Object tracking using SIFT features and mean shift

A scale invariant feature transform (SIFT) based mean shift algorithm is presented for object tracking in real scenarios. SIFT features are used to correspond the region of interests across frames. Meanwhile, mean shift is applied to conduct similarity search via color histograms. The probability distributions from these two measurements are evaluated in an expectation–maximization scheme so as to achieve maximum likelihood estimation of similar regions. This mutual support mechanism can lead to consistent tracking performance if one of the two measurements becomes unstable. Experimental work demonstrates that the proposed mean shift/SIFT strategy improves the tracking performance of the classical mean shift and SIFT tracking algorithms in complicated real scenarios.

Novel loading tests on full-scale tapered member portal frames

This paper presents the experimental results of loading tests on two 18m span tapered member portal frames designed to BS 5950. Deflection test results for vertical, lateral and combined loading cases are compared with the predictions given by elastic analysis to BS 5950 and shown to be favourable. The predicted ultimate capacities and modes of failure, which were by lateral-torsional buckling of the columns, were also found to agree with the experimental behaviour. It was found that the method of modelling the tapered members as a series of prismatic elements gave good comparison with test results.

Hylambatin and (Thr)11-hylambatin from the skin secretion of the African hyperoliid frog, Kassina maculata: Structural characterisation, precursor cDNA cloning and preliminary pharmacological testing

The tachykinins hylambatin and (Thr)11-hylambatin have been isolated from the defensive skin secretion of the African hyperoliid frog, Kassina maculata,. Hylambatin (DPPDPNRFYGMMamide) is revised in structure from the original sequence by a single site substitution (Asn/Asp at position 6), and (Thr)11-hylambatin, a novel tachykinin, differs in structure from hylambatin by a single Thr/Met substitution. (Thr)11-hylambatin is five- to ten-fold more abundant than hylambatin in secretions. Synthetic replicates of both peptides were active in smooth muscle preparations including the rat tail artery, rat ileum and bovine trachea. While hylambatin displayed activity consistent with an NK1-receptor ligand, (Thr)11-hylambatin was more active than either substance P or neurokinin A in both NK1- and NK-2 receptor rich preparations. Incorporation of a threoninyl residue rather than the canonical leucyl residue at the penultimate position in both substance P and neurokinin A, generated active ligands in both arterial and intestinal smooth muscle preparations. Hylambatin precursor cDNAs, designated HYBN-1 and HYBN-2, respectively, were cloned from a skin library by 3'- and 5'-RACE reactions. Both were highly-homologous containing open-reading frames of 66 amino acids encoding single copies of either hylambatin or (Thr)11-hylambatin. These data reveal a hitherto unrecognized structure/activity attribute of mammalian tachykinin receptors revealed though discovery of a novel amphibian skin-derived, site-substituted peptide ligand.

Spatial Perception and Cognition in Multichannel Audio for Electroacoustic Music

Listeners experience electroacoustic music as full of significance and meaning, and they experience spatiality as one of the factors contributing to its meaningfulness. If we want to understand spatiality in electroacoustic music, we must understand how the listener’s mental processes give rise to the experience of meaning. In electroacoustic music as in everyday life, these mental processes unite the peripheral auditory system with human spatial cognition. In the discussion that follows we consider a range of the listener’s mental processes relating space and meaning from the perceptual attributes of spatial imagery to the spatial reference frames for places and navigation. When considering multichannel loudspeaker systems in particular, an important part of the discussion is focused on the distinctive and idiomatic ways in which this particular mode of sound production contributes to and situates meaning. These idiosyncrasies include the phenomenon of image dispersion, the important consequences of the precedence effect and the influence of source characteristics on spatial imagery. These are discussed in close relation to the practicalities of artistic practice and to the potential for artistic meaning experienced by the listener.

On-line Emotion Recognition in a 3-D Activation-Valence-Time Continuum using Acoustic and Linguistic Cues

For many applications of emotion recognition, such as virtual agents, the system must select responses while the user is speaking. This requires reliable on-line recognition of the user’s affect. However most emotion recognition systems are based on turnwise processing. We present a novel approach to on-line emotion recognition from speech using Long Short-Term Memory Recurrent Neural Networks. Emotion is recognised frame-wise in a two-dimensional valence-activation continuum. In contrast to current state-of-the-art approaches, recognition is performed on low-level signal frames, similar to those used for speech recognition. No statistical functionals are applied to low-level feature contours. Framing at a higher level is therefore unnecessary and regression outputs can be produced in real-time for every low-level input frame. We also investigate the benefits of including linguistic features on the signal frame level obtained by a keyword spotter.

