Search

In the electronic nose (e-nose), a stable feature representation of the gas sensor’s response is a key step to realize subsequent odor identification algorithms. However, the noises in gas sensors hinder the acquisition of such features. In order to solve this problem, this article proposes a stable feature extraction algorithm which takes the impulse response of the e-nose system as the feature. The impulse response is estimated from a nonparametric model constrained by a multiscale wavelet kernel regularization matrix. The kernel regularization matrix equips the proposed feature extraction method with an ability in resistance to random noise. A numerical experiment proves that compared with single-scale kernel regularization, the use of multiscale wavelet kernel helps to achieve more stable and accurate impulse response estimation. Then, a field experiment is conducted to demonstrate the performance of the proposed features. This experiment aims to identify four different whiskies measured by a self-designed e-nose with four commercial gas sensors. Under the framework of transfer learning, the classification result based on the proposed features outperforms those using other considered features. The accuracy of whisky identification reaches 92.00%, showing a good potential of applying the proposed feature representations in the area of e-noses. [ABSTRACT FROM AUTHOR]

The absence or abnormality of fidgety movements of joints or limbs is strongly indicative of cerebral palsy in infants. Developing computer-based methods for assessing infant movements in videos is pivotal for improved cerebral palsy screening. Most existing methods use appearance-based features and are thus sensitive to strong but irrelevant signals caused by background clutter or a moving camera. Moreover, these features are computed over the whole frame, thus they measure gross whole body movements rather than specific joint/limb motion. Addressing these challenges, we develop and validate a new method for fidgety movement assessment from consumer-grade videos using human poses extracted from short clips. Human poses capture only relevant motion profiles of joints and limbs and are thus free from irrelevant appearance artifacts. The dynamics and coordination between joints are modeled using spatio-temporal graph convolutional networks. Frames and body parts that contain discriminative information about fidgety movements are selected through a spatio-temporal attention mechanism. We validate the proposed model on the cerebral palsy screening task using a real-life consumer-grade video dataset collected at an Australian hospital through the Cerebral Palsy Alliance, Australia. Our experiments show that the proposed method achieves the ROC-AUC score of 81.87%, significantly outperforming existing competing methods with better interpretability. [ABSTRACT FROM AUTHOR]

In recent years, the native food industry in Australia has increased in both value and volume due to the discovery of a wide range of phytochemicals (e.g., vitamin C, polyphenols) that have potential health benefits. Thus, plant organs and tissues of these native plants are used in a wide range of applications. In particular, the kernel of a native plum, the Kakadu plum (Terminalia ferdinandiana, Combretaceae) is considered to be rich in lipids and other phytochemical compounds. The aim of this study was to evaluate the use of NIR spectroscopy to analyze and characterize kernel samples and tissues of wild harvest fruit samples. The Fourier transform near-infrared reflectance spectra of cracked kernels, seeds cover tissues, and dry powder Kakadu plum kernels were acquired. Both principal component analysis and partial least squares discriminant analysis were used to analyze and interpret the spectral data. A correct classification rate of 93%, 86%, and 80% was achieved for the identification of kernel provenance using all tissues, seed coats, and the whole nuts, respectively. The results of this study reported for the first time the analysis of Kakadu plum kernels and their tissues using NIR spectroscopy. Graphical Abstract [ABSTRACT FROM AUTHOR]

The Australian almond industry desires to improve storage of harvested almonds awaiting processing. The present work quantified some storage-related physical properties of almonds. The sample used in the study was Nonpareil almonds with a kernel moisture content of 4.5% d.b. The mass composition of the sample was 55% hull, 32% kernel and 13% shell. Tests showed that the bulk stored in-shell almonds had only 41% of the volume of in-hull almonds and 45% of the mass. Thus removing hulls before storage would result in saving both storage and subsequent transportation costs. Tests simulating various storage heights of almonds showed that a 10 m storage height of almonds increased the bulk density of in-hull almonds from 320 to 355 kg/m3, of in-shell almonds (hull removed) from 356 to 378 kg/m3, and kernels (hull and shell removed) from 604 to 649 kg/m3. A 10 m storage height of almonds reduced the porosity of in-hull almonds from 67% to 64%, of in-shell almonds from 58% to 55%, and of the kernel from 48% to 44%. Observation showed that the change in bulk density and porosity occurred in an exponential manner with fitted curves that provided R2 between 0.97 and 0.99 [ABSTRACT FROM AUTHOR]

The only specialized ant-eating mammal in Australia and New Guinea is the egg-laying short-beaked echidna (Tachyglossus aculeatus), and this single species occurs throughout Australia in a wide range of habitats. Despite the diversity of habitats and density and distribution of prey species, home-range sizes throughout Australia seem remarkably similar. We radiotracked echidnas in a population in Tasmania over a 13-year period and calculated home-range sizes using the fixed kernel method and the minimum convex polygon method. No relationship was found between body mass and home-range size, and mean annual home-range size of males (90% kernels) was 107 ha ± 48 SD, twice that of females (48 ± 28 ha). Male home ranges overlapped considerably and also overlapped with those of several females. The echidna follows the pattern seen in many solitary eutherian mammals: both sexes are promiscuous, and males have larger home ranges than females. Echidnas show a high degree of home-range fidelity but can make rare excursions out of their normal area. Hibernating echidnas move between shelters during their periodic arousals, resulting in home-range sizes similar to those of the active period. Consistent with their very low metabolic rate, echidnas have home-range sizes considerably smaller than predicted for carnivorous or omnivorous mammals. Examination of data from other ant-eating mammals shows that as a group anteaters not only have smaller than predicted home ranges but they depart significantly from the normal relationship between home-range size and body mass. [ABSTRACT FROM AUTHOR]

Sedentary Australian owlet-nightjars Aegotheles cristatus roost in cavities year-round, and are thought to mate for life and maintain stable home ranges throughout the year. These factors lead to the prediction that they should be highly territorial, yet nothing is known about their home range requirements or level of site fidelity. I used radio-telemetry coupled with GPS recordings to determine home range size in the semi-arid zone of central Australia (13 birds over two winters) and in a eucalypt woodland in the Northern Tablelands of NSW (14 birds over one calendar year). The mean home range in the eucalypt woodland was 17.7 ha based on the minimum convex polygon (MCP) method and 17.4 ha based on the fixed kernel contour (95%), whereas it was 23.8 ha and 24.1 ha based on the MCP and kernel methods in the desert respectively. With the exception of mated pairs (range overlap 41.9%), there was little overlap among individuals in home range areas (<13.0%), even in the densely populated woodland. Home range size did not differ significantly between seasons or study sites; nor was it correlated with arthropod abundance. Owlet-nightjars exhibited high site fidelity, using the same home range throughout the year with a 68% overlap between seasons. These data, plus anecdotal records of aggression and vocalizations, suggest that this species is highly territorial, yet individuals live in close proximity to conspecifics. The degree of territoriality may be due to the obligate cavity roosting behaviour of owlet-nightjars and the need to defend areas with known roost hollows. [ABSTRACT FROM AUTHOR]