Skip to Main Content
Retinal fundus images acquired with non-mydriatic digital fundus cameras are a versatile tool for the diagnosis of various retinal diseases. Even with relative ease of use, the images produced sometimes suffer from reflectance artefacts mainly due to the nerve fibre layer (NFL) or camera lens related reflections. We propose a technique that employs multiple fundus images to obtain a single higher quality image without these reflectance artefacts, which also compensates for a sub- optimal illumination. The removal of bright artefacts, can have great benefits for the reduction of false positives in the detection of retinal lesions by automatic systems or manual inspection. The fundus images are acquired by changing the stare point of the patient but keeping the camera fixed. Between each shot, the apparent shape and position of all the retinal structures that do not exhibit isotropic reflectance (e.g. bright artefacts) change. This physical effect is exploited by our algorithm.