Image Sensing Apparatus for Low-Power User Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing user detection methods based on image recognition require high-quality cameras, significant computational power, and continuous analysis of pixel relations, making it difficult to provide active services efficiently due to high power consumption and complex computations.

Innovation Solution

An image sensing apparatus using a low-power image sensor that analyzes images by obtaining first and second difference images between current and previous frames, and background images, allowing for efficient user existence and position determination with minimal computation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If object-recognition algorithm is used to detect user from images, then user detection accuracy is improved, but computation power requirement increases significantly

Engineering Contradiction:
Improveuser detection accuracyVSAvoidcomputation power requirement
Core Design Contradiction:
Measurement precisionVSPower

Solution Approach 1:

The patent segments the image analysis process into two distinct difference calculation stages: first calculating difference between current and previous frames to detect motion, then calculating difference between current frame and background to identify static objects. This segmentation avoids the need for complex object-recognition algorithms while maintaining effective user detection capability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent extracts only the essential motion and presence information from images by calculating simple pixel difference values between frames and background, discarding the need for complex feature extraction and recognition. This extraction approach maintains detection effectiveness while dramatically reducing computational requirements.

Inventive Principle:
Principle #2Taking out (Extraction)

2Measurement precision

If continuous image analysis is performed to track user position, then user position tracking accuracy is improved, but power consumption increases

Engineering Contradiction:
Improveuser position tracking accuracyVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent performs image analysis at periodic intervals by comparing current frames with previously stored frames and background images, rather than continuously analyzing every pixel relationship. This periodic difference calculation maintains position tracking accuracy while significantly reducing power consumption compared to continuous real-time analysis.

Inventive Principle:
Principle #19Periodic action

3Reliability

If high-quality camera is used for user detection, then detection reliability is improved, but device cost and complexity increase

Engineering Contradiction:
Improvedetection reliabilityVSAvoidcamera quality requirement
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates difference images as simplified copies of the original image data, extracting only the essential motion and presence information through pixel comparison. This copying approach allows reliable user detection using simple image sensors without requiring high-quality cameras, as the difference calculation process compensates for lower image quality.

Inventive Principle:
Principle #26Copying

4Speed

If object-chase algorithm is applied in addition to object-recognition, then real-time tracking is improved, but computation complexity increases

Engineering Contradiction:
Improvereal-time tracking speedVSAvoidalgorithm complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent merges the object detection and tracking functions into a single unified process by continuously calculating difference images between current frames and reference frames. This combination eliminates the need for separate object-recognition and object-chase algorithms, achieving real-time tracking with reduced computational complexity.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3121790B1Image sensing apparatus, object detecting method thereof and non-transitory computer readable recording medium
Publication Date: 2019.07.03 SAMSUNG ELECTRONICS CO LTD
  • EP3121790B1 patent drawingFigure 1~2
  • EP3121790B1 patent drawingFigure 3
  • EP3121790B1 patent drawingFigure 4

AI summary

An apparatus and a method of detecting an object of the apparatus are provided. The apparatus includes a sensing part configured to photograph an image, a storage configured to store a background image frame, and a controller configured to obtain a first difference image and a second difference image from the photographed image and to determine an existence and a position of an object using the first and the second difference images, wherein the first difference image is an image indicating a difference between a currently photographed image frame and a previously photographed image frame, and the second difference image is an image indicating a difference between the currently photographed image frame and the background image frame stored in the storage.