Radiographic Image Noise Removal via Body Thickness Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Radiographic images suffer from noticeable quantum noise due to increased scattered radiation when the subject's body thickness is large, which degrades image quality and makes noise removal challenging during contrast enhancement processes.

Innovation Solution

A radiographic image processing device that acquires body thickness information and uses it to adjust the noise removal process, increasing the degree of noise removal as body thickness increases, by estimating noise amounts and applying a smoothing filter tailored to the converted noise levels.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If contrast enhancement or frequency processing is performed to compensate for scattered radiation, then image contrast is improved, but quantum noise is enhanced and becomes more noticeable

Engineering Contradiction:

Improveimage contrastVSAvoidquantum noise visibility

Core Design Contradiction:

Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent applies noise removal processing before contrast enhancement or frequency processing. By removing quantum noise in advance and creating a noise-removed image, subsequent contrast enhancement operations can be performed without amplifying the noise, thus resolving the contradiction between improving contrast and preventing noise visibility.

Inventive Principle:

Principle #10Preliminary action

2Quantity of substance

If body thickness increases, then scattered radiation increases, but quantum noise becomes more noticeable and image quality deteriorates

Engineering Contradiction:

Improvescattered radiation amountVSAvoidquantum noise

Core Design Contradiction:

Quantity of substanceVSObject-affected harmful factors

Solution Approach 1:

The patent performs noise removal processing as a preliminary step before any contrast enhancement or frequency processing. This preliminary noise removal creates a clean image foundation that allows subsequent processing to improve contrast without amplifying the quantum noise that increases with body thickness.

Inventive Principle:

Principle #10Preliminary action

3Object-affected harmful factors

If smoothing process is applied to remove quantum noise, then noise is reduced, but edge components may be deteriorated

Engineering Contradiction:

Improvequantum noiseVSAvoidedge component

Core Design Contradiction:

Object-affected harmful factorsVSShape

Solution Approach 1:

The patent applies smoothing filter processing as a preliminary noise removal step before edge enhancement or frequency processing. By removing noise in advance, the subsequent edge enhancement operations can sharpen edges without being confounded by the presence of quantum noise, thus resolving the contradiction between noise removal and edge preservation.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS10194881B2Radiographic image processing device, method, and recording medium

Publication Date: 2019.02.05 FUJIFILM CORP

AI summary

A radiographic image which is captured by irradiating a subject with radiation is acquired. A body thickness information acquisition unit acquires body thickness information of a subject. The radiographic image may be analyzed to acquire the body thickness information. A noise removal unit removes quantum noise included in the radiographic image on the basis of the body thickness information, using a filtering process using, for example, a smoothing filter.