BOUNDARY ELEMENT METHOD BASED BINARY IMAGES CLASSifiCATION USING LAPLACIAN OPERATOR WITH DIRICHLET BOUNDARY CONDITION

The eigenvalues of the Dirichlet Laplacian are used to generate three different sets ofjeatures for shape recognition and classification in binary images. The generated features are rotation, translation and size-invariant and are shown to be tolerant of boundary deformation. The effectiveness of these features is demonstrated by using them to classify; five--types of computer generated and boundary distorted shapes. The classification was done using 4 to 20 features fed to a simple feed-forward neural network. Correct classification rates ranging from 94. 4% to 100% were obtained on computer generated shapes and 675% to 95.5% on boundary distorted shapes. To simulate the model, a new approach used; the boundary element method.