Image Stabilizer Control via Motion Vector Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image stabilization systems struggle to accurately differentiate between motion vectors of still and moving objects, leading to ineffective image stabilization when a moving object is detected, as they rely on accurately detecting background motion vectors.
Innovation Solution
A control apparatus and method that includes a camera system with a MPU that processes motion vector information by comparing blur amounts from an angular velocity sensor and a motion vector, adjusting feedback to the image stabilizer to prevent erroneous detection, setting the motion vector to 0 or multiplying it by 0.5 when it exceeds a certain value, ensuring accurate image stabilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If motion vector information is fed back to the image stabilizer to improve stabilization accuracy, then image stabilization accuracy is improved, but when a moving object is detected the image stabilizer follows the moving object causing erroneous stabilization
Solution Approach 1:
The patent implements feedback control by detecting motion vectors from captured images and feeding this information back to the image stabilizer. The control apparatus calculates motion vectors between sequential frames and uses this feedback to adjust the stabilizer's correction amounts, thereby improving stabilization accuracy while maintaining system reliability through proper differentiation of moving and still objects.
Solution Approach 2:
The patent introduces an intermediary processing step between motion vector detection and stabilizer control. A control apparatus acts as an intermediary that analyzes motion vectors, determines whether they represent still or moving objects, and selectively applies or suppresses feedback based on this analysis, preventing erroneous stabilization when moving objects are detected.
2Measurement precision
If the system uses motion vector detection to calculate blur residual, then blur residual detection capability is improved, but accurate detection of background motion vector becomes difficult when moving objects are present
Solution Approach 1:
The patent segments the motion detection process into distinct analysis stages. The control apparatus divides the image into multiple regions and calculates motion vectors for each region separately. By segmenting the analysis, the system can identify regions with significant motion (moving objects) versus regions with minimal motion (background), thereby accurately detecting background motion vectors even when moving objects are present.
Solution Approach 2:
The patent applies partial feedback by selectively using motion vector information only from regions identified as still objects. Rather than using all motion vectors uniformly, the system performs partial action by filtering and applying feedback only where appropriate, preventing moving objects from corrupting the background motion vector detection.
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
This approach enables highly accurate image stabilization by properly controlling motion vector information and reducing the influence of erroneous motion vector feedback, enhancing the stability of captured images.
Implementation Method 1
an angular velocity sensor that detects a shake angular velocity
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A control apparatus (118) includes a first acquirer (1181) configured to acquire first blur information based on a motion vector calculated using an imaging signal from an image sensor, a second acquirer (1182, 1189) configured to acquire second blur information based on a blur signal from a blur detector, and a controller (118a) configured to control driving of an image stabilizing element in accordance with third blur information obtained by using the first blur information and the second blur information. A weight for the first blur information in acquiring the third blur information is determined based on a comparison between the first blur information and the second blur information.