It is known that one of the most critical issues for the implementation of a fully automatic processing dedicated to the detection of oil spills from SAR imagery is the extraction of the oil spill candidate. In fact, the segmentation of the image is the first of three necessary steps, the other two being the characterization of the extracted black spot by using a set of features and the classification between oil spill and look-alike. In this paper we investigate an unsupervised neural network approach for automatically extracting oil spill candidates from ERS and ENVISAT SAR images. The technique is based on the use of Pulse-Coupled Neural Networks (PCNN) which is a relatively novel technique based on models of the visual cortex of small mammals. When applied to image processing, it yields a series of binary pulsed signals, each associated to one pixel or to a cluster. In literature, interesting results have been already reported by several authors in applications of this model to image segmentation, including, in few cases, the use of satellite data. The architecture of PCNN is rather simpler than most other neural network implementations. PCNN do not have multiple layers and receive input directly from the original image, forming a resulting “pulse” image. The network consists of multiple nodes coupled together with their neighbors within a definite distance, forming a grid (2D-vector). The PCNN neuron has two input compartments: linking and feeding. The feeding compartment receives both an external and a local stimulus, whereas the linking compartment only receives a local stimulus. When the internal activity becomes larger than an internal threshold, the neuron fires and the threshold sharply increases. Afterward, it begins to decay until once again the internal activity becomes larger. This process gives rise to the pulsing nature of PCNN, forming a wave signature which is invariant to rotation, scale, shift or skew of an object within the image. This study discusses the use of PCNN technique in a fully automatic chain for oil spill detection from SAR images. The objects segmented by the PCNN are successively processed by a more standard Multi-Layer Perceptron Neural Network, which provides the classification response between real oil spill and look-alike. The performance yielded by the PCNN-MLP chain is evaluated and critically discussed for a set of ERS-SAR and ENVISAT ASAR images. The application of the methodology to the very-high resolution SAR images taken by COSMO-Skymed and TerraSAR-X satellites will be also considered.
Del Frate, F., Latini, D., & Pratola, C. (2010). Pulse coupled neural networks for automatic oil spill detection from satellite SAR images. Paper presented at ESA SEASAR workshop 2010, Frascati, Rome, Italy.
|Autori:||Del Frate, F; Latini, D; Pratola, C|
|Titolo:||Pulse coupled neural networks for automatic oil spill detection from satellite SAR images|
|Nome del convegno:||ESA SEASAR workshop 2010|
|Luogo del convegno:||Frascati, Rome, Italy|
|Anno del convegno:||2010|
|Enti collegati al convegno:||ESA|
|Data di pubblicazione:||2010|
|Settore Scientifico Disciplinare:||Settore ING-INF/02 - Campi Elettromagnetici|
|Tipologia:||Intervento a convegno|
|Citazione:||Del Frate, F., Latini, D., & Pratola, C. (2010). Pulse coupled neural networks for automatic oil spill detection from satellite SAR images. Paper presented at ESA SEASAR workshop 2010, Frascati, Rome, Italy.|
|Appare nelle tipologie:||02 - Intervento a convegno|