85000 research outputs found
Sort by
Chroma Intra Prediction With Lightweight Attention-Based Neural Networks
No full textNeural networks can be successfully used for cross-component prediction in video coding. In particular, attention-based architectures are suitable for chroma intra prediction using luma information because of their capability to model relations between difierent channels. However, the complexity of such methods is still very high and should be further reduced, especially for decoding. In this paper, a cost-effective attention-based neural network is designed for chroma intra prediction. Moreover, with the goal of further improving coding performance, a novel approach is introduced to utilize more boundary information effectively. In addition to improving prediction, a simplification methodology is also proposed to reduce inference complexity by simplifying convolutions. The proposed schemes are integrated into H.266/Versatile Video Coding (VVC) pipeline, and only one additional binary block-level syntax flag is introduced to indicate whether a given block makes use of the proposed method. Experimental results demonstrate that the proposed scheme achieves up to -0.46%/-2.29%/-2.17% BD-rate reduction on Y/Cb/Cr components, respectively, compared with H.266/VVC anchor. Reductions in the encoding and decoding complexity of up to 22% and 61%, respectively, are achieved by the proposed scheme with respect to the previous attention-based chroma intra prediction method while maintaining coding performance
Machine Learning-Based Reliability Analysis of Structural Concrete Cracking Considering Realistic Nonuniform Corrosion Development
No full textCorrosion-induced concrete cracking significantly weakens the integrity, serviceability and durability of reinforced concrete (RC) structures. Existing reliability analysis of corrosion-induced concrete cracking often considers the corrosion of reinforcement as a uniform process, in favor of implementing the analytical formulation of corrosion rust progression. However, corrosion distribution in RC structures is seldom uniform around the steel reinforcement, hence the corrosion-induced pressure. Thus, considering the nonuniform corrosion process in the reliability analysis becomes important. This paper develops a time-dependent reliability methodology, combining mesoscale heterogeneous fracture modeling and a state-of-the-art machine learning algorithm, to assess the serviceability of the RC structures subjected to nonuniform development of corrosion. The effects of critical crack width, corrosion nonuniformity, chloride content, temperature, and relative humidity on the failure probability are investigated. The worked example demonstrates the importance of considering the nonuniformity of the corrosion product distribution, which provides reliable evaluation of the remaining safe life of RC structures compared with the use of a uniform corrosion model. The developed unified assessing methodology for corrosion of RC structures can serve as a useful tool for engineers, designers, and asset managers for their decision making with regard to repair and maintenance of corrosion-affected RC structures
Thermal and acoustic performance in textile fibre-reinforced concrete: An analytical review
No full textTextile fibre-reinforced concrete based reviews have explored various engineering properties, such as strengthening of concrete, enhancing strain capacity, crack control, durability, and energy absorption. An essential missing component is a comprehensive analysis of the thermal and acoustic insulation performance of textile fibre-reinforced concrete. The paper provides a large-scale analytical database by analysing prior literature on the thermal and acoustic performance of textile fibre-reinforced concrete. It further reviews the microstructural and pore-structural aspects of concrete to provide an overview of the underlying mechanisms driving these properties. This review explores the impact of textile fibre inclusion from 0–20 as a mass percentage (wt%) and 0–40 as a volume percentage (v%). The key findings of the review are that jute fibre-reinforced mortar demonstrated superior thermal conductivity, achieving 0.068 W/mK at 20 wt% inclusion, followed by 0.08 W/mK of basalt fibres at 20 v% inclusion, demonstrating that fibres possess commendable insulation qualities. Notably, inclusion of 30 v% of 2–4 mm miscanthus fibre in concrete showed outstanding dual performance, achieving optimal thermal conductivity of 0.09 W/mK and 90% acoustic absorption at 841 Hz. Finally, the study suggests directions to address identified gaps that can be utilised in the design of future research focusing end-user applications
4D printed pH-responsive labels of methacrylic anhydride grafted konjac glucomannan for detecting quality changes in respiring climacteric fruits
