Detection and classification of road signs in natural environments

An automatic road sign recognition system first locates road signs within images captured by an imaging sensor on-board of a vehicle, and then identifies the detected road signs. This paper presents an automatic neural-network-based road sign recognition system. First, a study of the existing road sign recognition research is presented. In this study, the issues associated with automatic road sign recognition are described, the existing methods developed to tackle the road sign recognition problem are reviewed, and a comparison of the features of these methods is given. Second, the developed road sign recognition system is described. The system is capable of analysing live colour road scene images, detecting multiple road signs within each image, and classifying the type of road signs detected. The system consists of two modules: detection and classification. The detection module segments the input image in the hue-saturation-intensity colour space, and then detects road signs using a Multi-layer Perceptron neural-network. The classification module determines the type of detected road signs using a series of one to one architectural Multi-layer Perceptron neural networks. Two sets of classifiers are trained using the Resillient-Backpropagation and Scaled-Conjugate-Gradient algorithms. The two modules of the system are evaluated individually first. Then the system is tested as a whole. The experimental results demonstrate that the system is capable of achieving an average recognition hit-rate of 95.96% using the scaled-conjugate-gradient trained classifiers.

Determination of plastic yield stress from spherical indentation slope curve

The indentation slope curve from a spherical indentation on elastic-plastic materials is examined. By comparing it with that of an linear elastic material of the same elastic properties, we found that the start point of plastic yielding for an elastic-plastic material can be easily located from the indentation slope curve. Based on this analysis, a simple but effective method is proposed to measure the plastic yield stress of very small samples from a spherical nano-indentation slope curve.

Casting yield improvement in graphite iron castings

A well designed runner and feeding system should produce castings with minimal defects and low pour weight. This thesis investigates how the filling regime and solidification of the mould influences defects in the castings produced from that mould. Design guidelines to reduce such defects are proposed and tested. An existing shrinkage fault in a Grey Iron disc brake casting is simulated using a commercial finite-difference computer program. Three criteria are used to predict the defect and the effect of changes to the feeder geometry. Critical Fraction Solidification analysis is used to determine whether the feeder remains in liquid contact with the casting during solidification and this approach is shown to correctly predict the presence or absence of porosity* The feeder block is extended below the ingate of the casting to improve liquid contact between the casting and feeder without significantly increasing the feeder mass. Plant trials confirm the change to the feeder eliminates the porosity defect. The runner system and mould venting for a thin walled Ductile Iron casting are investigated. Trials show that by setting the total mould vent area to be greater than the net ingate area of the castings, the cold-shut frequency is halved. A method for runner system design based on peak linear flow velocity in the runner during mould filling is proposed. A new pressurised runner system produces castings with significantly fewer defects and reduced pour weight when runner areas are designed to maintain peak velocity below 1 m/s. Peak velocity and magnesium levels are demonstrated to be critical factors in the elimination of cold-shut defects. A pressurised runner system is also shown to isolate inclusion defects from castings more effectively than an unpressurised system. From this work, a technique is proposed which allows the yield of an existing runner and feeder system for iron castings to be improved with confidence in the results.

TOF-SIMS analysis using C60. Effect of impact energy on yield and damage

C₆₀ has been shown to give increased sputter yields and, hence, secondary ions when used as a primary particle in SIMS analysis. In addition, for many samples, there is also a reduction in damage accumulation following continued bombardment with the ion beam. In this paper, we report a study of the impact energy (up to 120 keV) of C₆₀ on the secondary ion yield from a number of samples with consideration of any variation in yield response over mass ranges up to m/z 2000. Although increased impact energy is expected to produce a corresponding increase in sputter yield/rate, it is important to investigate any increase in sample damage with increasing energy and, hence, efficiency of the ion beams. On our test samples including a metal, along with organic samples, there is a general increase in secondary ion yield of high-mass species with increasing impact energy. A corresponding reduction in the formation of low-mass fragments is also observed. Depth profiling of organic samples demonstrates that when using C₆₀, there does not appear to be any increase in damage evident in the mass spectra as the impact energy is increased.

