Improved Automatic Classification of Biological Particles from Electron-Microscopy Images Using Genetic Neural Nets

In this paper several neural network classification algorithmshave been applied to a real-world data case of electron microscopy imageclassification in which it was known a priori the existence of twodifferentiated views of the same specimen. Using several labeled sets asa reference, the parameters and architecture of the classifier (both LVQtrained codebooks and BP trained neural-nets) were optimized using agenetic algorithm. The automatic process of training and optimizationis...