Non-linear Mapping Circuit for Image Blur Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Digital cameras often produce images with blur due to variations in the point spread function (PSF) of their objective optics, which existing methods struggle to fully compensate for, leading to suboptimal image quality.

Innovation Solution

An image restoration circuit performs non-linear mapping and linear convolution filtering on pixel values to generate output pixel streams with reduced blur, using a deconvolution filter with a kernel determined by the PSF, and optionally includes a second non-linear mapping to cancel initial mapping distortions, employing piecewise linear functions or look-up tables to mitigate artifacts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If linear convolution filtering is applied to deblur images, then blur reduction is achieved, but artifacts appear in low-illumination regions due to noise amplification

Engineering Contradiction:
Improveimage sharpnessVSAvoidnoise artifacts
Core Design Contradiction:
Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

The patent applies non-linear mapping to pixel values before linear convolution filtering to pre-condition the data. This preliminary transformation adjusts the distribution of pixel values to reduce noise amplification effects during subsequent deconvolution, particularly in low-illumination regions where noise is most problematic.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent transforms pixel value parameters through non-linear mapping functions that change the statistical distribution of intensities. By modifying how pixel values are represented before filtering, the system alters the noise characteristics and reduces artifact generation during blur correction.

Inventive Principle:
Principle #35Parameter changes

2Illumination intensity

If non-linear mapping is applied to pixel values, then dynamic range of low-illumination pixels is improved, but distortion is introduced that requires additional processing

Engineering Contradiction:
Improvelow-illumination pixel dynamic rangeVSAvoidpixel value distortion
Core Design Contradiction:
Illumination intensityVSLoss of information

Solution Approach 1:

The patent introduces an intermediate non-linear mapping step that acts as a mediator between the original pixel values and the final deblurred output. This intermediate transformation improves low-illumination pixel representation while the subsequent inverse mapping serves as a corrective intermediary to remove introduced distortions.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent divides the image processing into separate stages: first applying non-linear mapping to specific pixel value ranges (particularly low-illumination regions), then applying linear filtering, and finally applying inverse mapping. This segmentation allows targeted improvement of specific pixel ranges without uniformly distorting the entire image.

Inventive Principle:
Principle #1Segmentation

3Manufacturing precision

If deconvolution filtering is used to compensate for PSF variations, then image restoration is achieved, but computational complexity increases

Engineering Contradiction:
Improveimage restoration qualityVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent replaces complex adaptive deconvolution algorithms with a simpler pipeline combining non-linear mapping and fixed linear convolution filtering. By substituting the mechanical complexity of adaptive PSF estimation and variable filtering with pre-computed mapping functions, the system achieves comparable restoration quality with reduced computational burden.

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

Data Source

PatentEP2162863B1Non-linear transformations for enhancement of images
Publication Date: 2014.07.30 DIGITALPTICS CORP INT
  • EP2162863B1 patent drawingFigure 1
  • EP2162863B1 patent drawingFigure 2
  • EP2162863B1 patent drawingFigure 3~4

AI summary

Imaging apparatus (26) is provided for use with an image sensor (24). The apparatus includes a non-linear mapping circuit (42), which is configured to receive a raw stream of input pixel values generated by the image sensor and to perform a non-linear mapping of the input pixel values. to generate a mapped stream of mapped pixel values, and a linear convolution filter (44), which is arranged to filter the mapped stream of mapped pixel values to generate a filtered stream of filtered pixel values. Other embodiments are also described.