Influence of stocking rate and mixed grazing of Angora goats and Merino sheep on animal and pasture production in southern Australia : 2 : liveweight, body condition score, carcass yield and mortality

The effects of animal species (AS; Angora goats, Merino sheep, mixed-grazed goats and sheep at the ratio of 1:1) and stocking rate (SR; 7.5, 10 and 12.5 animals/ha) on the liveweight, body condition score, carcass yield and mortality of goats and sheep were determined in a replicated experiment on improved annual temperate pastures in southern Australia from 1981 to 1984. The pattern of liveweight change was similar for both species with growth from pasture germination in autumn until maturation in late spring followed by weight loss. In winter, sheep grew faster than goats (65 versus 10 g/day, P < 0.05). In mixed-grazed treatments between November and December goats either grew when sheep were losing weight or goats lost less weight than sheep (P < 0.01). Both AS (P < 0.001) and SR (P < 0.001) affected liveweight of sheep and an AS SR interaction (P <  0.05) affected liveweight of goats. Mixed-grazed sheep were heavier than separately grazed sheep at all SR with a mean difference at 10 and 12.5/ha of 4.6 kg. Mixed-grazed goats at 10/ha were heavier than separately grazed goats from the end of the second year of the experiment, but at 12.5/ha, separately grazed goats maintained an advantage over mixed-grazed goats, with a 9.4-kg mean difference in December (P < 0.05). Body condition scores of goats and sheep declined with increasing SR; they were highest in late spring and were highly correlated with liveweight (r² > 0.8). Both AS and SR affected (P < 0.001) carcass weight and GR tissue depth as a direct result of differences in liveweight. Adjusting for differences in carcass weight negated AS effects on GR tissue depth. The carcass weights of sheep and goats increased by similar amounts for each 1-kg increase in liveweight. Mortality of sheep (3.1% p.a.) was unaffected by AS or SR. An AS SR interaction indicated mortality of separately grazed goats at 12.5/ha and mixed-grazed goats at 10 and 12.5/ha were higher (P < 0.05) than all other goat (29 versus 9%) and sheep treatments, primarily because of increased susceptibility to cold stress. Disease prevalence differed between sheep and goats. Mixed grazing of Merino sheep and Angora goats produced complementary and competitive effects depending upon the SR. Goats used summer pasture better but winter pasture less well for liveweight gain than sheep. Angora goats should not be grazed alone or mixed grazed with sheep on annual temperate pastures at SR greater than that recommended for Merino sheep and the evidence indicates a lower SR will reduce risks associated with mortality.

Analysis of environmental sounds as indexical signs in film

In this paper, we investigate the problem of classifying a subset of environmental sounds in movie audio tracks that indicate specific indexical semiotic use. These environmental sounds are used to signify and enhance events occurring in film scenes. We propose a classification system for detecting the presence of violence and car chase scenes in film by classifying ten various environmental sounds that form the constituent audio events of these scenes using a number of old and new audio features. Experiments with our classification system on pure test sounds resulted in a correct event classification rate of 88.9%. We also present the results of the classifier on the mixed audio tracks of several scenes taken from The Mummy and Lethal Weapon 2. The classification of sound events is the first step towards determining the presence of the complex sound scenes within film audio and describing the thematic content of the scenes.

Detecting indexical signs in film audio for scene interpretation

In this paper, we study the sound tracks in films and their indexical semiotic usage by developing a classification system that detects complex sound scenes and their constituent sound events in cinema. We investigate two main issues in this paper: Determination of what constitutes the presence of a high level sound scene and inferences about the thematic content of the scene that can be drawn from this presence, and classification of environmental sounds in the audio track of the scene, to assist in the automatic detection of the high level scene. Experiments with our classification system on pure sounds resulted in a correct event classification rate of 88.9%. When the audio content of a number of film scenes was examined, though a lower accuracy resulted with sound event detection due to the presence of mixed sounds, the film audio samples were generally classified with the correct high-level sound scene label, enabling correct inferences about the story content of the scenes.

Comparisons between two empirical yield criteria for rock masses

In general, rock masses are inhomogeneous, discontinuous media composed of rock material and naturally occurring discontinuities such as joints, fractures and bedding planes. Because of these features, the strength of rock masses is notoriously difficult to assess. Nonetheless, many criteria have been proposed for estimating rock mass strength. Based on the finite element upper and lower bound limit analysis methods, this study examined two empirical yield criteria for rock masses, the Hoek-Brown failure criterion (2002) and the Douglas criterion (2002). The comparisons showed that very different results may be obtained using the same input parameters. Therefore, it is interesting to discuss the source of these differences.