Retinal Position Tracking with Adaptive Reference Image
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ophthalmic devices face challenges in achieving satisfactory positional confidence, repeatability, and accuracy in retinal position tracking due to low dynamic range images and small eye movements, especially under low infrared illumination, leading to poor signal-to-noise and feature variation across image frames.
Innovation Solution
A method involving cross-correlation between reference and received images to calculate offsets, with an adaptive reference image modification process that accentuates valid retinal features and attenuates noise, using a processor-based image processing apparatus to generate retinal position tracking information and stabilize the OCT imaging beam.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If images are acquired at high frame rates under low infrared illumination, then image acquisition speed is improved, but signal-to-noise ratio deteriorates
Solution Approach 1:
The patent combines multiple consecutive image frames through accumulation to create a composite reference image. This merging process integrates signal information across multiple frames while averaging out random noise, thereby improving the signal-to-noise ratio of the reference image used for tracking calculations.
Solution Approach 2:
The patent implements continuous image accumulation over multiple frames to build the reference image, maintaining a continuous integration process that maximizes signal capture while minimizing the impact of transient noise in individual frames.
2Device complexity
If standard cross-correlation is used on low-quality images, then processing simplicity is maintained, but tracking accuracy deteriorates
Solution Approach 1:
The patent performs preliminary image accumulation to create a high-quality reference image before conducting cross-correlation calculations. This pre-processing step enhances the quality of input data for tracking, ensuring that subsequent correlation operations achieve higher accuracy even with low-quality individual frames.
Solution Approach 2:
The patent transforms the reference image by accumulating multiple frames, effectively changing the quality parameters of the reference data. This parameter transformation improves the signal-to-noise ratio and feature visibility, enabling more accurate tracking without increasing the complexity of the cross-correlation algorithm itself.
3Object-affected harmful factors
If low infrared illumination is used, then patient comfort and safety are improved, but image quality deteriorates
Solution Approach 1:
The patent merges information from multiple low-illumination frames through accumulation to create a reference image with sufficient signal quality. This approach enables the system to maintain low infrared exposure levels while still achieving the image quality necessary for accurate tracking by combining weak signals across multiple acquisitions.
Solution Approach 2:
The continuous accumulation process allows the system to integrate signal information over time, enabling the use of lower illumination levels per frame while maintaining overall image quality through temporal integration of multiple frames.
Data Source
AI summary
A method of processing a sequence of images of a retina acquired by an ophthalmic device to generate retinal position tracking information indicative of retina movement during acquisition. The method includes (i) receiving one or more images of the retina; (ii) calculating a cross-correlation between a reference image and an image based on the received image(s) to acquire an offset between the image and reference image; and repeating processes (i) and (ii) to acquire, as the tracking information, respective offsets for images that are based on the respective received image(s). Another step includes modifying the reference image during the repeating, by determining a measure of similarity between correspondingly located regions of pixels in two or more received images and accentuating features in the reference image representing structures of the imaged retina in relation to other features in the reference image based on the determined measure of similarity.