Image Stabilization Apparatus Selective Frame Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image stabilization techniques deteriorate image quality when correcting image blur, especially when image data with high or low spatial frequency, moving subjects, or camera shake are included, as these data sets are not suitable for superposition-based correction methods.

Innovation Solution

An image stabilization apparatus that detects image blur and selectively chooses suitable image data for correction, excluding unsuitable data to prevent image quality deterioration, by using a detection unit, evaluation unit, and correction unit to adjust exposure and superpose image data appropriately.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If image stabilization correction is performed using superposition of multiple image data frames, then image blur is reduced, but image quality deteriorates when unsuitable image data (high/low spatial frequency, moving subjects, rotation camera shake) is included

Engineering Contradiction:

Improveimage blur correction precisionVSAvoidimage quality

Core Design Contradiction:

Manufacturing precisionVSReliability

Solution Approach 1:

The evaluation unit performs preliminary evaluation of each image data frame to determine suitability for correction before the actual superposition correction is performed. This preliminary screening identifies and excludes unsuitable frames (those with high/low spatial frequency, moving subjects, or rotation camera shake) from the correction process, preventing quality deterioration while maintaining correction effectiveness.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Different image data frames are treated differently based on their individual characteristics. The system evaluates and selects only the suitable frames for correction, applying the correction process locally to the appropriate subset of data rather than uniformly to all frames. This ensures that correction is performed only on images that will improve quality, not those that would deteriorate it.

Inventive Principle:

Principle #3Local quality

2Productivity

If all image data frames are used for correction superposition, then correction completeness is improved, but image quality deteriorates due to inclusion of unsuitable data

Engineering Contradiction:

Improvecorrection processing efficiencyVSAvoidimage quality

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

Before performing correction superposition, the system preliminarily evaluates each image data frame to assess its suitability. This preliminary action filters out unsuitable frames (those with high/low spatial frequency, moving subjects, or rotation camera shake) from the correction process, ensuring that only appropriate data is used for superposition and thus maintaining image quality while still achieving comprehensive correction where applicable.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS7773825B2Image stabilization apparatus, method thereof, and program product thereof

Publication Date: 2010.08.10 RENESAS ELECTRONICS CORP

AI summary

An image stabilization apparatus of the present invention includes a detection unit which detects an amount of an image blur between a plurality of image data, an evaluation unit which selects at least one image data to use in correcting image blur from the plurality of image data according to a result of the detection of the amount of image blur, and a correction unit which generates image data with its image blur corrected using the image data selected by the evaluation unit.