Month: April 2024

Hilbert Domain Analysis of Wavelet Packets for Emotional Speech Classification

Latest Results

Abstract

This work investigates the significance of Hilbert domain characterization of wavelet packets in classifying different emotion of speech signal. The goal of this paper is to create a new emotional speech database and introduce a new feature extraction approach that can recognize various emotions. The proposed feature, wavelet cepstral coefficients (WCC) are based on Hilbert spectrum analysis of the wavelet packet of the speech signal. The speaker-independent machine learning models are developed using multiclass support vector machine (SVM) and k-nearest neighbourhood (KNN) classifier. The approach is tested with newly developed Telugu Indian database and the EMOVO (Italian emotional speech) database. Our proposed wavelet features achieve a peak accuracy of 73.5%, further boosted by NCA feature selection by 3–5%, resulting in an improved unweighted average recall (UAR) of 78% for database 1 and 87.50% for database 2, employing optimal wavelet features in conjunction with SVM classification. The proposed features outperformed the baseline Mel-frequency cepstral coefficients (MFCC) feature. The performance of newly formulated features is better than other existing methodologies tested with different language databases.

Comparative evaluation of performances of algae indices, pixel- and object-based machine learning algorithms in mapping floating algal blooms using Sentinel-2 imagery

Latest Results

Abstract

One of the main threats to freshwater resources is pollution from anthropogenic activities such as rapid urbanization and excessive agricultural nutrient runoff. Remote sensing technologies have been effectively used in monitoring and mapping rapid changes in the marine environment and assessing the overall health of freshwater ecosystems. The main goal of this study is to comparatively evaluate the performance of index-based and classification-based approaches in mapping dense floating algal blooms observed in Lake Burdur using Sentinel-2 imagery. For index-based mapping, algae-specific indices, namely the Floating Algae Index (FAI), Adjusted Floating Algae Index, Surface Algal Blooms Index (SABI), and Algal Blooms Detection Index (ABDI), were used. At the same time, pixel- and object-based Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Long Short-Term Memory Network (LSTM) were utilized for classification-based algal mapping. For this purpose, seven Sentinel-2 images, selected through time series analysis performed on the Google Earth Engine platform, were used as the primary dataset in the application. The results show that high-density floating algae formations can be detected over 99% by both indices and classification-based approaches, whereas pixel-based classification is more successful in mapping low-density algal blooms. When two-class thematic maps representing water and floating algae classes were considered, the maps produced by index-based FAI using an appropriate threshold value and the classification-based RF algorithm reached an overall accuracy of over 99%. The highest algae density in the lake was observed on July 13, 2021, and was determined to be effective in ~ 45 km2 of the lake’s surface.

Indigenous languages & education: Do we have the right agenda?

Latest Results

Abstract

The language and cultural priorities in Australian Indigenous education have been priority areas since the inaugural national Indigenous education policy was launched in 1989. For over thirty years, these priorities have sat awkwardly in the largely non-Indigenous teaching profession and classroom teachers continue to struggle with how to embed these priorities into the education process, despite the efforts of Australian Curriculum, Assessment and Reporting Authority and their elaborations for application at curriculum and practice levels. In this article, I suggest that these language and cultural priorities are at cross-purposes with education priorities, and neither have helped to curb the demise of our Indigenous languages nor improved Indigenous students’ educational outcomes.

Green building practices to integrate renewable energy in the construction sector: a review

Latest Results

Abstract

The building sector is significantly contributing to climate change, pollution, and energy crises, thus requiring a rapid shift to more sustainable construction practices. Here, we review the emerging practices of integrating renewable energies in the construction sector, with a focus on energy types, policies, innovations, and perspectives. The energy sources include solar, wind, geothermal, and biomass fuels. Case studies in Seattle, USA, and Manama, Bahrain, are presented. Perspectives comprise self-sufficiency, microgrids, carbon neutrality, intelligent buildings, cost reduction, energy storage, policy support, and market recognition. Incorporating wind energy into buildings can fulfill about 15% of a building's energy requirements, while solar energy integration can elevate the renewable contribution to 83%. Financial incentives, such as a 30% subsidy for the adoption of renewable technologies, augment the appeal of these innovations.

