Image Blur Correction Interference with Object Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image capturing apparatuses face interference between image blur correction and object tracking functions, particularly when electronic blur correction is performed based on motion vectors that include background movement, leading to incorrect object tracking.
Innovation Solution
An image capturing apparatus with an object detection unit, movement amount acquisition unit, object tracking unit, motion vector detection unit, image blur correction unit, and switching unit that allows for the suppression of electronic blur correction during object tracking by adjusting the optical element movement and image memory readout position based on the motion vector, ensuring the object tracking mode has a reduced blur correction effect compared to the blur correction mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If electronic blur correction is performed based on motion vectors detected from background movement, then image blur is corrected, but object tracking accuracy deteriorates
Solution Approach 1:
The patent divides the image into a motion vector detection area (background region) and an object tracking area (foreground region). By segmenting the image processing into separate functional zones, the system can detect motion vectors from background movement for blur correction while simultaneously tracking objects in the foreground without interference, thus resolving the contradiction between blur correction accuracy and object tracking accuracy.
Solution Approach 2:
The patent applies different processing qualities to different regions of the image. The motion vector detection area uses standard motion vector detection for blur correction, while the object tracking area uses enhanced tracking algorithms that are less sensitive to background movement. This local differentiation allows both blur correction and object tracking to function accurately in their respective regions.
2Measurement precision
If the correction lens is moved to track the main object, then object tracking is improved, but the position of background objects changes causing false motion vectors
Solution Approach 1:
The patent extracts the motion vector detection function from the entire image and confines it to a specific motion vector detection area in the background region. By taking out the detection function from the full image context and placing it in a dedicated area, the system can accurately detect background motion without being influenced by object tracking movements, thus preventing false motion vectors while maintaining object tracking accuracy.
Solution Approach 2:
The patent introduces an intermediary region (motion vector detection area) that acts as a buffer between the object tracking area and the blur correction processing. This intermediary zone allows the system to separate the effects of object tracking from background movement detection, mediating the interaction between these two functions to prevent interference.
3Manufacturing precision
If motion vector detection area is enlarged to include more background, then blur correction is improved, but object tracking is interfered with
Solution Approach 1:
The patent segments the image into distinct functional areas: a motion vector detection area for blur correction and an object tracking area for foreground object monitoring. This segmentation allows the system to maximize the motion vector detection area for accurate blur correction while preventing it from interfering with object tracking by spatially separating the two functions.
Solution Approach 2:
The patent applies different detection sensitivities and processing qualities to different regions. The motion vector detection area uses algorithms optimized for background movement detection, while the object tracking area uses algorithms optimized for foreground object tracking. This local quality differentiation allows both functions to perform accurately without mutual interference.
Data Source
AI summary
An image capturing apparatus comprises an object detection unit configured to detect a predetermined object and a movement amount acquisition unit configured to acquire an amount of movement of the predetermined object. An object tracking unit suppresses the amount of movement of the predetermined object by moving an optical element constituting an image capturing optical system. A motion vector detection unit detects a motion vector indicating an image blur amount. An image blur correction unit corrects image blur based on the motion vector. A switching unit switches whether or not to perform object tracking using the object tracking unit. The image blur correction unit makes effect of the image blur correction in a case of performing the object tracking smaller than effect of the image blur correction in a case of not performing the object tracking.