Coded Light Range Camera Depth Reconstruction via Maximum Likelihood Binarization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional stereo cameras face computational intensity and depth blindness issues in determining object depth due to correspondence processing requirements, especially in featureless scenes or with varying lighting conditions, which limits real-time image processing and increases costs.
Innovation Solution
Implementing a coded light range camera system with an infrared projector and RGB camera that projects one-dimensional binary patterns, allowing for active stereo imagery and triangulation without correspondence calculations, using maximum likelihood image binarization to improve depth reconstruction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional stereo cameras use correspondence processing to determine depth, then depth information can be obtained, but computational intensity increases and real-time processing is eliminated
Solution Approach 1:
The system projects coded light patterns (binary sequences) onto the scene before capture, encoding depth information directly into the illumination. This preliminary encoding eliminates the need for post-capture correspondence matching, enabling real-time depth extraction while maintaining accuracy.
Solution Approach 2:
The patent replaces the computational correspondence matching process with an optical encoding system. By using coded light projection and direct pattern recognition, the system substitutes complex image processing algorithms with simpler optical-mechanical depth extraction methods.
2Measurement precision
If conventional stereo cameras perform correspondence processing, then depth can be determined, but additional computing hardware is required which increases cost
Solution Approach 1:
The system replaces complex computing hardware with a simpler optical encoding and decoding mechanism. The coded light patterns carry depth information that can be extracted through straightforward pattern recognition rather than intensive computational matching algorithms.
Solution Approach 2:
The patent changes the fundamental parameter being measured from image intensity correlations to coded light pattern recognition. By encoding depth information directly into the light patterns' spatial-temporal structure, the system enables depth extraction with minimal computational resources.
3Measurement precision
If correspondence processing is used in stereo cameras, then depth information is obtained, but the system fails in featureless scenes
Solution Approach 1:
The system applies coded light patterns to the scene before capture, actively illuminating featureless surfaces with known spatial-temporal codes. This preliminary encoding ensures that even featureless scenes contain distinguishable depth information embedded in the projected patterns, enabling reliable depth measurement where passive stereo methods fail.
Solution Approach 2:
The coded light patterns serve as an intermediary that bridges the gap between the camera and featureless scenes. By projecting structured light codes onto the scene, the system creates artificial features that facilitate depth extraction without requiring natural scene features.
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
This approach reduces computational intensity and enhances depth reconstruction accuracy by eliminating the need for correspondence calculations, enabling real-time processing and improved performance in featureless scenes.
Implementation Method 1
projecting via a projector, a collection of planes, each at a different angle of projection, onto a scene via a plurality of coded pattern images
Implementation Method 2
a detector to capture the plurality of coded pattern images from the scene
Data Source
AI summary
In accordance with disclosed embodiments, there are provided systems, methods, and apparatuses for implementing maximum likelihood image binarization in a coded light range camera. For instance, a depth camera is described having therein a projector to project a collection of planes, each at a different angle of projection, onto a scene via a plurality of coded pattern images, each of the coded pattern images having encoded therein via a plurality of stripes, the angle of projection for the plane of projection within which the respective coded pattern image is projected; a detector to capture the plurality of coded pattern images from the scene; a processing component to adjust for ambient illumination and reflection properties of the scene; in which the processing component to further output a bit value for each pixel in the captured plurality of coded pattern images and to output a sub-pixel offset for the pixels positioned upon transitions of the plurality of stripes in the captured plurality of coded pattern images; a decoder to decode each of the plurality of coded pattern images and to adjust the decoded plurality of coded image patterns based on the sub-pixel offsets to determine the angle of projection for the corresponding plane of projection; and a triangulator to determine a position of an object in the scene based on an intersection of the determined angle of projection for the corresponding plane of projection with a known ray emanating from the detector that detected the plurality of the coded pattern images from the scene. Other related embodiments are disclosed.