Depth-Aware False Image Detection in Indoor Electronic Devices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Electronic devices face operational errors due to detecting false images of objects reflected in mirrors or displayed on screens, which are mistaken for actual objects in 2D color images, leading to incorrect object recognition.

Innovation Solution

The electronic device employs a camera module and processor to capture color and depth images, using depth information to differentiate between actual objects and false images by comparing the depth of detected areas with reference surfaces, thereby identifying and neglecting false images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If the electronic device detects objects based on 2D color images, then the device can recognize objects in real-time, but the device detects false images reflected in mirrors or displayed on TV as actual objects

Engineering Contradiction:
Improvereal-time detection speedVSAvoidobject recognition accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent transitions from 2D color image detection to 3D depth-aware detection by introducing a depth image acquisition unit. This adds the depth dimension (Z-axis) to the traditional 2D (X-Y plane) detection, enabling the system to distinguish actual objects from false images based on their spatial position relative to reference surfaces such as mirrors or screens.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Device complexity

If the electronic device uses only 2D color images for object detection, then the device complexity remains low, but the device cannot distinguish between actual objects and false images

Engineering Contradiction:
Improvedetection system complexityVSAvoidobject detection reliability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent merges the color image acquisition unit and depth image acquisition unit into an integrated detection system. The color image provides visual identification while the depth image provides spatial verification, and both are processed together by the object detection unit to reliably distinguish actual objects from false images.

Inventive Principle:
Principle #5Merging (Combining)

3Ease of operation

If the electronic device detects objects without depth information, then the detection process is simple, but the device mistakenly processes false images as actual objects

Engineering Contradiction:
Improvedetection process simplicityVSAvoidobject location accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The patent introduces a reference surface detection unit that detects mirrors, screens, or other reflective surfaces as intermediaries. By identifying these reference surfaces and their spatial positions, the system can determine whether detected objects are actual objects or false images reflected in these surfaces, thereby improving location accuracy without significantly complicating the detection process.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP3772701B1Electronic device and method of identifying a false image of an object attributable to reflection in an indoor environment
Publication Date: 2024.05.01 KOREA ADVANCED INST OF SCI & TECH
  • EP3772701B1 patent drawingFigure 1
  • EP3772701B1 patent drawingFigure 2
  • EP3772701B1 patent drawingFigure 3(a)~3(b)

AI summary

Disclosed herein are an electronic device and an operating method thereof. The electronic device is for identifying an interested false image attributable to reflection in an indoor environment, and may be configured to obtain a color image and a depth image, detect an interested area indicative of at least one object in the color image, detect at least one reference surface in the depth image, and compare an interested depth of the interested area, detected based on the depth image, with a reference depth of the reference surface and process the interested area.