Laser Scanner Image-Guided Scan Settings for Point Cloud Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing terrestrial laser scanners face challenges with high data volume processing, registration errors due to moving objects and reflective surfaces, and inconsistent color information due to light conditions, leading to inefficient and inaccurate 3D scanning.
Innovation Solution
A terrestrial laser scanner with a camera that uses machine learning to categorize scan regions based on image significance, adjusting scan parameters such as density, speed, and radiation intensity in real-time to enhance scanning efficiency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the laser scanner performs continuous scanning of the entire scan sphere, then complete spatial coverage is achieved, but the data volume becomes excessively large and processing time increases
Solution Approach 1:
The scan sphere is divided into multiple regions based on camera image analysis. The processing unit identifies regions of interest (ROIs) where objects are detected and segments the scanning task to focus only on these relevant areas, excluding background regions from detailed scanning.
Solution Approach 2:
A camera captures an image of the scan sphere before the laser scanning begins. This preliminary image is processed to identify objects and their locations, allowing the scanning system to pre-determine which regions require detailed measurement and which can be skipped.
2Ease of operation
If the laser scanner uses fixed scan parameters, then the scanning process is simple to control, but scanning efficiency varies and registration errors occur due to moving objects
Solution Approach 1:
The scan parameters (such as scan density, speed, and radiation intensity) are made dynamic and adaptable. The processing unit modifies these parameters in real-time based on the detected scene content, object types, and potential moving objects, allowing the system to optimize scanning efficiency for each specific situation.
Solution Approach 2:
The system incorporates feedback mechanisms where the camera image analysis continuously informs the scanning process. The processing unit uses the detected object information to adjust scan parameters, creating a closed-loop system that adapts to changing conditions and reduces registration errors.
3Reliability
If the laser scanner scans all regions with uniform density, then measurement completeness is maintained, but processing time increases due to unnecessary points
Solution Approach 1:
The scanning system applies different scan densities to different regions of the scan sphere. Regions containing detected objects receive high-density scanning for complete measurement, while background regions are scanned with low density or skipped entirely, optimizing the balance between measurement completeness and processing efficiency.
4Quantity of substance
If the laser scanner processes all detected scan points, then data completeness is maintained, but registration errors increase due to moving objects and reflective surfaces
Solution Approach 1:
The system extracts and identifies problematic scan points corresponding to moving objects and reflective surfaces based on camera image analysis. These extracted points are then excluded from the final point cloud processing, preventing them from causing registration errors while maintaining the integrity of stable, non-reflective objects.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves scanning efficiency by reducing unnecessary data points, minimizing registration errors, and ensuring consistent colorization, resulting in a more accurate and optimized 3D point cloud generation.
Implementation Method 1
a radiation source or emitter for generating optical measurement radiation, often laser radiation
Implementation Method 2
Distance determination is based on reflected measurement radiation, reflected from the irradiated target object so that at least a part of the measuring radiation is reflected back to the laser scanner and detected
Implementation Method 3
the scanner has a receiver embodied as an optoelectronic sensor, which is designed for time-resolved detection of the reflected measuring radiation
Implementation Method 4
The measurement beam of the distance measuring device is deflected in one spatial direction by rotating aforementioned deflector about one axis (elevation axis) and-independent therefrom-in another spatial direction by rotation of a top part or body
Data Source
AI summary
A stationary terrestrial laser scanner and method with image processing, based on a machine learning algorithm, of a 2D-image of a scan sphere, captured with a camera of the stationed laser scanner before measurement of scan points, in such a way that the image is partitioned in multiple clusters of different predefined categories of measurement significance and setting at least one adaptable scan parameter according to the presence and/or absence of clusters of a significance category in the first 2D-image.