Optical Mouse Tracking Method Using Dual-Window Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional correlation techniques for optical input devices, such as optical mice, face challenges in achieving accurate and efficient tracking of movement due to high computational intensity and power consumption, with pixel-by-pixel correlation being time-consuming and block correlation potentially missing the actual peak correlation.
Innovation Solution
A tracking method that captures multiple frames and computes correlation between a reference frame and a compare frame within a first correlation window, with further verification using a second correlation window centered on the peak correlation value to confirm its validity, ensuring accurate displacement calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pixel by pixel correlation scheme is used, then tracking accuracy is improved, but computational time and power consumption increase
Solution Approach 1:
The patent divides the correlation computation into two stages: first performing block correlation to identify a candidate peak location, then performing pixel-by-pixel correlation only in a localized search window around that candidate. This segmentation allows the system to use the efficient block method for most of the search space while reserving the accurate pixel-by-pixel method for a small focused region, thereby maintaining high accuracy without the full computational cost of pixel-by-pixel correlation across the entire image.
2Productivity
If block correlation scheme is used, then computational efficiency is improved, but tracking accuracy deteriorates due to potential peak misidentification
Solution Approach 1:
The patent introduces an intermediary verification step between block correlation and final peak identification. After obtaining a candidate peak from block correlation, the system performs pixel-by-pixel correlation in a search window centered on that candidate to verify whether it is indeed the true peak. This intermediary verification resolves the contradiction by using the efficient block method to narrow down candidates while ensuring accuracy through focused pixel-level verification.
3Measurement precision
If longer image processing time is used, then tracking accuracy is improved, but power consumption increases
Solution Approach 1:
The patent segments the image processing into two phases with different computational intensities. The first phase uses efficient block correlation to quickly identify a candidate peak location, consuming minimal power. The second phase performs pixel-by-pixel correlation only in a small search window around the candidate, maintaining accuracy while limiting power consumption. This segmented approach allows adequate processing time for accuracy while avoiding the excessive power consumption of full-image pixel-by-pixel correlation.
Data Source
AI summary
A tracking method of an optical input device includes: capturing a first image as a reference frame and an input image subsequent to the first image as a compare frame while the optical input device operates; computing correlation between the compare frame and the reference frame for a first correlation window to obtain a first peak correlation value; computing another correlation between the compare frame and the reference frame for a second correlation window when the first peak correlation value identified is located at a corner of the first correlation window; when a second peak correlation value of the second correlation window is determined to be greater than the first peak correlation value, computes a displacement information of the optical input device based on the pixel position associated with the second peak correlation value identified in the compare frame.