Object Masking with Border Visibility for Error Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing systems struggle to accurately determine whether automatic masking of objects in images is erroneous, leading to potential coverage of adjacent relevant areas, which hinders user analysis.

Innovation Solution

An image processing apparatus and method that detects objects, generates a mask smaller than the object, and positions it to cover the object's body while keeping the border area visible, allowing users to infer masking errors by preserving relevant information in the border area.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If a mask is applied to an object in an image to reduce user distraction, then the visibility of the object is reduced, but the user cannot determine whether the automatic masking was erroneous

Engineering Contradiction:

Improveuser distractionVSAvoidmasking accuracy verification

Core Design Contradiction:

Object-affected harmful factorsVSReliability

Solution Approach 1:

The mask is applied selectively to different regions of the object with different opacity levels. The body of the object is masked with high opacity to reduce distraction, while the border area is kept unmasked or with low opacity to allow verification of masking accuracy. This local differentiation resolves the contradiction by maintaining both distraction reduction and error detectability.

Inventive Principle:

Principle #3Local quality

2Object-affected harmful factors

If a mask covers the entire object to maximize distraction reduction, then user distraction is minimized, but diagnostic information and border area details are lost

Engineering Contradiction:

Improveuser distractionVSAvoidborder area and diagnostic information

Core Design Contradiction:

Object-affected harmful factorsVSLoss of information

Solution Approach 1:

The object is segmented into two distinct regions: the body and the border area. The mask is applied differently to each segment - fully applied to the body for distraction reduction, and not applied or partially applied to the border area to preserve diagnostic information. This segmentation allows simultaneous achievement of both distraction reduction and information preservation.

Inventive Principle:

Principle #1Segmentation

3Loss of information

If the mask size is reduced to keep the border area visible, then diagnostic information is preserved, but the masking effect on the object body is reduced

Engineering Contradiction:

Improveborder area visibilityVSAvoiduser distraction

Core Design Contradiction:

Loss of informationVSObject-affected harmful factors

Solution Approach 1:

Different regions of the object are treated with different mask qualities. The body region receives full masking to maximize distraction reduction, while the border region receives no mask or partial mask to preserve diagnostic information. This local quality differentiation ensures that the reduced mask size does not compromise the overall masking effectiveness.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS11263732B2Imaging processing apparatus and method for masking an object

Publication Date: 2022.03.01 KONINKLIJKE PHILIPS NV

AI summary

Image processing apparatus 110 for applying a mask to an object, comprising an input 120 for obtaining an image 122, a processor 130 for (i) detecting the object in the image, and (ii) applying the mask to the object in the image for obtaining an output image 60, and the processor being arranged for said applying the mask to the object by (j) establishing an object contour of the object, (jj) generating, based on the object contour, a mask being smaller than the object, and (jjj) positioning the mask over the object for masking a body of the object while keeping clear a border area of the object.