Image Processing Apparatus Target Region Selection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing techniques fail to appropriately decide the output target region in images, often including unnecessary objects like passers or the user's finger, which can distract from the intended target regions such as family members, leading to unsatisfactory output.

Innovation Solution

An image processing apparatus and method that determines the output target region by evaluating and filtering out unnecessary regions based on predetermined conditions, such as the presence of unrelated individuals or objects, ensuring that only relevant regions are included in the output.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If automatic trimming is performed based on target region detection, then the output includes all detected target regions, but unnecessary objects such as passers or user's finger are also included in the output

Engineering Contradiction:
Improvetarget region detection accuracyVSAvoidunnecessary objects in output
Core Design Contradiction:
Measurement precisionVSObject-generated harmful factors

Solution Approach 1:

The patent extracts and removes unnecessary objects from the output by introducing a determination unit that identifies and excludes regions containing passers or user's finger from the final output, while preserving the detected target regions. This selective extraction resolves the contradiction by maintaining detection accuracy while eliminating harmful elements.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary determination unit that acts as a mediator between the target region detection and the final output generation. This intermediary unit evaluates each detected region and makes a determination on whether to include it in the output, thereby filtering out unnecessary objects while maintaining target regions.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If all detected regions are included in the output, then complete information is preserved, but the output quality deteriorates due to unnecessary objects

Engineering Contradiction:
Improveinformation completenessVSAvoidoutput quality
Core Design Contradiction:
Loss of informationVSManufacturing precision

Solution Approach 1:

The patent selectively extracts only the necessary information (target regions) while removing unnecessary information (passers, user's finger) from the output. This selective extraction maintains information completeness regarding the intended targets while improving output quality by eliminating irrelevant content.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The determination unit provides feedback on whether each detected region should be included in the output based on predetermined conditions. This feedback mechanism ensures that only regions meeting the criteria are included, thereby maintaining information completeness for valid targets while filtering out low-quality or irrelevant regions.

Inventive Principle:
Principle #23Feedback

3Manufacturing precision

If manual selection of output target is required, then output quality can be controlled, but processing time and operation complexity increase

Engineering Contradiction:
Improveoutput quality controlVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent implements self-service by enabling the system to automatically determine and select the output target regions based on predetermined conditions. The determination unit autonomously evaluates detected regions and makes decisions without requiring manual user intervention, thereby maintaining output quality control while significantly reducing processing time and operational complexity.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent changes the parameters of the selection process from manual to automated by introducing predetermined conditions that the determination unit evaluates automatically. This parameter change enables the system to autonomously make decisions based on objective criteria, resolving the contradiction between quality control and processing time.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS9292760B2Apparatus, method, and non-transitory computer-readable medium
Publication Date: 2016.03.22 CANON KK
  • US9292760B2 patent drawing
  • US9292760B2 patent drawing
  • US9292760B2 patent drawing

AI summary

An apparatus comprises a determination unit configured to determine, based on a region that satisfies a predetermined condition in an image as a candidate of an output target, whether to set the region as the output target; and a decision unit configured to, in a case where a plurality of regions that satisfy the predetermined condition exists in the image, decide a portion of the image including one or a plurality of regions determined, out of the plurality of regions, as the output target by said determination unit as a portion of the output target.