Collection of multispectral images of the fundus

School of Computer Science, The University of Birmingham

These images have been collected as a part of a research project

Physics based image interpretation to aid the detection of early signs of retinopathies

funded by the EPSRC Grant GR/S09906/01.

Each set comprises 17 images at the following wavelengths (in nm):

480, 507, 515, 525, 540, 548, 552, 560, 565, 569, 575, 585, 596, 611, 640, 675 and 705.

Images were acquired using a fundus camera (Zeiss RCM250, 40 degrees field of view), modified as follows. The camera port was equipped with bespoke corrective optics. The standard flash illumination was disabled and camera optics modified to enable fundus illumination with a halogen viewing lamp. Images were recorded on Retiga EXi 12-bit monochrome digital camera (Qimaging, Canada) after passing through VariSpec tunable LCD optical filter (Cambridge Research Instruments, USA). Images were normalised by removing the effects of the image acquisition system [1] . To compensate for eye movements, images were registered using specialised fundus registration software [2].

Shown below are the thumbprints of images at 560 nm from each data set. The images have been acquired from healthy young volunteers from a variety of ethnic backgrounds including Caucasians, African, Afro-Carribean and Mediterranean (first two rows) and from patients with a range of retinopathies.

The images are freely available to academic researchers. If you wish to receive a set, please email


1. Styles IB, Calcagni A, Claridge E, Orihuela Espina F, Gibson JM (submitted) Quantitative analysis of multispectral fundus images. Medical Image Analysis . Also available as Technical Report CSR-05-08, School of Computer Science, The University of Birmingham, 2005. CSR-05-08.pdf

2. Stewart CV, Tsai C-L, Roysam B (2003) The dual-bootstrap iterative closest point algorithm with application to retinal image registra-tion. IEEE Trans. Med. Im. 22, 1379-1394.

Page maintained by Ela Claridge