Kassorins: Novel innate immune system peptides from skin secretions of the African hyperoliid frogs, Kassina maculata and Kassina senegalensis

From defensive skin secretions acquired from two species of African hyperoliid frogs, Kassina maculata and Kassina senegalensis, we have isolated two structurally related, C-terminally amidated tridecapeptides of novel primary structure that exhibit a broad spectrum of biological activity. In reflection of their structural novelty and species of origin, we named the peptides kassorin M (FLEGLLNTVTGLLamide; 1387.8 Da) and kassorin S (FLGGILNTITGLLamide; 1329.8 Da), respectively. The primary structure and organisation of the biosynthetic precursors of kassorins M and S were deduced from cloned skin secretion-derived cDNA. Both open-reading frames encoded a single copy of kassorin M and S, respectively, located at the C-terminus. Kassorins display limited structural similarities to vespid chemotactic peptides (7/13 residues), temporin A (5/13 residues), the N-terminus of Lv-ranaspumin, a foam nest surfactant protein of the frog, Leptodactylus vastus, and an N-terminal domain of the equine sweat surfactant protein, latherin. Both peptides elicit histamine release from rat peritoneal mast cells. However, while kassorin S was found to possess antibacterial activity against Staphylococcus aureus, kassorin M was devoid of such activity. In contrast, kassorin M was found to contract the smooth muscle of guinea pig urinary bladder (EC50 = 4.66 nM) and kassorin S was devoid of this activity. Kassorins thus represent the prototypes of a novel family of peptides from the amphibian innate immune system as occurring in defensive skin secretions.

Effect of column base strength on steel portal frames in fire

This paper presents the findings of a project part sponsored by an ICE Research and Development grant on portal frames in fire. The research reported here has also lead to a sucessful research grant from the IStructE. The paper describes a non-linear elasto plastic dynamic finite element model that captures the collapse of a portal frame in fire. It demonstrates that current guidance on the column base stiffness and strength, to prevent collapse, may in some cases be unconservative.

Moving Object Scanning Apparatus and Method

Apparatus for scanning a moving object includes a visible waveband sensor oriented to collect a series of images of the object as it passes through a field of view. An image processor uses the series of images to form a composite image. The image processor stores image pixel data for a current image and predecessor image in the series. It uses information in the current image and its predecessor to analyse images and derive likelihood measures indicating probabilities that current image pixels correspond to parts of the object. The image processor estimates motion between the current image and its predecessor from likelihood weighted pixels. It generates the composite image from frames positioned according to respective estimates of object image motion. Image motion may alternatively be detected be a speed sensor such as Doppler radar sensing object motion directly and providing image timing signals

Luminosity and mass limits for the progenitor of the type Ic supernova 2004gt in NGC 4038

We report on our attempts to locate the progenitor of the Type Ic supernova SN 2004gt in NGC 4038. We use high-resolution HST ACS images of SN 2004gt and have compared them with deep pre-explosion HST WFPC2 F336W, F439W, F555W, and F814W images. We identify the SN location on the pre-explosion frames with an accuracy of 5 mas. We show that the progenitor is below the detection thresholds of all the pre-explosion images. These detection limits are used to place luminosity and mass limits on the progenitor. The progenitor of SN 2004gt seems to be restricted to a low-luminosity high-temperature star, either a single WC star with an initial mass of > 40 M-circle dot or a low-mass star in a binary. The pre-explosion data cannot distinguish between the two scenarios.

