Long-term diabetes may lead to a diabetic retinopathy, an eye disease, which is one of themost frequent causes of blindness. Its early symptoms can be detected from the photosof retina, thus, making the investigation into the automatic lesion detection method to bea valuable undertaking. Hyperspectral images provide additional information about thecharacteristics of the imaging target, which may improve accuracy in comparison to sim-ple color images, but at the same time introduce problems of higher dimensionality. Thepurpose of this work is to develop an effective algorithm for detecting diabetic retinopathylesions and find out whether hyperspectral images are beneficial for diabetic retinopathydiagnosis in comparison to colour images. The experiments provide evidence that properalgorithmic use of spectral images allows to achieve significantly better results.
展开▼
