AI Image Stabilization Processor for Multi-Sensor Tremor Compensation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image stabilization methods using simple feedback algorithms are limited in performance and consume high power, failing to effectively compensate for various tremor patterns in different usage environments.

Innovation Solution

An artificial intelligence-based image stabilization method utilizing an artificial neural network (ANN) model trained through reinforcement learning, which acquires tremor detection data from multiple sensors and outputs stabilization data to compensate for image shaking, thereby improving processing speed and reducing power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If simple feedback algorithms are used for image stabilization, then the device complexity is reduced, but the stabilization performance and accuracy deteriorate

Engineering Contradiction:

Improvealgorithm complexityVSAvoidstabilization accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces traditional mechanical feedback control algorithms with an artificial neural network-based intelligent system. The neural network learns optimal stabilization patterns from training data and automatically adjusts compensation signals, substituting complex mathematical calculations with trained model inference that achieves higher accuracy while maintaining real-time performance.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the operational parameters of the stabilization system by using a neural network that can dynamically adjust multiple parameters simultaneously (gain values, phase corrections, frequency responses) based on learned patterns. This allows the system to adapt to different usage scenarios and tremor characteristics, improving stabilization accuracy beyond fixed-parameter feedback algorithms.

Inventive Principle:

Principle #35Parameter changes

2Use of energy by moving object

If traditional feedback algorithms are used for image stabilization, then power consumption is reduced, but processing speed and stabilization performance deteriorate

Engineering Contradiction:

Improvepower consumptionVSAvoidprocessing speed

Core Design Contradiction:

Use of energy by moving objectVSProductivity

Solution Approach 1:

The patent applies preliminary action by training the neural network model beforehand with extensive stabilization data covering various tremor patterns and usage scenarios. This pre-training phase performs the computationally intensive learning work in advance, allowing the deployed system to achieve high processing speed through efficient inference without consuming excessive power during actual image stabilization operations.

Inventive Principle:

Principle #10Preliminary action

3Device complexity

If simple feedback algorithms are used, then the device complexity is reduced, but the ability to compensate for various tremor patterns deteriorates

Engineering Contradiction:

Improvealgorithm structureVSAvoidtremor pattern compensation

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent implements universality by designing a neural network model that can handle multiple tremor patterns and usage scenarios through a single unified architecture. The network is trained on diverse data including different hand shake patterns, camera types, and shooting conditions, enabling it to universally adapt to various stabilization requirements without requiring separate algorithms for each scenario.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The AI-based image stabilization method significantly enhances stabilization speed and accuracy by considering multiple variables such as temperature, defocus amount, and modulation transfer function, while also reducing power consumption and enabling efficient focus adjustment.

Implementation Method 1

outputting stabilization data for compensating for an image shaking, the stabilization data outputted using an artificial neural network (ANN) model trained to output the stabilization data based on the tremor detection data

Methodology Applied

Scientific EffectArtificial Neural Network:

Data Source

PatentUS20250030946A1Processor for compensating for shaking of the image based on artificial intelligence and device including the same

Publication Date: 2025.01.23 DEEPX CO LTD

AI summary

A method for stabilizing an image based on artificial intelligence includes acquiring tremor detection data with respect to the image, the tremor detection data acquired from two or more sensors; outputting stabilization data for compensating for an image shaking, the stabilization data outputted using an artificial neural network (ANN) model trained to output the stabilization data based on the tremor detection data; and compensating for the image shaking using the stabilization data. A camera module includes a lens; an image sensor to output an image captured through the lens; two or more sensors to output tremor detection data with respect to the image; a controller to output stabilization data based on the tremor detection data using an ANN model; and a stabilization unit to compensate for an image shaking using the stabilization data. The ANN model is trained to output the stabilization data based on the tremor detection data.