No full textTo enhance the monitoring of quality changes in respiring climacteric fruits during storage, pH-responsive labels were developed using methacrylic anhydride (MA)-grafted konjac glucomannan (KGM) through a combination of 4D printing and light-curing technology. Anthocyanin was used as the color indicator. The response of these labels, prepared through 4D printing and molding, to changes in the quality of kiwi fruits was determined and compared. MA was successfully grafted onto KGM, achieving the highest grafting degree of 15%. The concentration of the maximally MA-grafted KGM (KGM-MA-HDG) had a significant effect on printability and color response of the printed labels. A 3% (w/w) concentration of KGM-MA-HDG was chosen to produce 4D printed pH-responsive labels. Compared to casted and 4D printed labels with higher infill, the printed labels with a lower infill rate better responded to the different quality levels of kiwi fruits in terms of color change, which was attributed to the stronger ability to absorb water and the lower masking degree of discolored areas from internal anthocyanin migration
Revolutionizing construction and demolition waste sorting: Insights from artificial intelligence and robotic applications
No full textThe growing environmental concerns have emerged the necessity of sustainable waste management of construction and demolition (C&D) wastes. This review explores the advancements in artificial intelligence (AI) and robotics to automate C&D waste sorting. A comprehensive examination of this domain is conducted by structuring the paper around six research questions. Current trends and potential future directions are revealed by performing methodology and data analysis involving bibliometric and scientometric studies. Notably, recent research emphasises circular economy, AI, and robotics, underscoring the importance to enhance AI for precise categorisation. The scarcity of publicly available datasets is a central challenge in the C&D waste domain, that hinders effective AI applications. However, data augmentation, data synthesis, generative AI, and transfer learning have been identified as crucial techniques to enhance dataset quality and categorization accuracy. While AI draws significant attention in the C&D waste domain, this review shows a lack of AI-enabled robotics systems due to the complex nature of waste sorting and collection. In summary, this study's findings highlight the need for new methods and techniques integrating multisensory fusion, unsupervised machine learning and robotics intelligence to continuously learn and adapt to new waste streams and materials, making them highly efficient in sustainable waste management
Effect of calcium-sequestering salts and heat treatment on the rheological and textural properties of acid gels from blends of skimmed buffalo and bovine milk
No full textThe influence of adding 5 mM trisodium citrate (TSC) or disodium hydrogen phosphate (DSHP) and heat treatment (85 °C or 95 °C for 5 min) on the acid gelation properties of blends of skim buffalo and bovine milk (0:100, 25:75, 50:50, 75:25, 100:0) was investigated. Significant increases in gelation pH, final G′ values, firmness, and water-holding capacity of gels were observed with increasing proportion of buffalo skim milk and with higher heating temperature. Differences in gel firmness were linked to gel microstructure, where milk blends containing higher proportion of buffalo skim milk formed gels with denser protein network clusters. The addition of TSC or DSHP reduced the gelation pH, final G′ values and gel firmness, but increased gel water-holding capacity. These results provide a better understanding of acid gelation of buffalo and bovine milk blends which will subsequently promote the potential of using milk mixtures in modulating the gel texture
Multivariate solar power time series forecasting using multilevel data fusion and deep neural networks
No full textAccurate forecasting of regional solar photovoltaic power (SPVP) generation is essential for efficient energy management and planning. Existing approaches have shown the effectiveness of decomposing the time series to model the stochastic variability in SPVP data. However, these approaches have limitations in extracting and exploiting both spatial and temporal information from complex and high-dimensional data from multiple sources with intricate relationships, which can impact the accuracy of predictions. In this paper, we propose a novel approach called multilevel data fusion and neural basis expansion analysis (MF-NBEA) for forecasting aggregated regional-level SPVP generation. MF-NBEA integrates exogenous data at multiple levels, uses supervised and unsupervised encoders to provide compact data representation, and enhances model learning from complex data by incorporating spatial information. It also includes a sequence analyser module based on a neural network decomposition mechanism to learn the variability in data and incorporates a residuals learner module to improve overall predictions. We evaluate MF-NBEA using two real-world datasets and find that it outperforms state-of-the-art deep learning methods in terms of forecast accuracy. Furthermore, MF-NBEA facilitates information fusion and knowledge extraction to provide interpretable predictions regarding trend, seasonality, and residual components. The insights gained from our approach inform decision-making for energy management and planning, and can lead to more efficient and sustainable resource utilisation
Planar-thick panels and 3D-printed gap fillers: A hybrid digital fabrication approach to curved surface approximations
