An Extension of Self-Organizing Maps to Categorical Data

Self-organizing maps (SOM) have been recognized as a powerful tool in data exploratoration, especially for the tasks of clustering on high dimensional data. However, clustering on categorical data is still a challenge for SOM. This paper aims to extend standard SOM to handle feature values of categorical type. A batch SOM algorithm (NCSOM) is presented concerning the dissimilarity measure and update method of map evolution for both numeric and categorical features simultaneously.

Automated, objective texture segmentation of multibeam echosounder data - Seafloor survey and substrate maps from James Island to Ozette Lake, Washington Outer Coast

Without knowledge of basic seafloor characteristics, the ability to address any number of critical marine and/or coastal management issues is diminished. For example, management and conservation of essential fish habitat (EFH), a requirement mandated by federally guided fishery management plans (FMPs), requires among other things a description of habitats for federally managed species. Although the list of attributes important to habitat are numerous, the ability to efficiently and effectively describe many, and especially at the scales required, does not exist with the tools currently available. However, several characteristics of seafloor morphology are readily obtainable at multiple scales and can serve as useful descriptors of habitat. Recent advancements in acoustic technology, such as multibeam echosounding (MBES), can provide remote indication of surficial sediment properties such as texture, hardness, or roughness, and further permit highly detailed renderings of seafloor morphology. With acoustic-based surveys providing a relatively efficient method for data acquisition, there exists a need for efficient and reproducible automated segmentation routines to process the data. Using MBES data collected by the Olympic Coast National Marine Sanctuary (OCNMS), and through a contracted seafloor survey, we expanded on the techniques of Cutter et al. (2003) to describe an objective repeatable process that uses parameterized local Fourier histogram (LFH) texture features to automate segmentation of surficial sediments from acoustic imagery using a maximum likelihood decision rule. Sonar signatures and classification performance were evaluated using video imagery obtained from a towed camera sled. Segmented raster images were converted to polygon features and attributed using a hierarchical deep-water marine benthic classification scheme (Greene et al. 1999) for use in a geographical information system (GIS). (PDF contains 41 pages.)

Detecting fish aggregations from reef habitats mapped with high resolution side scan sonar imagery

As part of a multibeam and side scan sonar (SSS) benthic survey of the Marine Conservation District (MCD) south of St. Thomas, USVI and the seasonal closed areas in St. Croix—Lang Bank (LB) for red hind (Epinephelus guttatus) and the Mutton Snapper (MS) (Lutjanus analis) area—we extracted signals from water column targets that represent individual and aggregated fish over various benthic habitats encountered in the SSS imagery. The survey covered a total of 18 km2 throughout the federal jurisdiction fishery management areas. The complementary set of 28 habitat classification digital maps covered a total of 5,462.3 ha; MCDW (West) accounted for 45% of that area, and MCDE (East) 26%, LB 17%, and MS the remaining 13%. With the exception of MS, corals and gorgonians on consolidated habitats were significantly more abundant than submerged aquatic vegetation (SAV) on unconsolidated sediments or unconsolidated sediments. Continuous coral habitat was the most abundant consolidated habitat for both MCDW and MCDE (41% and 43% respectively). Consolidated habitats in LB and MS predominantly consisted of gorgonian plain habitat with 95% and 83% respectively. Coral limestone habitat was more abundant than coral patch habitat; it was found near the shelf break in MS, MCDW, and MCDE. Coral limestone and coral patch habitats only covered LB minimally. The high spatial resolution (0.15 m) of the acquired imagery allowed the detection of differing fish aggregation (FA) types. The largest FA densities were located at MCDW and MCDE over coral communities that occupy up to 70% of the bottom cover. Counts of unidentified swimming objects (USOs), likely representing individual fish, were similar among locations and occurred primarily over sand and shelf edge areas. Fish aggregation school sizes were significantly smaller at MS than the other three locations (MCDW, MCDE, and LB). This study shows the advantages of utilizing SSS in determining fish distributions and density.

Use of generalised additive models to categorise continuous variables in clinical prediction

13 P.

High-Strength Composite Fibers: Realizing True Potential of Carbon Nanotubes in Polymer Matrix through Continuous Reticulate Architecture and Molecular Level Couplings

Carbon nanotubes have unprecedented mechanical properties as defect-free nanoscale building blocks, but their potential has not been fully realized in composite materials due to weakness at the interfaces. Here we demonstrate that through load-transfer-favored three-dimensional architecture and molecular level couplings with polymer chains, true potential of CNTs can be realized in composites as Initially envisioned. Composite fibers with reticulate nanotube architectures show order of magnitude improvement in strength compared to randomly dispersed short CNT reinforced composites reported before. The molecular level couplings between nanotubes and polymer chains results in drastic differences in the properties of thermoset and thermoplastic composite fibers, which indicate that conventional macroscopic composite theory falls to explain the overall hybrid behavior at nanoscale.