How Can We Help?

Prescan Function

You are here:

The Prescan feature in Arago Control software leverages the Intel RealSense Depth Camera D415 mounted on an Arago Photogrammetry rig. It moves along 3 axes and streams a volumetric view in real-time, allowing users to create a quick 3D scan of an object. This proxy 3D model is then used to determine optimal camera positions for photogrammetric capture. Below, we address common issues and best practices to ensure successful use of the Prescan feature.


Why is my Prescan not working properly?

 

1. Hardware Requirements Not Met

 

Ensure your system meets the necessary hardware requirements. The Prescan process involves real-time depth frame streaming and 3D reconstruction, which can be computationally intensive. Slower computers may drop frames or lose tracking, leading to a lower-quality 3D model. Check the minimum hardware requirements here.


2. Unstable USB Connection

 

Verify that the Intel RealSense Depth Camera is connected via a stable USB 3.1 powered connection. An unstable connection can slow down the depth frame transfer, impacting Prescan performance.


3. Why are certain objects difficult to scan?

 

Some objects pose challenges for the Prescan sensor, including:

  • Highly symmetrical objects

  • Very shiny or transparent surfaces

  • Featureless objects (single flat color)

  • Small objects (under 5-10 cm in size) 

4. Solutions for Scanning Problematic Objects

 

  • Add Features for Better Tracking
    Place the object on a feature-rich surface, such as a box with random markings and high-contrast points. Prescan both the object and the surface together. Use the “Crop using bounding box” tool in the Prescan Post-processing panel to digitally remove the assisting object after the scan. Learn more about cropping here.
    In case the assisting box/objects is not required for the capture step, it can be removed. Keep in mind, that this would change the location of your object and the position of the prescanned 3D model would have to be adjusted to match the new location.

  • Slow Down the Turntable Speed
    Reducing the turntable speed (e.g., to half-speed or less) allows the sensor to gather more depth frames, improving tracking and overall 3D model quality.

  • Optimal Sensor Placement
    Ensure the camera is not too close or too far from the object. The object should not be cropped out of the field of view (FOV) and should occupy a substantial portion of the frame without being too small.

  • Reconstruction Region
    Define the reconstruction region carefully. Ensure it does not crop the object during turntable rotation. A cube-shaped region larger than the object is recommended, as empty space around the object will not affect the reconstruction process. 

5. How can I address exposure issues during Prescan?

 
Set Depth Sensor Exposure Region of Interest.
High contrast between the object and its background (e.g., a white object against a black background) can cause over- or underexposure. Use the “RGB Camera” live view panel to define a custom exposure region that ensures proper illumination of the object.
 

6. Best Practices Summary

 

  • Confirm hardware requirements are met.

  • Use a stable USB 3.1 powered connection.

  • Place challenging objects on feature-rich surfaces to aid tracking.

  • Adjust turntable speed to improve depth frame capture.

  • Frame the object properly, ensuring it occupies an optimal portion of the FOV.

  • Define the reconstruction region to fully encompass the object.

  • Adjust exposure settings for better performance in high-contrast scenarios

Table of Contents