No full textCurved surfaces may be approximated using polygonal panels to simplify complex architectural designs, yet their constructions can be expensive due to single-use moulds producing unique panels. To reduce costs, strategies like planarizing and reducing the number of different panels have been suggested, which are increasingly achievable with advancing digital fabrication technologies. However, current subtractive and additive manufacturing techniques face challenges in producing complex 3D shapes and large volumes, respectively. Combining these two techniques has been attempted, mainly with small 3D-printed nodes, leaving the direct realization of curved surface approximations unexplored. This paper extends the thick-panel origami theory to introduce planar-thick panels and 3D-printed gap fillers for cost-effective constructions. It presents a computational workflow to transform meshes into non-intersecting panels while preserving edge connectivity and frame-like gap fillers with minimized volumes. Physical prototypes demonstrate significant cost savings compared to direct 3D printing. All prototypes are accurate within a 10 mm material thickness due to the guidance of the gap fillers. The finite element analysis shows that structural performance is greatly influenced by the material properties, especially of panels, due to their relatively larger volume. Numerical examples are also presented using the k-means clustering-optimization method to reduce the number of different panels
A comparative study of programs to predict direct photolysis rates in wastewater systems
No full textA wide range of contaminants of emerging concern (CECs) are known to photodegrade in the surface layers of natural waters and wastewater systems. Computer programs such as GCSolar, ABIWAS, APEX, EXAMS and WASP model the direct photolysis rates and half-lives of CECs, usually as a function of the solar irradiance, water molar light extinction, chemical molar light absorption and reaction quantum yield. These programs have been used extensively for studies in natural water systems in the northern hemisphere. However, their applicability to wastewater treatment systems such as waste stabilisation ponds and/or southern hemisphere conditions is not well studied. Here we present a comparative review of the major software used and their potential applicability to predicting direct photolysis rates and half-lives in wastewater. The newer equivalent monochromatic wavelength, approach, which enables the approximation of polychromatic photodegradation via a monochromatic wavelength is also discussed. Current software appears to be less suitable for modelling photodegradation in wastewater systems in the southern hemisphere than the northern hemisphere as their internal databases are based on data from natural waters in the northern hemisphere. This may be because there have been few attempts to model CEC photolysis in wastewater systems, particularly in the southern hemisphere. This indicates that either new software needs to be developed, or these programs need to be updated with data on wastewater matrices and/or the southern hemisphere. We anticipate this review will promote the adaptation of these programs as tools to further the understanding CEC photodegradation in wastewater treatment plants
Comparative analysis of symptom profile and risk of death associated with infection by SARS-CoV-2 and its variants in Hong Kong
No full textThe recurrent multiwave nature of coronavirus disease 2019 (COVID-19) necessitates updating its symptomatology. We characterize the effect of variants on symptom presentation, identify the symptoms predictive and protective of death, and quantify the effect of vaccination on symptom development. With the COVID-19 cases reported up to August 25, 2022 in Hong Kong, an iterative multitier text-matching algorithm was developed to identify symptoms from free text. Multivariate regression was used to measure associations between variants, symptom development, death, and vaccination status. A least absolute shrinkage and selection operator technique was used to identify a parsimonious set of symptoms jointly associated with death. Overall, 70.9% (54 450/76 762) of cases were symptomatic with 102 symptoms identified. Intrinsically, the wild-type and delta variant caused similar symptoms among unvaccinated symptomatic cases, whereas the wild-type and omicron BA.2 subvariant had heterogeneous patterns, with seven symptoms (fatigue, fever, chest pain, runny nose, sputum production, nausea/vomiting, and sore throat) more frequent in the BA.2 cohort. With ≥2 vaccine doses, BA.2 was more likely than delta to cause fever among symptomatic cases. Fever, blocked nose, pneumonia, and shortness of breath remained jointly predictive of death among unvaccinated symptomatic elderly in the wild-type-to-omicron transition. Number of vaccine doses required for reducing occurrence varied by symptoms. We substantiate that omicron has a different clinical presentation compared to previous variants. Syndromic surveillance can be bettered with reduced reliance on symptom-based case identification, increased weighing on symptoms predictive of death in outcome prediction, individual-based risk assessment in care homes, and incorporating free-text symptom reporting