Researchers from the Tokyo Institute of Technology used an artificial intelligence program termed a Convolutional Neural Network. Applying this technology the researchers demonstrated how artificial intelligence can be trained to catalog and analyze volcanic ash particle shapes. The significance is that as the form and shape of volcanic particles are connected to the type of volcanic eruption, this systematic approach can assist scientists in providing information on eruptions. Such insights can aid volcanic hazard mitigation efforts.
In machine learning, a convolutional neural network is a type of deep, feed-forward artificial neural network. This form of artificial intelligence is best equipped for analyzing visual imagery. The process uses multilayer perceptrons, which are designed to require minimal preprocessing. The development of convolutional networks was inspired by biological processes given that the connectivity pattern between neurons is equivalent to the pattern of the animal visual cortex.
It is common for volcanologists look at the ash produced by eruptions for different eruptions generate ash particles of differing shapes. However, the process of examining thousands of tiny samples objectively is a time consuming task and one potentially prone to error. This is where the artificial intelligence presents an advantage.
Trials so far with the artificial intelligence platform have succeeded in teaching the program to classify basal shapes of ash with a success rate of 92 percent and to assign probability ratios to each particle. For the next step the platform will be trained to seek useful, complex information about other tiny particles with vast geological value.
The research has been published in the journal Scientific Reports, with the article titled “Classification of volcanic ash particles using a convolutional neural network and probability.”
