Series feature aggregation for content-based image retrieval

Feature aggregation is a critical technique in content-based image retrieval systems that employ multiple visual features to characterize image content. One problem in feature aggregation is that image similarity in different feature spaces can not be directly comparable with each other. To address this problem, a new feature aggregation approach, series feature aggregation (SFA), is proposed in this paper. In contrast to merging incomparable feature distances in different feature spaces to get aggregated image similarity in the conventional feature aggregation approach, the series feature aggregation directly deal with images in each feature space to avoid comparing different feature distances. SFA is effectively filtering out irrelevant images using individual features in each stage and the remaining images are images that collectively described by all features. Experiments, conducted with IAPR TC-12 benchmark image collection (ImageCLEF2006) that contains over 20,000 photographic images and defined queries, have shown that SFA can outperform the parallel feature aggregation and linear distance combination schemes. Furthermore, SFA is able to retrieve more relevant images in top ranked outputs that brings better user experience in finding more relevant images quickly.

Properties of series feature aggregation schemes

Feature aggregation is a critical technique in content-based image retrieval (CBIR) that combines multiple feature distances to obtain image dissimilarity. Conventional parallel feature aggregation (PFA) schemes failed to effectively filter out the irrelevant images using individual visual features before ranking images in collection. Series feature aggregation (SFA) is a new scheme that aims to address this problem. This paper investigates three important properties of SFA that are significant for design of systems. They reveal the irrelevance of feature order and the convertibility of SFA and PFA as well as the superior performance of SFA. Furthermore, based on Gaussian kernel density estimator, the authors propose a new method to estimate the visual threshold, which is the key parameter of SFA. Experiments, conducted with IAPR TC-12 benchmark image collection (ImageCLEF2006) that contains over 20,000 photographic images and defined queries, have shown that SFA can outperform conventional PFA schemes.

Properties of series feature aggregation schemes

Feature aggregation is a critical technique in content-based image retrieval (CBIR) that combines multiple feature distances to obtain image dissimilarity. Conventional parallel feature aggregation (PFA) schemes failed to effectively filter out the irrelevant images using individual visual features before ranking images in collection. Series feature aggregation (SFA) is a new scheme that aims to address this problem. This paper investigates three important properties of SFA that are significant for design of systems. They reveal the irrelevance of feature order and the convertibility of SFA and PFA as well as the superior performance of SFA. Furthermore, based on Gaussian kernel density estimator, the authors propose a new method to estimate the visual threshold, which is the key parameter of SFA. Experiments, conducted with IAPR TC-12 benchmark image collection (ImageCLEF2006) that contains over 20,000 photographic images and defined queries, have shown that SFA can outperform conventional PFA schemes.

Integrating time-series hydroacoustics and video observations for detecting changes in benthic habitats

The ability to quantify change in marine benthic habitats must be considered a key goal of marine habitat mapping activities. Changes in distribution of distinct suites of benthic biological species may occur as a result of natural or human induced processes and these processes may operate at a range of temporal and spatial scales. It is important to understand natural small scale inter-annual patterns of change in order to separate these signals from potential patterns of longer term change. Work to describe these processes of change from an acoustic remote sensing stand point has thus far been limited due to the relatively recent availability of full coverage swath acoustic datasets and cost pressures associated with multiple surveys of the same area. This paper describes the use of landscape transition analysis as a means to differentiate seemingly random patterns of habitat change from systematic signals of habitat transition at a shallow (10–50 m depth) 18 km2 study area on the temperate Australian continental shelf between the years 2006 and 2007. Supervised classifications for each year were accomplished using independently collected high resolution (3 m cell-size) multibeam echosounder (MBES) and video-derived reference data. Of the 4 representative biotic classes considered, signals of directional systematic changes were observed to occur between a shallow kelp dominated class, a deep sessile invertebrate dominated class and a mixed class of kelp and sessile invertebrates. These signals of change are interpreted as inter-annual variation in the density and depth related extent of canopy forming kelp species at the site, a phenomenon reported in smaller scale temporal studies of the same species. The methods applied in this study provide a detailed analysis of the various components of the traditional change detection cross tabulation matrix allowing identification of the strongest signals of systematic habitat transitions across broad geographical regions. Identifying clear patterns of habitat change is an important first step in linking these patterns to the processes that drive them.

Corrosion inhibition of 7000 series aluminium alloys with cerium diphenyl phosphate

Cerium diphenyl phosphate (Ce(dpp)3) has previously been shown to be a strong corrosion inhibitor for aluminium-copper magnesium alloy AA2024-T3 and AA7075 in chloride solutions. Surface characterisation including SEM and ToF-SIMS coupled with electrochemical impedance spectroscopy (EIS) measurements are used to propose a mechanism of corrosion inhibition which appears to involve the formation of a complex oxide film of aluminium and cerium also incorporating the organophosphate component. The formation of a thin complex film consisting of hydrolysis products of the Ce(dpp)3 compound and aluminium oxide is proposed to lead to the observed inhibition. SEM analysis shows that some intermetallics favour the creation of thicker deposits predominantly containing cerium oxide compounds.

