Enable ICP Optimization**: Refine the automatic registration result with ICP.

* **Select Class**: Point cloud class used in ICP optimization.

* **Registration Type**: ICP registration type. Point-to-point and point-to-plane registration are supported.

* **Sampling Point Limit**: Maximum number of points sampled for ICP optimization.

* **Maximum Point Search Distance**: Maximum correspondence search distance. The default value is 1 m.

* **Overlap Threshold**: Minimum overlap threshold used during ICP optimization. The default value is 40%.

* **Normal Filtering Threshold**: Normal filtering threshold used for point-to-plane ICP. The default value is 60%.

### Output ###

The tool outputs the registered point cloud data and the corresponding transformation result.