Hybrid Sensor Structured Light Depth Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3D point cloud data acquisition methods, such as laser scanning and moiré fringe techniques, face limitations including increased cycle time due to motion requirements, extensive calibration processes, and sensing issues with surface discontinuities, necessitating an improved sensor system for accurate feature analysis.
Innovation Solution
A sensor system comprising a first laser source projecting a laser line onto a feature, a sensor imaging the laser line, and a second laser source projecting a pattern intersecting the laser line, utilizing precalibrated relationships between sensors and laser sources to determine feature position and orientation within a contiguous sensing volume.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laser scanning is used to acquire 3D point cloud data, then reliable depth information is obtained, but cycle time increases due to motion requirements
Solution Approach 1:
The patent replaces the mechanical motion system of traditional laser scanners with a static sensor system that uses structured light projection. Instead of moving the laser scanner to scan different areas, the system projects laser lines and uses image sensors to capture depth information simultaneously across the entire field of view, eliminating motion-related cycle time while maintaining depth measurement capability through optical triangulation
2Productivity
If photogrammetric targets are used for calibration, then 3D data can be obtained without sensor translation, but extensive calibration is required
Solution Approach 1:
The patent extracts and removes the photogrammetric target calibration step entirely from the measurement process. By using a static sensor configuration with pre-calibrated intrinsic parameters and a known geometric relationship between the sensor and laser sources, the system achieves immediate 3D measurement capability without requiring external calibration targets or complex calibration procedures
Solution Approach 2:
The system performs self-calibration through the geometric relationship between the sensor and laser sources. The intrinsic parameters of the sensor and the known configuration of the laser projection system allow the system to determine depth information directly from the captured laser line patterns without external intervention or calibration objects
3Measurement precision
If multiple images are used to solve for absolute depth, then depth information can be obtained, but calibration becomes extensive
Solution Approach 1:
The patent performs preliminary calibration of the sensor and laser source relationship during manufacturing, storing the intrinsic parameters in the sensor's memory. This pre-calibration eliminates the need for extensive field calibration and allows the system to compute absolute depth directly from the captured laser line patterns using the pre-stored geometric parameters
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
The system enhances measurement efficiency by reducing cycle time, improving calibration accuracy, and overcoming sensing challenges with surface discontinuities, enabling precise feature analysis and increased depth sensitivity through triangulation and structured light principles.
Implementation Method 1
The first laser source projects a laser line into the sensing volume and onto the feature forming a laser stripe on the feature
Implementation Method 2
The sensor images the laser stripe where the laser line intersects with the feature
Implementation Method 3
The second laser source projects a pattern onto the feature such that the pattern intersects the laser stripe on the feature
Data Source
AI summary
A sensor system and method for analyzing a feature in a sensing volume. The system projecting a pattern onto the feature and imaging the pattern where the pattern intersects with the feature, where the pattern is a series of lines that are encoded to identify at least one line of the series of lines.