Image Blur Correction with Position Compensation During Part Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image blur correction techniques in imaging apparatuses fail to maintain image quality when switching object parts for detection, leading to unnatural changes in object positions within the image.
Innovation Solution
An image blur correction apparatus that includes an object detection unit, a selection unit, and an object blur correction amount calculation unit, which calculates and adjusts the blur correction amount based on the difference between the selected object part and a predetermined target position, and optionally adjusts the target position, to stabilize the object's position in the image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If object detection is performed by switching between different parts (e.g., face to trunk) based on detection availability, then the reliability of object tracking is improved, but the position stability of the object in the image deteriorates
Solution Approach 1:
The patent introduces a position correction mechanism that acts as an intermediary between object detection and image blur correction. When the detected object part changes (e.g., from face to trunk), the system calculates the position difference between the new part and the original target position, then applies correction to maintain stable object positioning in the image while still allowing part switching for reliable tracking.
Solution Approach 2:
The system dynamically adjusts the target position parameter based on which object part is detected. When switching from detecting one part (e.g., face) to another (e.g., trunk), the target position is updated to compensate for the positional difference, ensuring that the object remains stably positioned in the image despite the change in detected part.
2Speed
If the position of the detected object part is used directly for blur correction control, then the responsiveness of blur correction is improved, but the image quality deteriorates due to unnatural position changes when switching parts
Solution Approach 1:
The system performs preliminary calculation of position differences when object parts are switched. By pre-calculating the offset between the newly detected part and the original target position, the system can apply smooth corrections without causing unnatural jumps in object position, thereby maintaining image quality while preserving responsive blur correction.
Data Source
AI summary
An image blur correction apparatus comprises an object detection unit configured to output a position of the part of the object, a selection unit configured to select a part of an object from among parts of one or more objects, and an object blur correction amount calculation unit configured to calculate an object blur correction amount, wherein the object blur correction amount calculation unit, when a part of an object selected by the selection unit is switched from a first part to a second part, calculates the object blur correction amount based on a difference between a position of the selected part of the object and the target position, a position of the second part in an image, and a position of the first part in an image acquired before switching.