Green building practices to integrate renewable energy in the construction sector: a review

Latest Results

Abstract

The building sector is significantly contributing to climate change, pollution, and energy crises, thus requiring a rapid shift to more sustainable construction practices. Here, we review the emerging practices of integrating renewable energies in the construction sector, with a focus on energy types, policies, innovations, and perspectives. The energy sources include solar, wind, geothermal, and biomass fuels. Case studies in Seattle, USA, and Manama, Bahrain, are presented. Perspectives comprise self-sufficiency, microgrids, carbon neutrality, intelligent buildings, cost reduction, energy storage, policy support, and market recognition. Incorporating wind energy into buildings can fulfill about 15% of a building's energy requirements, while solar energy integration can elevate the renewable contribution to 83%. Financial incentives, such as a 30% subsidy for the adoption of renewable technologies, augment the appeal of these innovations.

Ultrapure thin films of CdPbS and PbS and photodetectors based on them were obtained and studied for the first time

Latest Results

Abstract

The ultrapure and stoichiometric semiconductor PbS-st thin film is manufactured for the first time by hydrochemical method. The width of the band gap Eg (295 K) = 0.39 eV is in a 0.3–0.7μm  thick PbS-st film. The drift mobility of carriers of 320–480 cm2/(V s) and the carrier lifetime of 7–20 µs correspond to the best natural single crystals of PbS. When cooled to 140 K, the conductivity of PbS-st films decreases 10,000 times. This is already a dielectric. From the conductivity graph in the range of 140–295 K, the Hall band gap of Eg (4 K) = 0.34 eV is determined, which sharply differs from the known Eg (4 K) = 0.286 eV. The temperature coefficient of change is in the width of the optical zone dE/dT (140–323 K) =  + 1.6 × 10–4 eV/K, which is 2.5 times less than the known one. Ultra-fast photodetectors with a time constant of τ = 8 µs and a detection ability of D*(2.4, 1000, 1) = 1.7 × 1010 cm Hz1/2/W were made. As well as photodetectors are similar to vacuum ones with τ = 40–70 µs and D*(2.4, 1000, 1,) = (4–6) × 1010 cm Hz1/2/W with a dark resistance of 40–120 kOhm/square. Oxygen-free photodetectors PbS have no analogues.

Downscaling and reconstruction of high-resolution gridded rainfall data over India using deep learning-based generative adversarial network

Latest Results

Abstract

To expedite regional-scale climate change impact research and assessments, the downscaling of climate data is a crucial prerequisite. Image super-resolution, which is analogous to gridded data downscaling, is the concept of improving the pixel quality of images using deep learning techniques. In this study, the performance of a Super-Resolution Generative Adversarial Network (SRGAN), a cutting-edge deep learning-based image super-resolution technique, is assessed in producing perceptually realistic high-resolution rainfall data over India from the low-resolution input. The main component of SRGAN is a generator network that takes abstract information from low-resolution (LR) rainfall data to infer potential high-resolution (HR) counterparts. A Super-Resolution Residual Neural Network (SRResNet) is used as the generator network. It is trained using a supervised learning strategy (SRResNet) and adversarial learning strategy (several variants created, e.g., SRGAN-MSE, SRGAN-VGGB2, SRGAN-VGGB3 and SRGAN-VGGB4). A statistical downscaling method called bias correction and spatial disaggregation (BCSD) is also employed to compare with the deep learning-based downscaling methods. All these methods are rigorously assessed for their ability to reconstruct distribution, mean, and extreme rainfall during the test period. Our results show that the supervised learning-based SRResNet and adversarial learning-based SRGAN-MSE variant has an upper hand over the BCSD method for gridded rainfall downscaling. These findings have important implications for enhancing the precision and quality of regional climate data in the context of climate change impact assessment.

