Dynamic Motion Vector Weighting for Zoom-Adaptive Image Stabilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing camera systems face difficulties in dynamically switching between image blur correction and subject tracking operations, especially at high zoom magnifications, where slight camera shakes result in significant changes in the angle of view, making it challenging for photographers to capture subjects effectively.
Innovation Solution
An image processing apparatus that includes a processor or circuit configured to detect motion vectors from captured images, calculate representative motion vectors by weighting vectors from specific regions, and switch between subject tracking and image stabilization based on electronic zoom magnification thresholds, ensuring appropriate operations at varying zoom levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of moving object
If electronic zoom magnification is increased to achieve high-magnification zoom, then the number of pixels in the readout region decreases, but the ability to capture subjects deteriorates due to slight camera shakes causing large changes in angle of view
Solution Approach 1:
The patent dynamically switches between two operational modes (image blur correction mode and subject tracking mode) based on the zoom magnification level. At low zoom magnifications, the system performs image blur correction to stabilize the image. At high zoom magnifications, the system switches to subject tracking to automatically follow the subject and compensate for camera shakes, thereby maintaining subject capture stability across varying zoom levels.
Solution Approach 2:
The patent changes the operational parameter (mode of operation) based on the zoom magnification threshold. When the zoom magnification exceeds a predetermined threshold, the system transitions from using motion vectors for image blur correction to using motion vectors for subject tracking. This parameter change allows the system to adapt its behavior to the current zoom level, resolving the contradiction between achieving high magnification and maintaining subject capture stability.
2Reliability
If subject tracking is always performed to maintain subject capture at high zoom, then the photographer's intended camera work is inhibited at low zoom magnification
Solution Approach 1:
The system dynamically adjusts its operational mode based on the current zoom magnification. At low zoom magnifications where the photographer can easily track subjects manually, the system performs image blur correction without automatic subject tracking, preserving photographer control freedom. At high zoom magnifications where manual tracking becomes difficult, the system automatically switches to subject tracking mode, providing reliable subject capture only when needed.
Solution Approach 2:
The patent uses the zoom magnification level as a threshold parameter to determine when to activate subject tracking. By setting a predetermined threshold value, the system automatically switches between image blur correction mode and subject tracking mode. This parameter-based switching ensures that subject tracking is activated only when the zoom magnification exceeds the threshold, preventing unnecessary interference with photographer control at lower zoom levels while ensuring reliable subject capture at high zoom levels.
3Measurement precision
If motion vector detection region is fixed for image blur correction, then subject tracking performance deteriorates at high zoom magnification
Solution Approach 1:
The patent dynamically changes the purpose and region of motion vector detection based on the zoom magnification level. At low zoom magnifications, the motion vector detection is performed for image blur correction using the entire image or a standard region. At high zoom magnifications, the system switches to subject tracking mode where motion vectors are detected specifically from the subject region to enable accurate subject following. This dynamic adaptation allows the motion vector detection to serve different functions appropriate to each zoom level.
Solution Approach 2:
In subject tracking mode at high zoom magnification, the patent applies local quality by focusing motion vector detection specifically on the subject region rather than the entire image. This localized approach improves the accuracy of subject tracking by concentrating computational resources on the relevant area. The system determines a subject region based on motion vectors and uses this region for tracking, thereby achieving both high measurement precision for motion detection and adaptability to different zoom levels.
Data Source
AI summary
An image processing apparatus includes an electronic zoom unit, a motion vector detection unit, a first representative vector calculation unit, a second representative vector calculation unit, and a third representative vector calculation unit, wherein the third representative vector calculation unit determines the third representative motion vector by setting a weight of the first representative motion vector greater than a weight of the second representative motion vector when an enlargement magnification in the electronic zoom unit is equal to or greater than a threshold, and determines the third representative motion vector by setting a weight of the second representative motion vector greater than a weight of the first representative motion vector when an enlargement magnification in the electronic zoom unit is less than the threshold.