Computational Inverse Imaging for Simple Lens Aberration Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Modern camera optics are complex and costly due to the need for multiple lens elements to compensate for aberrations, leading to increased weight and lens flare, while simpler optics suffer from significant distortions that affect image quality.

Innovation Solution

A method and system that utilize computational image processing to compensate for distortions by sharing information between color channels, allowing for post-capture correction of images using cross-channel analysis and prior information to minimize aberrations, enabling high-quality photography with simpler lens designs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If modern complex lens designs with multiple elements are used, then geometric and chromatic aberrations are compensated, but device complexity, cost, and weight increase significantly

Engineering Contradiction:
Improveaberration compensationVSAvoidlens element complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent replaces the mechanical/optical system of multiple lens elements with a computational system. Instead of using complex physical lens assemblies to correct aberrations, the invention captures images through simple optics and then applies digital image processing algorithms to compensate for geometric distortion, chromatic aberration, and other optical defects post-capture. This substitution of computational methods for mechanical optical complexity directly resolves the contradiction between aberration compensation and device complexity.

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

Solution Approach 2:

The patent performs preliminary characterization of the optical system's aberrations during manufacturing or calibration. By pre-measuring and storing the distortion maps and aberration profiles of simple lenses, the system can then apply pre-computed correction algorithms to images captured with these lenses. This preliminary action enables simple optics to achieve complex correction capabilities without requiring complex lens designs.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If modern complex lens designs with multiple elements are used, then image quality is improved, but weight and cost increase significantly

Engineering Contradiction:
Improveimage qualityVSAvoidlens weight
Core Design Contradiction:
Manufacturing precisionVSWeight of moving object

Solution Approach 1:

The patent replaces heavy mechanical lens assemblies with lightweight simple lenses combined with computational correction. Instead of using multiple heavy glass elements and metal mountings to achieve high image quality, the invention uses a single simple lens element whose weight is minimal compared to modern multi-element lenses, while digital processing algorithms compensate for the optical deficiencies to maintain high image quality.

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

3Manufacturing precision

If modern complex lens designs with multiple elements are used, then geometric and chromatic aberrations are compensated, but lens flare increases

Engineering Contradiction:
Improveaberration compensationVSAvoidlens flare
Core Design Contradiction:
Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

The patent extracts and separates the aberration compensation function from the optical path. Instead of using additional lens elements that physically interact with light and generate flare, the invention captures the raw image with simple optics and then digitally extracts and corrects the aberration effects through image processing algorithms. This extraction of the correction function to the digital domain eliminates the source of lens flare while maintaining aberration compensation.

Inventive Principle:
Principle #2Taking out (Extraction)

4Device complexity

If simple lens designs are used, then device complexity and weight are reduced, but significant geometric and chromatic distortions occur

Engineering Contradiction:
Improvelens simplicityVSAvoidimage distortion
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent substitutes computational correction for mechanical optical precision. Simple lenses inherently produce geometric and chromatic distortions, but the invention replaces the need for complex optical precision with digital image processing that characterizes and corrects these distortions. The computational system compensates for the optical imperfections, enabling simple lenses to produce high-quality images without requiring complex lens designs.

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

Solution Approach 2:

The patent creates a digital copy or model of the optical system's aberration characteristics through calibration procedures. By capturing test images through the simple lens and analyzing the distortion patterns, the system builds a digital representation of the lens's optical imperfections. This digital copy is then used to generate correction algorithms that reverse the distortions, allowing simple lenses to achieve precision comparable to complex lenses.

Inventive Principle:
Principle #26Copying

Data Source

PatentUS9307208B2System and method for solving inverse imaging problems
Publication Date: 2016.04.05 QUALCOMM INC
  • US9307208B2 patent drawing
  • US9307208B2 patent drawing
  • US9307208B2 patent drawing

AI summary

There is provided a computer-implemented method for solving inverse imaging problems to compensate for distortions in an image. The method comprises: minimizing a cost objective function containing a data fitting term and one or more image prior terms to each of the plurality of channels, the one or more image prior terms comprising cross-channel information for a plurality of channels derived from the image.