Intonation contours vs. syntactic structure as predictors of apparent segmentation

This paper is a review of a study on perception and comprehension of speech using syntactic, visual and acoustic information.

Fusion of motion segmentation algorithms

Many algorithms have been developed to achieve motion segmentation for video surveillance. The algorithms produce varying performances under the infinite amount of changing conditions. It has been recognised that individually these algorithms have useful properties. Fusing the statistical result of these algorithms is investigated, with robust motion segmentation in mind.

Infant feeding choices: experience, self-identity and lifestyle

In England 78% of mothers initiate breastfeeding and in the UK less than 1% exclusively breastfeed until 6 months, despite WHO recommendations to do so. This study investigated women’s infant feeding choices using in-depth interviews with 12 mothers of infants aged 7-18 weeks. Using content analysis, four themes emerged: (1) Information, Knowledge and Decision Making, (2) Physical Capability, (3) Family and Social Influences, (4) Lifestyle, Independence and Self-Identity. Whilst women were aware of the ‘Breast is Best’ message, some expressed distrust in this information if they had not been breastfed themselves. Women felt their own infant feeding choice was influenced by the perceived norm amongst family and friends. Women described how breastfeeding hindered their ability to retain their self-identities beyond motherhood as it limited their independence. Several second-time mothers felt they lacked support from health professionals when breastfeeding their second baby, even if they had previously encountered breastfeeding difficulties. The study indicates that experience of breastfeeding, and belief in the health benefits associated with it are important factors for initiation of breastfeeding, whilst decreased independence and self-identity may influence duration of breastfeeding. Intervention and support schemes should tackle all mothers, not just first-time mothers.

Effects of body size and lifestyle on evolution of mammal life histories

It has recently been proposed that life-history evolution is subject to a fundamental size-dependent constraint. This constraint limits the rate at which biomass can be produced so that production per unit of body mass is inevitably slower in larger organisms than in smaller ones. Here we derive predictions for how changes in body size and production rates evolve in different lifestyles subject to this constraint. Predictions are tested by using data on the mass of neonate tissue produced per adult per year in 637 placental mammal species and are generally supported. Compared with terrestrial insectivores with generalized primitive traits, mammals that have evolved more specialized lifestyles have divergent massspecific production rates: (i) increased in groups that specialize on abundant and reliable foods: grazing and browsing herbivores (artiodactyls, lagomorphs, perissoclactyls, and folivorous rodents) and flesh-eating marine mammals (pinnipeds, cetaceans); and (ii) decreased in groups that have lifestyles with reduced death rates: bats, primates, arboreal, fossorial, and desert rodents, bears, elephants, and rhinos. Convergent evolution of groups with similar lifestyles is common, so patterns of productivity across mammalian taxa reflect both ecology and phylogeny. The overall result is that groups with different lifestyles have parallel but offset relationships between production rate and body size. These results shed light on the evolution of the fast-slow life-history continuum, suggesting that variation occurs along two axes corresponding to body size and lifestyle.

Characterisation of an amphioxus Fringe gene and the evolution of the vertebrate segmentation clock

In mouse and chick embryos, cyclic expression of lunatic fringe has an important role in the regulation of mesoderm segmentation. We have isolated a Fringe gene from the protochordate amphioxus. Amphioxus is the closest living relative of the vertebrates, and has mesoderm that is definitively segmented in a manner that is similar to, and probably homologous with, that of vertebrates. AmphiFringe is placed basal to vertebrate Fringe genes in molecular phylogenetic analyses, indicating that the duplications that formed radical-, manic- and lunatic fringe are specific to the vertebrate lineage. AmphiFringe expression was detected in the anterior neural plate of early neurulae, where it resolved into a series of segmental patches by the mid-neurulae stage. No AmphiFringe transcripts were detected in the mesoderm. Based on these observations, we propose a model depicting a successive recruitment of Fringe in the maintenance then regulation of segmentation during vertebrate evolution.

Texture segmentation in Williams syndrome

Williams syndrome (WS) is a developmental disorder in which visuo-spatial cognition is poor relative to verbal ability. At the level of visuo-spatial perception, individuals with WS can perceive both the local and global aspects of an image. However, the manner in which local elements are integrated into a global whole is atypical, with relative strengths in integration by luminance, closure, and alignment compared to shape, orientation and proximity. The present study investigated the manner in which global images are segmented into local parts. Segmentation by seven gestalt principles was investigated: proximity, shape, luminance, orientation, closure, size (and alignment: Experiment I only). Participants were presented with uniform texture squares and asked to detect the presence of a discrepant patch (Experiment 1) or to identify the form of a discrepant patch as a capital E or H (Experiment 2). In Experiment 1, the pattern and level of performance of the WS group did not differ from that of typically developing controls, and was commensurate with the general level of non-verbal ability observed in WS. These results were replicated in Experiment 2, with the exception of segmentation by proximity, where individuals with WS demonstrated superior performance relative to the remaining segmentation types. Overall, the results suggest that, despite some atypical aspects of visuo-spatial perception in WS, the ability to segment a global form into parts is broadly typical in this population. In turn, this informs predictions of brain function in WS, particularly areas V1 and V4. (c) 2006 Elsevier Ltd. All rights reserved.

Unsupervised segmentation using Gabor wavelets and statistical features in LIDAR data analysis

In this paper, we address issues in segmentation Of remotely sensed LIDAR (LIght Detection And Ranging) data. The LIDAR data, which were captured by airborne laser scanner, contain 2.5 dimensional (2.5D) terrain surface height information, e.g. houses, vegetation, flat field, river, basin, etc. Our aim in this paper is to segment ground (flat field)from non-ground (houses and high vegetation) in hilly urban areas. By projecting the 2.5D data onto a surface, we obtain a texture map as a grey-level image. Based on the image, Gabor wavelet filters are applied to generate Gabor wavelet features. These features are then grouped into various windows. Among these windows, a combination of their first and second order of statistics is used as a measure to determine the surface properties. The test results have shown that ground areas can successfully be segmented from LIDAR data. Most buildings and high vegetation can be detected. In addition, Gabor wavelet transform can partially remove hill or slope effects in the original data by tuning Gabor parameters.

Optimization algorithms in FMRF model-based segmentation for LIDAR data and co-registered bands

In this paper, a fuzzy Markov random field (FMRF) model is used to segment land-objects into free, grass, building, and road regions by fusing remotely, sensed LIDAR data and co-registered color bands, i.e. scanned aerial color (RGB) photo and near infra-red (NIR) photo. An FMRF model is defined as a Markov random field (MRF) model in a fuzzy domain. Three optimization algorithms in the FMRF model, i.e. Lagrange multiplier (LM), iterated conditional mode (ICM), and simulated annealing (SA), are compared with respect to the computational cost and segmentation accuracy. The results have shown that the FMRF model-based ICM algorithm balances the computational cost and segmentation accuracy in land-cover segmentation from LIDAR data and co-registered bands.