Subject Tracking Apparatus Position Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional subject tracking methods often result in wrong tracking due to background interference or changes in the subject's orientation, leading to shifted subject regions and continued incorrect tracking.
Innovation Solution
A subject tracking apparatus that registers a reference image, extracts subject regions based on correlation, detects characteristic pixels, and corrects positional misalignment to ensure accurate tracking, updating the reference image for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of stationary object
If template matching is used for subject tracking, then tracking continuity is maintained, but wrong tracking occurs due to background interference or subject orientation changes
Solution Approach 1:
The patent segments the subject region into multiple blocks and selects characteristic pixels from specific blocks (e.g., eye region blocks) to represent the subject. This segmentation allows the system to focus on distinctive features rather than the entire subject region, improving reliability by reducing the influence of background interference and orientation changes on tracking accuracy.
Solution Approach 2:
Instead of directly matching the entire subject region template with the current frame, the patent inverts the approach by detecting characteristic pixels first and then using their distribution to determine subject position. This inversion from direct template matching to characteristic pixel-based detection resolves the contradiction by maintaining tracking continuity while improving accuracy through selective feature detection.
2Ease of manufacture
If the reference image includes background regions, then matching processing is simplified, but subject position detection becomes inaccurate
Solution Approach 1:
The patent applies local quality by assigning different importance to different regions within the subject template. Characteristic pixels are selected from specific local blocks (e.g., eye regions) rather than uniformly from the entire template. This allows the system to maintain simple matching processing while achieving accurate position detection by focusing on locally distinctive features.
Solution Approach 2:
The patent performs preliminary action by pre-selecting characteristic pixels from specific blocks of the subject template before matching processing. This preliminary selection of key feature pixels simplifies the subsequent matching process while ensuring accurate position detection, as the characteristic pixels are already identified and ready for use in position calculation.
3Loss of information
If characteristic pixels are selected from the entire subject region, then subject representation is comprehensive, but processing complexity increases
Solution Approach 1:
The patent extracts only the essential characteristic pixels from specific blocks of the subject region rather than using all pixels. By taking out only the necessary characteristic pixels (e.g., from eye regions) needed for accurate tracking, the system maintains comprehensive subject representation while significantly reducing processing complexity compared to using the entire subject region.
Data Source
AI summary
A subject tracking apparatus extracts a subject region which is similar to a reference image on the basis of a degree of correlation with the reference image for tracking a predetermined subject from images supplied in a time series manner. Further, the subject tracking apparatus detects the position of the predetermined subject in the subject region on the basis of the distribution of characteristic pixels representing the predetermined subject contained in the subject region, and corrects the subject region so as to reduce a shift in position of the predetermined subject in the subject region. Moreover, the corrected subject region is taken as the result of tracking the predetermined subject, and the reference image is updated with the corrected subject region as the reference image to be used for the next supplied image.