Estimation of mean grain size of seafloor sediments using neural network
- PDF / 337,397 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 4 Downloads / 197 Views
ORIGINAL RESEARCH PAPER
Estimation of mean grain size of seafloor sediments using neural network Chanchal De • Bishwajit Chakraborty
Received: 7 October 2011 / Accepted: 28 December 2011 / Published online: 13 January 2012 Ó Springer Science+Business Media B.V. 2012
Abstract The feasibility of an artificial neural network based approach is investigated to estimate the values of mean grain size of seafloor sediments using four dominant echo features, extracted from acoustic backscatter data. The acoustic backscatter data were collected using a dual-frequency (33 and 210 kHz) single-beam, normal-incidence echo sounder at twenty locations in the central part of the western continental shelf of India. Statistically significant correlations are observed between the estimated average values of mean grain size of sediments and the ground-truth data at both the frequencies. The results indicate that once a multi-layer perceptron model is trained with back-propagation algorithm, the values of mean grain size can reasonably be estimated in an experimental area. The study also revealed that the consistency among the estimated values of mean grain size at different acoustic frequencies is considerably improved with the neural network based method as compared to that with a model-based approach. Keywords Mean grain size Seafloor sediments Neural networks Acoustic backscatter Echo features
Introduction A quantitative knowledge of mean grain size of seafloor sediment is of great importance for a wide range of C. De (&) G-FAST, P-1, Metcalfe House, Delhi 110 054, India e-mail: [email protected] B. Chakraborty National Institute of Oceanography, Council of Scientific and Industrial Research, Dona Paula, Goa 403 004, India e-mail: [email protected]
applications in the field of marine geology, marine engineering, hydrographic, and environmental monitoring. The most reliable and accurate assessment on mean grain size of seafloor sediments can be obtained from the laboratory analyses of sediment samples. However, collection of sediment samples with grabs and/or cores or in situ measurements is expensive as well as time consuming process. In addition, this conventional approach can give information on the seafloor characteristics only at pre-selected discrete locations in an experimental area. As an alternative approach, remote sensing by acoustic means has long been recognized as a rapid and cost-effective method for characterization and classification of seafloor sediments over a wide area of interest. The acoustic remote sensing essentially relies on the backscatter strength of the acoustic signal reflected from the seafloor. Since the backscatter strength contains information on the properties of the material at which the signal is scattered, it can be used for assessing the mean grain size of seafloor sediments. Acoustic backscatter data, obtained from common seafloor depth measurements equipments such as single-beam and multi-beam echo sounders, could be used for this purpose. A number of approaches concerning the
Data Loading...
