Planar Purge Imaging to Separate 3D Edges From 2D Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image analysis techniques struggle to accurately identify 3D features of objects on planar surfaces due to interference from 2D features such as color, graphics, and tape, leading to erroneous robot grasping and failed operations.
Innovation Solution
A method involving a fixed camera capturing three images under different lighting conditions (ambient and supplemental) to compute an output image that eliminates 2D features by subtracting and dividing the images, retaining only 3D features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If 2D image analysis is used to identify objects, then the process is simple and fast, but 2D features like color and graphics cause misidentification of edges and corners
Solution Approach 1:
The patent transforms 2D image analysis into a pseudo-3D analysis by introducing multiple light sources at different positions. This creates depth information through shading variations, allowing the system to distinguish between 2D features (color, graphics) and 3D features (edges, corners) by analyzing how light interacts with surface geometry from multiple angles.
Solution Approach 2:
The patent introduces intermediate images captured under different lighting conditions as mediators between the original 2D image and the final analysis. These intermediate images serve as computational tools that, when combined through arithmetic operations, filter out 2D features while preserving 3D structural information.
2Measurement precision
If 3D methods like point cloud data are used to identify object shapes, then depth information is obtained, but the data is noisy and has sparse data points with drop outs
Solution Approach 1:
The patent creates multiple computational copies of the scene from different lighting perspectives without requiring physical multiple cameras or complex 3D scanning. These virtual copies are synthesized through image arithmetic, providing redundant information that compensates for noise and gaps in any single view.
3Measurement precision
If multiple light sources are used to eliminate 2D features, then color-based shading variations are removed, but the device complexity increases
Solution Approach 1:
The patent replaces complex mechanical 3D sensing systems with a simpler optical system using standard cameras and multiple light sources. The complexity is shifted from hardware to software through computational image processing, where arithmetic operations on captured images achieve the desired feature separation.
Solution Approach 2:
The patent changes the lighting parameters (position, intensity, direction) to create distinct illumination patterns that interact differently with 2D versus 3D features. By varying these parameters across multiple captures, the system encodes geometric information that can be decoded through image arithmetic.
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 output image effectively purges 2D features like color and tape, enhancing 3D features like box edges, enabling robust object identification for robotic operations.
Implementation Method 1
one image (IA) is taken with only ambient lighting, one image (I1) has ambient lighting plus a first added light source at a first position, and one image (I2) has ambient lighting plus a second added light source at a second position
Data Source
AI summary
A method and system for eliminating 2D features from planar surfaces in 2D images. Three digital images are taken, from a single camera at a fixed position, of a subject such as a pallet of boxes. One image (IA) is taken with ambient lighting, one image (I1) has ambient lighting plus a first added light source, and one image (I2) has ambient lighting plus a second added light source. An output image Q is then computed by Q=(I1−IA)/(I2−IA). Subtracting the ambient image removes ambient diffuse and specular reflections. Division eliminates all variations in the output image caused by color. The only variations that remain are those due to the angle between each surface point's normal direction and the direction from the light to that point. The output image Q, devoid of all colors and 2D features, is well suited for computing a robot grasp of an object in the image.