Pattern Recognition And Diagnosis Of Retina OCT Image

Optical Coherence Tomography is a recently developed imaging technology. With OCT technology one could obtain clear tomography pictures of retina and macula lutea. The unique capability of OCT is to explore microstructure within transparent media by using backscattered or backreflected light. OCT performs high resolution, non-invasion.A preprocessing and identification technique of OCT image of retina has been developed and demonstrated in this paper. According to the characteristics of OCT image, integrated image segmentation has been adopted to better detect contour area. The edge and contour were extracted automatically through image segmentation and enhancement. The feature extraction is performed by K-L transform. Comparing and analyzing several identification plans of OCT image, we obtained different results by using varied processing method in image preprocessing, feature extracting and identification. Finally, the optimal project is established.Processing is performed with MATLAB 7.0. The automatic diagnosis of eye ground diseases is realized. It is very significant to improve OCT system and enhance the usability.