Image Processing Module Subject Detection via Regional Weighting

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional cameras lack the ability to detect the subject of an image automatically, leading to blurring when the desired object is not centered, resulting in poor image quality.

Innovation Solution

An imaging device with an image capturing module and an image processing module that captures a temporal image, defines regions based on the image center, assigns initial values to objects, modifies these values, sorts them, and selects the subject for focusing operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If traditional cameras rely on user to direct focusing direction, then device complexity is reduced, but image quality deteriorates when desired object is not centered

Engineering Contradiction:
Improveimage qualityVSAvoiddevice complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The imaging device performs automatic subject detection and selection without requiring user intervention. The image processing module automatically identifies objects, calculates scores based on regional weights, and determines the subject, allowing the device to serve itself in the focusing process

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system performs preliminary subject detection and scoring before the actual photographing operation. By pre-identifying the subject and calculating regional weights in advance, the system prepares the focusing information needed for high-quality imaging

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If user does not direct focusing direction precisely, then ease of operation is improved, but image quality deteriorates due to blurring

Engineering Contradiction:
Improveease of operationVSAvoidimage quality
Core Design Contradiction:
Ease of operationVSManufacturing precision

Solution Approach 1:

The imaging device automatically detects subjects and determines focusing directions without requiring precise user operation. The system self-corrects the focusing direction by identifying the subject's position and calculating appropriate regional weights, eliminating the need for user precision while maintaining image quality

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system provides visual feedback by displaying the detected subject and regional weights on the display unit. This allows users to see what the system has identified and makes adjustments if needed, creating a feedback loop that improves both ease of operation and image quality

Inventive Principle:
Principle #23Feedback

3Manufacturing precision

If automatic subject detection is implemented, then image quality is improved through accurate focusing, but device complexity increases

Engineering Contradiction:
Improveimage qualityVSAvoiddevice complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The image processing module divides the captured image into multiple regions and assigns different weights to each region. By segmenting the image and applying regional weighting, the system efficiently identifies the subject without requiring complex artificial intelligence algorithms, thus improving image quality while controlling device complexity

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes parameters by assigning different weights to different regions of the image. By adjusting these regional weight parameters, the system optimizes subject detection accuracy without adding significant hardware complexity, achieving improved image quality through software-based parameter optimization

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS8897589B2Method of detecting subject of image and imaging device thereof
Publication Date: 2014.11.25 ALTEK SEMICON
  • US8897589B2 patent drawing
  • US8897589B2 patent drawing
  • US8897589B2 patent drawing

AI summary

A method of detecting the subject of an image and imaging device thereof are disclosed. The method comprises the following steps of providing an image capturing module to capture the temporal image, providing the image processing module to receive the temporal image and to determine a plurality of objects from the temporal image, using the image processing module to define a plurality of regions on the temporal image based on the center of temporal image, using the image processing module to assign an initial setup value to each of the objects according to the region corresponding to each of the objects, using the imaging processing module to execute an operation to increase or decrease each of the initial setup values, using the image processing module to execute a sorting, and selecting at least one of the objects as a subject of the temporal image.