Optimisation of fishing predictions by means of artificial neural networks, anfis, functional networks and remote sensing images

摘要: This article presents the application of various Artificial Intelligence techniques to images that proceed from Remote Sensing and serve predict Prionace Glauca captures (the is a type shark). Our data remote sensors whose spectral signature allows us calculate products are useful for ecological modelling. After digitally processing images, we created database which extract necessary patterns training artificial neural networks (Backpropagation network, RBF, functional separability network) neuro-diffuse (ANFIS). These used our system with aforementioned algorithms. The results show this problems generalisation capacity reduced, probably due absence subjacent mathematical model. Finally, implementation was carried out multilayer perceptron trained Backpropagation algorithm (error backpropagation); method less complicated than ANFIS RBF networks.