CLASSIFICATION OF RED TIDE ALGAE IMAGES USING GENERALIZED FEEDFORWARD NEURAL NETWORK

Abstract

In this paper a new classification algorithm is proposed for the Classification of Red tide algae images. In order to develop algorithm 154 red tide algae SEM images have been considered, So as to create calculation 154 red tide algae SEM image have been considered, With a view to extricate highlights from the red tide green growth SEM images after image preparing, a calculation proposes WHT changed coefficients. The Efficient classifiers dependent on Generalized feed forward neural network(GFF). A different Cross-Validation dataset is utilized for legitimate assessment of the proposed characterization calculation as for essential execution measures, for example, MSE and arrangement exactness. The Average Classification Accuracy of GFF Neural Network including one hidden layers with 20 PE's sorted out in a run of the mill topology is observed to be unrivalled (96.66 %) for Training and crossvalidation. At last, ideal calculation has been created based on the best classifier execution. The calculation will give a powerful option in contrast to customary technique for Red tide algae SEM images examination for Classify the five sort of Red tide algae SEM images.