Electronic Apparatus AI Scene Selection Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing imaging devices struggle to select important scenes in moving images based on the state of the subject, often selecting scenes regardless of user interest.

Innovation Solution

An electronic apparatus with a control device that acquires moving images, determines important scenes using AI-driven analysis of image information such as subject, composition, and color tone, and generates scene information for selective extraction and management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If important scenes are selected based on user input information during picture capturing, then the selection process is simple and fast, but the selection accuracy regarding user interest is low

Engineering Contradiction:
Improvescene selection speedVSAvoidscene importance judgment accuracy
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary analysis of the captured moving image to pre-identify important scenes before user review. By automatically analyzing image information such as subject presence, composition quality, and color tone variations in advance, the system prepares candidate important scenes that align with user interest, thus improving both selection speed and accuracy without requiring extensive user input during capturing.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If all captured moving images are stored for later review, then no important scenes are missed, but storage space is wasted and retrieval efficiency is low

Engineering Contradiction:
Improvecompleteness of important scene preservationVSAvoidstorage space consumption
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The system extracts and separates important scenes from the complete moving image data by analyzing image information such as subject detection, composition evaluation, and color tone changes. Only the extracted important scene segments are stored in the database, while the full original moving images are minimized or discarded. This extraction approach ensures no important scenes are lost while dramatically reducing storage space requirements and improving retrieval efficiency through targeted storage of only relevant content.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If detailed image analysis is performed to accurately identify important scenes, then scene selection accuracy improves, but processing time and computational resources increase

Engineering Contradiction:
Improveimportant scene identification accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The image analysis process is segmented into multiple independent evaluation dimensions: subject presence detection, composition quality assessment, and color tone variation analysis. Each dimension is processed separately using optimized algorithms tailored to its specific requirements. This segmentation allows parallel processing of different analysis aspects, reducing overall processing time while maintaining comprehensive and accurate important scene identification through the combination of multiple evaluation criteria.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS11288512B2Electronic apparatus, control device, and control method
Publication Date: 2022.03.29 SHARP KK
  • US11288512B2 patent drawing
  • US11288512B2 patent drawing
  • US11288512B2 patent drawing

AI summary

An electronic apparatus includes an imaging device and a control device, and the control device performs moving image acquisition processing of acquiring a first moving image captured by the imaging device, important scene determination processing of determining whether or not each of frames included in the first moving image is an important scene, based on image information included in the first moving image, and scene information generation processing of generating important scene information including a result of the determination whether or not each of the frames is the important scene.