Adaptive Noise Filtering for Fixed Pattern and Random Noise Suppression

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional noise filtering techniques fail to effectively suppress structured fixed pattern noise (FPN) and random noise in images, especially in low-cost sensors and low signal-to-noise ratio imaging applications, leading to prominent artifacts.

Innovation Solution

A method and system that model and filter both power spectral densities of random noise and FPN components in spatiotemporal volumes, using adaptive filtering based on noise parameters and motion, to suppress noise effectively.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If conventional averaging or smoothing operations are applied to suppress noise, then random noise may be partially reduced, but structured fixed pattern noise becomes more visible and prominent artifacts are produced

Engineering Contradiction:

Improvenoise suppressionVSAvoidimage quality

Core Design Contradiction:

Object-affected harmful factorsVSManufacturing precision

Solution Approach 1:

The patent segments the noise into two distinct components: structured fixed pattern noise (FPN) and unstructured random noise. By separating the noise modeling approach into two independent filtering paths - one for FPN and one for random noise - the system can apply appropriate suppression techniques to each component without compromising the other, thereby resolving the contradiction between noise suppression and image quality

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the filtering parameters adaptively based on the local characteristics of the image and the estimated noise properties. By adjusting filtering strength and type according to measured noise parameters (such as noise variance and correlation structure), the system achieves effective noise suppression while preserving image details and avoiding prominent artifacts

Inventive Principle:

Principle #35Parameter changes

2Ease of manufacture

If conventional noise filtering techniques are used, then processing simplicity is maintained, but effectiveness is poor for low-cost sensors and low signal-to-noise ratio imaging applications

Engineering Contradiction:

Improvefiltering implementationVSAvoidnoise suppression effectiveness

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The patent performs preliminary estimation of noise parameters (such as FPN patterns and random noise statistics) before applying the filtering operation. By pre-characterizing the noise properties from the input image or calibration data, the system can then apply optimized filtering parameters tailored to the specific noise conditions, significantly improving suppression effectiveness for low-cost sensors and low SNR applications while maintaining implementation feasibility

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9635220B2Methods and systems for suppressing noise in images

Publication Date: 2017.04.25 TELEDYNE FLIR LLC

AI summary

Various techniques are disclosed to effectively suppress noise in images (e.g., video or still images). For example, noise in images may be more accurately modeled as having a structured random noise component and a structured fixed pattern noise (FPN) component. Various parameters of noise may be estimated robustly and efficiently in real time and in offline processing. Noise in images may be filtered adaptively, based on various noise parameters and motion parameters. Such filtering techniques may effectively suppress noise even in images that have a prominent FPN component, and may also improve effectiveness of other operations that may be affected by noise.