Effect of noise factors in energy management of series plug-in hybrid electric vehicles

It has been demonstrated that charge depletion (CD) energy management strategies are more efficient choices for energy management of plug-in hybrid electric vehicles (PHEVs). The knowledge of drive cycle as a priori can improve the performance of CD energy management in PHEVs. However, there are many noise factors which affect both drivetrain power demand and vehicle performance even in identical drive cycles. In this research, the effect of each noise factor is investigated by introducing the concept of power cycle instead of drive cycle for a journey. Based on the nature of the noise factors, a practical solution for developing a power-cycle library is introduced. Investigating the predicted power cycle, an energy management strategy is developed which considers the influence of temperature noise factor on engine performance. The effect of different environmental and geographic conditions, driver behavior, aging of battery and other components are considered. Simulation results for a modelled series PHEV similar to GM Volt show that the suggested energy management strategy based on the driver power cycle library improves both vehicle fuel economy and battery health by reducing battery load and temperature.

An in vitro comparative assessment with a series of new triphenyltin(IV) 2-/4-[(E)-2-(aryl)-1-diazenyl]benzoates endowed with anticancer activities: Structural modifications, analysis of efficacy and cytotoxicity involving human tumor cell lines

Four new triphenyltin(IV) complexes of composition Ph₃SnLH (where LH = 2-/4-[(E)-2-(aryl)-1-diazenyl]benzoate) (1–4) were synthesized and characterized by spectroscopic (¹H, ¹³C and ¹¹⁹Sn NMR, IR, ¹¹⁹Sn Mössbauer) techniques in combination with elemental analysis. The ¹¹⁹Sn NMR spectroscopic data indicate a tetrahedral coordination geometry in non-coordinating solvents. The crystal structures of three complexes, Ph₃SnL¹H (1), Ph₃SnL³H (3), Ph₃SnL⁴H (4), were determined. All display an essentially tetrahedral geometry with angles ranging from 93.50(8) to 124.5(2)°; ¹¹⁹Sn Mössbauer spectral data support this assignment. The cytotoxicity studies were performed with complexes 1–4, along with a previously reported complex (5) in vitro across a panel of human tumor cell lines viz., A498, EVSA-T, H226, IGROV, M19 MEL, MCF-7 and WIDR. The screening results were compared with the results from other related triphenyltin(IV) complexes (6–7) and tributyltin(IV) complexes (8–11) having 2-/4-[(E)-2-(aryl)-1-diazenyl]benzoates framework. In general, the complexes exhibit stronger cytotoxic activity. The results obtained for 1–3 are also comparable to those of its o-analogs i.e. 4–7, except 5, but the advantage is the former set of complexes demonstrated two folds more cytotoxic activity for the cell line MCF-7 with ID₅₀ values in the range 41–53 ng/ml. Undoubtedly, the cytotoxic results of complexes 1–3 are far superior to CDDP, 5-FU and ETO, and related tributyltin(IV) complexes 8–11. The quantitative structure-activity relationship (QSAR) studies for the cytotoxicity of triphenyltin(IV) complexes 1–7 and tributyltin(IV) complexes 8–11 is also discussed against a panel of human tumor cell lines.

New series of intramolecularly coordinated diaryltellurium compounds. Rational synthesis of the diarylhydroxytelluronium triflate [(8-Me 2NC 10H 6) 2Te(OH)](O 3SCF 3)

The reaction of 8-dimethylaminonaphthyllithium etherate with the tellurium(II) bis(dithiocarbamate) Te(S₂CNEt₂)₂ provided the diaryltelluride (8-Me₂NC₁₀H₆)₂Te (1). The oxidation of 1 with an excess of H₂O₂ did not afford the expected diaryltellurium(IV) oxide (8-Me₂NC₁₀H₆)₂TeO (2), but the diaryltellurium(VI) dioxide (8-Me₂NC₁₀H₆)₂TeO₂ (3). The preparation of 2 was achieved by the comproportionation reaction of 1 and 3. The protonation of 2 using triflic acid gave rise to the formation of diarylhydroxytelluronium triflate [(8-Me₂NC₁₀H₆)₂Te(OH)](O₃SCF₃) (4), which features the protonated diaryltellurium oxide [(8-Me₂NC₁₀H₆)₂Te(OH)]+ (4a). Compounds 1, 3·H₂O·H₂O₂, 3·2H₂O, and 4 were characterized by X-ray crystallography. The experimentally obtained molecular structures were compared to those calculated for 1–3, 4a, and (8-Me₂NC₁₀H₆)₂Te(OH)₂ (5) as well as the related diphenyltellurium compounds Ph₂Te (6), Ph₂TeO (7), Ph₂TeO₂ (8), [Ph₂Te(OH)]+ (9a), and Ph₂Te(OH)₂ (10) at the DFT/B3PW91 level of theory.

The application of an ensemble of boosted elman networks to time series prediction : a benchmark study

In this paper, the application of multiple Elman neural networks to time series data regression problems is studied. An ensemble of Elman networks is formed by boosting to enhance the performance of the individual networks. A modified version of the AdaBoost algorithm is employed to integrate the predictions from multiple networks. Two benchmark time series data sets, i.e., the Sunspot and Box-Jenkins gas furnace problems, are used to assess the effectiveness of the proposed system. The simulation results reveal that an ensemble of boosted Elman networks can achieve a higher degree of generalization as well as performance than that of the individual networks. The results are compared with those from other learning systems, and implications of the performance are discussed.