KNN ALGORITHM IN A FRAMEWORK OF SCALE-SPACE THEORY FOR RETINAL IMAGE ANALYSIS
Keywords:
Feature stability, K-nearest neighbor, object detection: scale-spaceAbstract
This topic presents a framework in the uses of the K-nearest neighbor algorithm in evaluating anobject detection method of scale-space theory with feature stability. A scale-space tree is constructedbased on the blobs that were created from a series of images after the blurring process. Features and spatial information provide the role within the scale-space tree construction. After the process of blob extraction, users determine each type of the blob that was detected within the image by distinguishing classes to create ground truth image data. Within the same process, the KNN algorithm is applied to distinguish classes of the image’s blobs in determining the performance.
References
Bowyer, K., Kranenburg, C., and Dougherty, S. (2001). Edge detector evaluation using empirical roc curves. Comput. Vis. Image Und., 84:77-103.
Carvalho, M.A.G., Lotufo, R.A., and Couprie, M. (2003). Segmentation of images of yeast cells by scale-space analysis. XVI Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI’03); 12-15 Oct, 2003; Sao Carlos, Brazil; p. 376-380.
Duanggate, C., Uyyanonvara, B., Makhanov, S.S., Barman, S., and Williamson, T. (2011). Parameter-free optic disc detection. Comput. Med. Imag. Grap., 35(1):51-63.
Forbes, L.A. and Draper, B.A. (2000). Inconsistencies in edge detector evaluation. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition; 13-15 June, 2000; Hilton Head Island, SC, USA; 2:398-404.
Inthajak, K., Duanggate, C., Uyyanonvara, B., Makhanov, S.S., Barman, S., and Williamson, T. (2010). Variable size blob detection with feature stability. Proceedings of the 3rd Biomedical Engineering International Conference on Medical Imaging, Image and Signal Processing; 27-28 Aug, 2010; Kyoto, Japan; p. 133-137.
Jalba, A.C., Wilkinson, M.H.F., and Roerdink, J.B.T.M. (2006). Shape representation and recognition through morphological curvature scale spaces. IEEE T. Image Process., 15:331-341.
Konishi, S., Yuille, A., and Coughlan, J. (2003). Statistical edge detection: Learning and evaluating edge cues. IEEE T. Pattern Anal., 25:57-74.
Lindeberg, T. (1994). Scale-Space Theory in Computer Vision. Kluwer Int Ser. Eng. C. Springer-Verlag, Berlin, Germany, 423p.
Lindeberg, T. (1998a). Edge detection and ridge detection with automatic scale selection. Int. J. Comput. Vision, 30:117-156.
Lindeberg, T. (1998b). Feature detection with automatic scale selection. Int. J. Comput. Vision, 30: 79-116.
Plaku, E. and Kavrati, L.E. (2007). Distributed computation of the knn graph for large high-dimensional point sets. J. Parallel Distr. Com., 67(3): 346-359.
Su, M-Y. (2011). Using clustering to improve the KNN-based classifiers for online anomaly network traffic identification. J. Netw. Comput. Appl., 34(2):722-730.
Tekmono, K. (2006). K-Nearest Neighbors Tutorial. http://people.revoledu.comkarditutorialKNN, Accessed date: June 8, 2010.
Thomas, B.F. (2008). K-Nearest Neighbors Algorithm: Prediction and Classification. http://faculty. smu. edu/tfomby/eco5385/lecture/K-NN%20 Method.pdf, Accessed date: April 14, 2012.
Wang, S., Ma, X., Zhang, X., and Jiao, L. (2007). Watershed-based textural image segmentation. Proceedings of International Symposium on Intelligent Signal Processing and Communication Systems; 28 Nov-1 Dec, 2007; Xiamen, China; p. 312-315.
Witkin, A.P. (1983). Scale-space filtering. Proceedings of the 8th International Joint Conference on Artificial Intelligence; Aug 8-12, 1983; Kalsruhe, Germany; p. 1019-1022.
