FEATURES CLASSIFIER image based on ART1-NETWORK
Main Article Content
Abstract
In order to create a medical image recognition system, we should take into account the range of possible changes of input signal that comes from the object. In this regard, the main requirement for pattern recognition is to provide a classifier which would be invariant under various transformations. The problem of image classification is solved experimentally based on the ART1-network system designed in MATLAB environment. It has been established that: 1) in case of not more than 40% of noise, for image classification ART1-network selects recorded in the associative memory prototype vector which is most correlated therewith; 2) the associative memory based on ART1-network performs equivalently to the associative memory in the form of a single layer binary linear associator based on pseudoinverse learning rule and it is unable to perform the functions of the invariant classifier.
Reference 6, figures 5, tables 6.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
References
Albert, A. (1972). Regression and the Moore-Penrose Pseudoinverse, New York: Academic Press. – 180 p.
Carpenter, G. A. and Grossberg, S. (1987). A massively parallel architecture for a self-organizing neu-ral pattern recognition machine. Computer Vision, Graphics, and Image Processing, vol. 37, pp. 54 – 115
Martin Hagan, Howard Demuth, B. Mark. Beale (2002). Neural Network Design. USA: Colorado Uni-versity Bookstore, Р. 734.
Dobrovskaya, L. M., Dobrovskaya, I. A. (2015). Theory and practice of neural networks, Ukrainian: NTU "KPI" Publisher Polytechnic, Р. 396. (Ukr.)
Haykin, S. (2006). Neural networks: a complete course. 2nd ed. Moscow, Williams Publ., Р. 1104. (Rus.).
Fadeev, D. K., Faddeev, V. N. (2002). Computational methods of linear algebra. 3rd ed., Sr. - St. Pe-tersburg: Lan, Р. 736. (Rus.).
