Laser Scanner Image Segmentation for Precise Scan Area Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for defining scanning areas in laser-based measuring devices are either imprecise with bounding boxes or cumbersome with manual polygon selection, leading to unnecessary scanning of background areas and increased data volume.
Innovation Solution
An image-based method using machine learning algorithms to automatically segment an area of interest based on visual coherence, allowing precise definition of scanning areas by selecting a pixel group and transforming image coordinates into targeting axis angles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual polygon selection is used to define scanning area, then scanning precision is improved, but ease of operation deteriorates
Solution Approach 1:
The system performs automatic segmentation of the area of interest using machine learning algorithms, allowing the system to define the scanning area autonomously based on image analysis rather than requiring manual user input for each boundary point
Solution Approach 2:
The manual mechanical process of selecting polygon vertices is replaced by an automated image processing system using machine learning algorithms that automatically identify and segment the area of interest from captured images
2Ease of operation
If bounding box method is used to define scanning area, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The system segments the area of interest from the surrounding background using machine learning-based image processing, automatically identifying precise boundaries of the target object rather than using a crude bounding box that includes unnecessary background areas
Solution Approach 2:
The system autonomously performs image analysis and automatic segmentation to define the scanning area boundaries, eliminating the need for manual adjustment while achieving precision beyond simple bounding boxes
3Measurement precision
If manual polygon selection is used to define scanning area, then scanning precision is improved, but productivity deteriorates
Solution Approach 1:
The system automatically segments the area of interest using machine learning algorithms, performing the precision-defining task autonomously without requiring time-consuming manual user input, thus maintaining high precision while significantly improving operational speed
Solution Approach 2:
The time-consuming manual process of selecting polygon vertices is replaced by automated image processing and machine learning-based segmentation, achieving both precision and speed through computational methods
Data Source
Figure 1~2d
Figure 3~5b
Figure 6a~6b
AI summary
Image-based method of defining a scanning area with a measuring radiation-based measuring device (10) such as a laser scanner, the method comprising roughly aligning a targeting axis (2) of the measuring device (10) towards an area of interest (1) to be scanned, the area of interest (1) having a coherent visual property, capturing with known orientation to the targeting axis (2) an image (3) covering the area of interest (1), displaying the image (3) to a user, manually selecting by the user of at least one pixel (p) of a group of pixels (7, 7s) representing the area of interest (1) in the image (3), automatically segmenting the group of pixels (7s) based on a similarity to the at least one selected pixel (p) using a machine learning algorithm, and automatically defining the scanning area based on border pixels (9i) of the first group of pixels (7s).