Single-Lens 3D Imaging with Central Aperture

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing three-dimensional (3-D) imaging techniques face challenges in achieving precise quantitative measurements due to issues with dynamic range, diffraction effects, and the complexity of processing blurred images, which are exacerbated by the need for multiple sensors and complex algorithms in single-lens, single-sensor systems.

Innovation Solution

A single-lens, single-sensor 3-D imaging device with a central aperture and off-axis defocusing apertures that use optical filters and a processor to distinguish between background and defocused images, allowing for the reconstruction of 3-D images by measuring the lateral offset and size of image triangles formed by the apertures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If a large aperture is used to capture more light, then the imaging system can capture sufficient light energy, but the defocused points appear blurred occupying large sensor space requiring sophisticated algorithms

Engineering Contradiction:
Improvelight energy captureVSAvoidalgorithm complexity
Core Design Contradiction:
Use of energy by moving objectVSDevice complexity

Solution Approach 1:

The aperture is segmented into multiple sub-apertures arranged in specific patterns. Each sub-aperture captures light from different angular directions, creating spatially separated image samples on the sensor. This segmentation allows the system to maintain a large effective aperture for light capture while avoiding the formation of large blurred discs, as each sub-aperture produces a smaller, distinct image region.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from measuring depth through blur size (2D intensity distribution) to measuring depth through lateral displacement of image samples (spatial position in 2D). By arranging sub-apertures to create laterally shifted image samples, the system encodes depth information in the spatial domain rather than the intensity domain, simplifying the measurement process and reducing algorithmic complexity.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If blur-based depth measurement is used, then depth information can be obtained from a single image, but the intensity measurement is sensitive to brightness differences requiring extremely high dynamic range sensors

Engineering Contradiction:
Improvedepth measurementVSAvoidsensor dynamic range requirement
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent changes the measurement dimension from intensity (brightness) to position (spatial location). Instead of measuring how bright a blurred region is to infer depth, the system measures the lateral position of image samples formed by different sub-apertures. This positional measurement is inherently more robust to intensity variations and does not require extreme dynamic range capabilities.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent replaces photometric measurement (intensity-based) with geometric measurement (position-based). By using the spatial arrangement of image samples rather than their intensity values, the system substitutes a measurement approach that is less sensitive to lighting conditions and object brightness, thereby reducing sensor dynamic range requirements.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If multiple defocusing apertures are used to improve 3-D imaging accuracy, then measurement precision improves, but sensor overcrowding increases making image processing more complex

Engineering Contradiction:
Improve3-D imaging accuracyVSAvoidsensor area utilization
Core Design Contradiction:
Measurement precisionVSArea of stationary object

Solution Approach 1:

The aperture is divided into multiple sub-apertures that are spatially distributed. Each sub-aperture creates a distinct, localized image sample on the sensor. This segmentation allows multiple depth-sampling channels to coexist on the same sensor without excessive overlap, as each sub-aperture's image contribution is spatially separated according to its position in the aperture plane.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent uses the spatial arrangement of sub-apertures in the aperture plane to create corresponding spatial patterns on the sensor. By mapping aperture geometry to sensor pattern geometry, the system efficiently packs multiple measurement channels into the available sensor area without requiring excessive sensor real estate, as the pattern is determined by the aperture configuration rather than requiring separate sensor regions.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

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 simplifies 3-D imaging by reducing the dynamic range issues and sensor overcrowding, enabling precise and cost-effective 3-D mapping and surface inspection with improved robustness and accuracy.

Implementation Method 1

a lens, a central aperture located along an optical axis for projecting an entire image of a target object

Methodology Applied
Scientific EffectRefraction: Refraction

Implementation Method 2

A single-lens, single-sensor 3-D imaging device with a central aperture and off-axis defocusing apertures that use optical filters

Methodology Applied
Scientific EffectOptical filtering: Filter (optical)

Data Source

PatentUS9736463B2Single-lens, single-sensor 3-D imaging device with a central aperture for obtaining camera position
Publication Date: 2017.08.15 CALIFORNIA INST OF TECH
  • US9736463B2 patent drawing
  • US9736463B2 patent drawing
  • US9736463B2 patent drawing

AI summary

A device and method for three-dimensional (3-D) imaging using a defocusing technique is disclosed. The device comprises a lens, a central aperture located along an optical axis for projecting an entire image of a target object, at least one defocusing aperture located off of the optical axis, a sensor operable for capturing electromagnetic radiation transmitted from an object through the lens and the central aperture and the at least one defocusing aperture, and a processor communicatively connected with the sensor for processing the sensor information and producing a 3-D image of the object. Different optical filters can be used for the central aperture and the defocusing apertures respectively, whereby a background image produced by the central aperture can be easily distinguished from defocused images produced by the defocusing apertures.