Slit-lamp based video mosaicking

To this day, the slit lamp remains the first tool used by an ophthalmologist to examine patient eyes. Imaging of the retina poses however a variety of problems, namely, a shallow depth of focus, reflections from the optical system, a small field of view and non-uniform illumination. For ophthalmologists, the use slit-lamp images for documentatation and analysis purposes however remain extremely challenging due to large image artifacts. For this reason, we propose an automatic retinal slit lamp video mosaicking, which enlarges the field of view and reduces amount of noise and reflections, thus enhancing image quality. Our method is composed of three parts: i) viable content segmentation, ii) global registration and iii) image blending. Frame content is segmented using gradient boosting with custom pixel-wise features. SURF is used to find pairwise translations between frames with robust RANSAC estimation and graph-based  SLAM for global bundle adjustment. Foreground-aware blending based on feathering merges video frames into comprehensive mosaic. Mosaicking results and state-of-the-art methods were compared and rated by ophthalmologists showing a strong preference for a large field of view provided by our method.


Full mosaic of slit lamp sequence of 1000 frames width a large field of view. The area covered by the separately acquired fundus photography (center) on the mosaic can be seen as a white circle. On the right an example of the algorithm by Richa et al. (TMI, 2014) is depicted showing the maximal extend of mosaic that can be achieved in this video sequence. Its area covered is delineated in the mosaic.