Climate change projections in Guatemala: temperature and precipitation changes according to CMIP6 models

Latest Results

Abstract

Projected changes in precipitation and temperature for Guatemala were examined using the phase 6 dataset of the Coupled Model Intercomparison Project (CMIP6). CMIP6 models project alterations in annual mean temperature and precipitation in Guatemala relative to the current climate. A set of 25 CMIP6 models project a continuous increase in annual mean temperature over Guatemala during the twenty-first century under four future scenarios. The data provided by WorldClim has a spatial resolution of 2.5 min (of a longitude/latitude degree) this means a 4.5 km × 4.5 km of area of each pixel approximately. for the climate horizons of 2021–2040, 2041–2060, 2061–2080, and 2081–2100, these were adjusted based on the average of 38 local stations in Guatemala from the period (1970–2000). The projected temperature shows a large increase over 5 °C under the SSP5-8.5 scenario, over the northern parts of Guatemala and the northwest. By the end of the twenty-first century, the annual mean temperature in Guatemala is projected to increase by on average 1.8 °C, 2.9 °C, 4.3 °C, and 5.4 °C under the SSP1_2.6, SSP2_4.5, SSP3_7.0, and SSP5_8.5 scenarios, respectively, relative to current climate (1990–2020). The warming is differentiated on a monthly time scale, with CMIP6 models projecting greater warming in July, August, and September, part of the summer and autumn season. Annual precipitation is projected to decrease in Guatemala during the twenty-first century under all scenarios. The rate of change in projected mean annual precipitation varies considerably among scenarios; − 5%, − 9%, − 18%, and − 22% under the SSP1_2.6, SSP2_4.5, SSP3_7.0, and SSP5_8.5 scenarios, respectively. Monthly precipitation projections show great variability, with projected precipitation for the months of May, June, and July, part of the spring and summer, showing a greater decrease than other months and specifically in the northern part of the country. On the other hand, mid-summer precipitation (July and August) shows a decrease in the central and eastern part of the country. The results presented in this study provide baseline information on CMIP6 models for Guatemala, which serve as a basis for developing climate change adaptation and mitigation strategies.

Knitting for conservation: a social practice perspective on a social and behaviour change communication intervention

Latest Results

Abstract

We critically reflect on a conservation project in the Ecuadorian Amazon that was designed to promote biodiversity conservation among lowland indigenous communities involved in eco-tourism initiatives by teaching them how to knit a particular set of local animals. We use interpretive qualitative research and draw on social practice theory to examine the ways that participants’ engagement with new knitting in participatory knitting workshops changed the understanding of environmental conservation and social entrepreneurship within an eco-tourism context. Eventually, the intervention pushed participants to adopt new and difficult-to-sustain conservation and entrepreneurial practices. The introduction of these new practices and a focus on a specific list of local species turned animals into commodities and created unsustainable connections with new materials and a disconnect between local and traditional know-how.

Induction of excitatory brain state governs plastic functional changes in visual cortical topology

Latest Results

Abstract

Adult visual plasticity underlying local remodeling of the cortical circuitry in vivo appears to be associated with a spatiotemporal pattern of strongly increased spontaneous and evoked activity of populations of cells. Here we review and discuss pioneering work by us and others about principles of plasticity in the adult visual cortex, starting with our study which showed that a confined lesion in the cat retina causes increased excitability in the affected region in the primary visual cortex accompanied by fine-tuned restructuring of neuronal function. The underlying remodeling processes was further visualized with voltage-sensitive dye (VSD) imaging that allowed a direct tracking of retinal lesion-induced reorganization across horizontal cortical circuitries. Nowadays, application of noninvasive stimulation methods pursues the idea further of increased cortical excitability along with decreased inhibition as key factors for the induction of adult cortical plasticity. We used high-frequency transcranial magnetic stimulation (TMS), for the first time in combination with VSD optical imaging, and provided evidence that TMS-amplified excitability across large pools of neurons forms the basis for noninvasively targeting reorganization of orientation maps in the visual cortex. Our review has been compiled on the basis of these four own studies, which we discuss in the context of historical developments in the field of visual cortical plasticity and the current state of the literature. Overall, we suggest markers of LTP-like cortical changes at mesoscopic population level as a main driving force for the induction of visual plasticity in the adult. Elevations in excitability that predispose towards cortical plasticity are most likely a common property of all cortical modalities. Thus, interventions that increase cortical excitability are a promising starting point to drive perceptual and potentially motor learning in therapeutic applications.