Image Processing Pipeline for De-Noising Encoded HDR Images

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Image quality is degraded during capture, conversion, and/or transmission, necessitating improved image processing methods to enhance and restore image quality.

Innovation Solution

An image processing method involving encoding, de-noising, and high dynamic range processing using machine learning algorithms to improve image quality, including operations such as downsampling, de-noising, and high dynamic range conversion.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If encoding is performed on input image to compress data, then transmission efficiency is improved, but image quality deteriorates due to noise introduction

Engineering Contradiction:

Improvetransmission efficiencyVSAvoidimage quality

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies preliminary de-noising to the encoded image before subsequent processing steps. By performing de-noising early in the processing pipeline, the method removes noise introduced during encoding, thereby recovering image quality while maintaining the transmission efficiency benefits of encoding.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent converts the harmful noise introduced by encoding into a beneficial process by applying de-noising algorithms. The noise, which initially degrades image quality, becomes an opportunity to demonstrate the effectiveness of de-noising techniques that restore and even enhance image quality beyond the original.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

2Manufacturing precision

If de-noising processing is applied to remove noise, then image quality is improved, but processing time increases

Engineering Contradiction:

Improveimage qualityVSAvoidprocessing time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent performs de-noising as a preliminary step before other image processing operations. By removing noise early, subsequent processing steps work with cleaner data, potentially reducing the computational burden and time required for later operations such as enhancement or compression.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent divides the image processing into distinct segments: encoding, de-noising, and subsequent processing. This segmentation allows each step to be optimized independently, with de-noising focused specifically on noise removal without the computational overhead of combined processing.

Inventive Principle:

Principle #1Segmentation

3Manufacturing precision

If high dynamic range conversion is performed to expand dynamic range, then image quality is improved, but device complexity increases

Engineering Contradiction:

Improveimage qualityVSAvoidprocessing complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies high dynamic range conversion after de-noising but as a distinct preliminary step before final image output. This sequencing allows the complex HDR processing to work on already-cleaned data, potentially simplifying the HDR algorithm requirements and reducing overall system complexity.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12602752B2Image processing method and image processing apparatus

Publication Date: 2026.04.14 GENESYS LOGIC INC

AI summary

An image processing method and an image processing apparatus are provided. In the method, first encoding is performed on an input image, to output a first noisy image. De-noising is performed on the first noisy image, to output a first de-noised image. De-noising is performed on the input image according to the first de-noised image, to output a first image.