Mass limits for the progenitor star of supernova 2001du and other Type II-P supernovae

The supernova SN 2001du was discovered in the galaxy NGC 1365 at a distance of 19 +/- 2 Mpc, and is a core-collapse event of Type II-P. Images of this galaxy, of moderate depth, have been taken with the Hubble Space Telescope approximately 6.6 yr before discovery and include the supernova position on the WFPC2 field of view. We have observed the supernova with the WFPC2 to allow accurate differential astrometry of SN 2001du on the pre-explosion frames. As a core-collapse event it is expected that the progenitor was a massive, luminous star. There is a marginal detection (3sigma) of a source close to the supernova position on the pre-discovery V -band frame, but it is not precisely coincident and we do not believe it to be a robust detection of a point source. We conclude that there is no stellar progenitor at the supernova position and derive sensitivity limits of the pre-discovery images that provide an upper mass limit for the progenitor star. We estimate that the progenitor had a mass of less than 15 M-circle dot . We revisit two other nearby Type II-P supernovae that have high-quality pre-explosion images, and refine the upper mass limits for the progenitor stars. Using a new distance determination for SN 1999gi from the expanding photosphere method, we revise the upper mass limit to 12 M-circle dot . We present new HST images of the site of SN 1999em, which validate the use of lower spatial resolution ground-based images in the progenitor studies and use a new Cepheid distance to the galaxy to measure an upper mass limit of 15 M-circle dot for that progenitor. Finally we compile all the direct information available for the progenitors of eight nearby core-collapse supernovae and compare their mass estimates. These are compared with the latest stellar evolutionary models of pre-supernova evolution which have attempted to relate metallicity and mass to the supernovae type. Although this is statistically limited at present, reasonable agreement is already found for the lower-mass events (generally the II-P), but some discrepancies appear at higher masses.

Phylogenetic analysis of porcine circovirus type 2 (PCV2) pre- and post-epizootic postweaning multisystemic wasting syndrome (PMWS)

The porcine circovirus type 2 (PCV2) genome encodes three major open reading frames (ORFs) encoding the replicase proteins (ORF1), the viral capsid protein (ORF2), and a protein with suggested apoptotic activity (ORF3). Previous phylogenetic analyses of complete genome sequences of PCV2 from GenBank have demonstrated 95-100% intra-group nucleotide sequence identity. However, although these isolates were readily grouped into clusters and clades, there was no correlation between the occurrence of specific PCV2 genotypes and the geographic origin or health status of the pig. In the present study, a unique dataset from a field study spanning the years pre and post the recognition of postweaning multisystemic wasting syndrome (PMWS) in Sweden was utilized. Using this dataset it was possible to discriminate three Swedish genogroups (SG1-3) of PCV2, of which SG1 was recovered from a pig on a healthy farm ten years before the first diagnosis of PMWS in Sweden. The SG1 PCV2/ORF2 gene sequence has been demonstrated to exhibit a high genetic stability over time and has subsequently only been demonstrated in samples from pigs on nondiseased farms. In contrast, SG2 was almost exclusively found on farms that had only recently broken down with PMWS whereas the SG3 genogroup predominated in pigs from PMWS-affected farms. These results further support the results obtained from earlier in vitro and in vivo experimental models and suggest the association of specific PCV2 genogroups with diseased and nondiseased pigs in the field.

Design optimization of cold-formed steel portal frames taking into account the effect of building topology

Cold-formed steel portal frames are a popular form of construction for low-rise commercial, light industrial and agricultural buildings with spans of up to 20 m. In this article, a real-coded genetic algorithm is described that is used to minimize the cost of the main frame of such buildings. The key decision variables considered in this proposed algorithm consist of both the spacing and pitch of the frame as continuous variables, as well as the discrete section sizes.A routine taking the structural analysis and frame design for cold-formed steel sections is embedded into a genetic algorithm. The results show that the real-coded genetic algorithm handles effectively the mixture of design variables, with high robustness and consistency in achieving the optimum solution. All wind load combinations according to Australian code are considered in this research. Results for frames with knee braces are also included, for which the optimization achieved even larger savings in cost.

Gaussian Approximation for Tracking Occluding and Interacting Targets

In this paper, we show how interacting and occluding targets can be tackled successfully within a Gaussian approximation. For that purpose, we develop a general expansion of the mean and covariance of the posterior and we consider a first order approximation of it. The proposed method differs from EKF in that neither a non-linear dynamical model nor a non-linear measurement vector to state relation have to be defined, so it works with any kind of interaction potential and likelihood. The approach has been tested on three sequences (10400, 2500, and 400 frames each one). The results show that our approach helps to reduce the number of failures without increasing too much the computation time with respect to methods that do not take into account target interactions.