Neural networks for optic nerve detection in digital optic fundus images

This paper presents a method used for detection of optic nerve in fundus digital images; for this purpose, initially there is a preprocessing and segmentation of digital images of fundus taken from databases Messidor and Stare in order to stand out the veins and blood vessels of ocular region. The processed image is used in an artificial neural network which has three layers; an input layer with 10000 neurons that belong to the vectorized and processed image, a hidden layer with 100 neurons and an output layer with 42 neurons, which define the pixel deviation between the center of the ROI and the original image. For the training of the Artificial Neural Network, is used an arrangement of fifteen selected images from databases and scaled to a suitable size, so it can fit inside the ROI; this training was performed using backpropagation. At the end of the training, there were two types of experiments for result's validation; the first one was the detection of optic discs in known databases with an error of 0%, the second one was the detection of optic discs in unknown databases, with an error of 5% during the tracking of optic nerve, thus it concludes that the presented method is efficient for the detection of optic disc in fundus digital images.

[1]  O. Chang,et al.  Artificial Vision Techniques to Optimize Strawberry's Industrial Classification , 2016, IEEE Latin America Transactions.

[2]  Anushikha Singh,et al.  Blood vessel inpainting based technique for efficient localization and segmentation of optic disc in digital fundus images , 2016, Biomed. Signal Process. Control..

[3]  Oscar Chang,et al.  Reliable object recognition by using cooperative neural agents , 2014, 2014 International Joint Conference on Neural Networks (IJCNN).

[4]  Huiqi Li,et al.  Automated feature extraction in color retinal images by a model based approach , 2004, IEEE Transactions on Biomedical Engineering.

[5]  Juan Xu,et al.  Optic disk feature extraction via modified deformable model technique for glaucoma analysis , 2007, Pattern Recognit..

[6]  E. R. Davies.  Computer and Machine Vision: Theory, Algorithms, Practicalities , 2012 .

[7]  Ahmed S. Fahmy,et al.  Fast Localization of the Optic Disc Using Projection of Image Features , 2010, IEEE Transactions on Image Processing.

[8]  Jarmo T. Alander,et al.  Integer-based accurate conversion between RGB and HSV color spaces , 2015, Comput. Electr. Eng..

[9]  A.M. Aibinu,et al.  Detection of vascular intersection in retina fundus image using modified cross point number and neural network technique , 2008, 2008 International Conference on Computer and Communication Engineering.

[10]  Guy Cazuguel,et al.  FEEDBACK ON A PUBLICLY DISTRIBUTED IMAGE DATABASE: THE MESSIDOR DATABASE , 2014 .

[11]  A. P. Ruhil,et al.  Prediction of sensory quality of UHT milk : A comparison of kinetic and neural network approaches , 2009 .

[12]  Robert Laganière,et al.  OpenCV 2 computer vision application programming cookbook : over 50 recipes to master this library of programming functions for real-time computer vision , 2011 .

[13]  John C. Russ,et al.  The Image Processing Handbook , 2016, Microscopy and Microanalysis.

[14]  Amir Akramin Shafie,et al.  Vascular intersection detection in retina fundus images using a new hybrid approach , 2010, Comput. Biol. Medicine.

[15]  Jacob Scharcanski,et al.  A morphologic two-stage approach for automated optic disk detection in color eye fundus images , 2013, Pattern Recognit. Lett..

[16]  Q. Y. Peng,et al.  Prediction of texture characteristics from extrusion food surface images using a computer vision system and artificial neural networks , 2013 .

[17]  Nasser Kehtarnavaz,et al.  Computationally efficient optic nerve head detection in retinal fundus images , 2014, Biomed. Signal Process. Control..