Search

Abstract Forecasting discharge (Q) and water level (H) are essential factors in hydrological research and flood prediction. In recent years, deep learning has emerged as a viable technique for capturing the non-linear relationship of historical data to generate highly accurate prediction results. Despite the success in various domains, applying deep learning in Q and H prediction is hampered by three critical issues: a shortage of training data, the occurrence of noise in the collected data, and the difficulty in adjusting the model’s hyper-parameters. This work proposes a novel deep learning-based Q–H prediction model that overcomes all the shortcomings encountered by existing approaches. Specifically, to address data scarcity and increase prediction accuracy, we design an ensemble learning architecture that takes advantage of multiple deep learning techniques. Furthermore, we leverage the Singular-Spectrum Analysis (SSA) to remove noise and outliers from the original data. Besides, we exploit the Genetic Algorithm (GA) to propose a novel mechanism that can automatically determine the prediction model’s optimal hyper-parameters. We conducted extensive experiments on two datasets collected from Vietnam’s Red and Dakbla rivers. The results show that our proposed solution outperforms current techniques across a wide range of metrics, including NSE, MSE, MAE, and MAPE. Specifically, by exploiting the ensemble learning technique, we can improve the NSE by at least $$2\%$$ 2 % . Moreover, with the aid of the SSA-based data preprocessing technique, the NSE is further enhanced by more than $$5\%$$ 5 % . Finally, thanks to GA-based optimization, our proposed model increases the NSE by at least $$6\%$$ 6 % and up to $$40\%$$ 40 % in the best case.

The latest advances in mobile platforms, such as robots, have enabled the automatic acquisition of full coverage point cloud data from large areas with terrestrial laser scanning. Despite this progress, the crucial post-processing step of registration, which aligns raw point cloud data from separate local coordinate systems into a unified coordinate system, still relies on manual intervention. To address this practical issue, this study presents an automated point cloud registration approach optimized for a stop-and-go scanning system based on a quadruped walking robot. The proposed approach comprises three main phases: perpendicular constrained wall-plane extraction; coarse registration with plane matching using point-to-point displacement calculation; and fine registration with horizontality constrained iterative closest point (ICP). Experimental results indicate that the proposed method successfully achieved automated registration with an accuracy of 0.044 m and a successful scan rate (SSR) of 100% within a time frame of 424.2 s with 18 sets of scan data acquired from the stop-and-go scanning system in a real-world indoor environment. Furthermore, it surpasses conventional approaches, ensuring reliable registration for point cloud pairs with low overlap in specific indoor environmental conditions.

This study investigated the determinants of online shopping continuance intention of Generation Y and Z during the new normal. A conceptual framework, which was an extension of the Technology Acceptance Model, was empirically tested using partial least squares structural equation modelling, multi-group analysis technique, and the data collected from 847 Gen Y-ers and Gen Z-ers in Hanoi, Vietnam during March 2022. The results revealed that facilitators of repurchase intention included perceived usefulness, perceived ease of use, satisfaction, and environmental awareness while perceived risks of online shopping served as a barrier. Notably, the barrier was found to affect Gen Y’s repurchase intention more severely. Personalization was not directly associated with the intention but had strong indirect effects through perceived usefulness, perceived ease of use, and satisfaction. The risk of COVID-19 was not a predictor of online repurchase intention. Understanding of the continuance intention of online shopping among consumers from different generations in an emerging country during the new normal may aid to enhance the quality of decision-making. Specifically, platforms and sellers should adopt customized marketing programs towards Gen Y and Gen Z. Additionally, a user-friendly and informative purchasing process with personalized features should be formulated. Demonstrating online shopping as a green behavior would be useful. This study differs from earlier research by considering and comparing factors influencing the intention to keep shopping online of Gen Y and Gen Z in a developing country when the COVID-19 is well controlled.

This study was conducted to compare the black and dark brown varieties of Noi purebred chickens raised in an intensive farming system. A total of 600 black and 600 dark brown Noi chickens were observed starting from 1 day after hatching. At 20 weeks of age, 30 black Noi cocks and 300 black Noi hens, as well as 30 dark brown Noi cocks and 300 dark brown Noi hens, were studied until the beginning of reproduction. All roosters and hens were kept in individual cages and mating was accomplished artificially. At 24 weeks of age, the black and dark brown Noi cocks had average weights of 2555 and 2600 g, respectively, and 1796 and 1830 g for hens, respectively. There was no difference in body weight between the two varieties. The first egg-laying ages of both varieties were relatively late. From the age of 25 weeks up to 50 weeks, the egg yields of the black and dark brown Noi hens were 74.33 and 77.98 eggs/hen, respectively, with average egg weights of 48.3 and 49.7 g, respectively. Embryonic egg rates were low at 80.3 and 81.9% for the black and dark brown varieties, respectively, and the rate of chick/incubated eggs was 73.4 and 77.0%, respectively. The Noi chickens, especially the dark brown variety, reached a relatively high egg yield in an intensive farming system, which creates great potential for the exploitation and development of this genetic source.

The concentration of fine particulate matter (PM2.5), which represents inhalable particles with diameters of 2.5 micrometers and smaller, is a vital air quality index. Such particles can penetrate deep into the human lungs and severely affect human health. This paper studies accurate PM2.5 prediction, which can potentially contribute to reducing or avoiding the negative consequences. Our approach’s novelty is to utilize the genetic algorithm (GA) and an encoder-decoder (E-D) model for PM2.5 prediction. The GA benefits feature selection and remove outliers to enhance the prediction accuracy. The encoder-decoder model with long short-term memory (LSTM), which relaxes the restrictions between the input and output of the model, can be used to effectively predict the PM2.5 concentration. We evaluate the proposed model on air quality datasets from Hanoi and Taiwan. The evaluation results show that our model achieves excellent performance. By merely using the E-D model, we can obtain more accurate (up to 53.7%) predictions than those of previous works. Moreover, the GA in our model has the advantage of obtaining the optimal feature combination for predicting the PM2.5 concentration. By combining the GA-based feature selection algorithm and the E-D model, our proposed approach further improves the accuracy by at least 13.7%.

The global positioning system (GPS) has motivated rapid advances in mobility data collection. A massive amount of spatio-temporal information has made it possible to know where a person was and when, but not how and why (s)he travelled, creating the need for inference models. Compared with mode detection, purpose imputation has been insufficiently studied. However, the relative lack of attention to purpose identification does not mean that this field has not emerged. For this paper, which is the first review dedicated to inferring trip purposes from GPS data, 1162 non-duplicate papers from four databases (Scopus, Web of Science, ScienceDirect and TRID) were screened, and a corpus of 25 publications was selected for examination. Based on these papers, the purpose imputation problem is defined in the contexts of the evolution of GPS-based travel surveys and two research domains, transportation science (TS) and human geography (HG). Subsequently, three steps of the purpose detection process, namely trip end detection, input feature selection and main algorithms and validation, are surveyed. During these procedures, the differences between studies in TS and those in HG are highlighted. Finally, unresolved issues related to data and feature selection, algorithms and assessment are discussed substantially to provide potential research directions. This review may be an informative reference for those newly accessing the GPS-based purpose imputation field and/or intending to develop solutions to this problem.

Coir (i.e., coconut fiber) is one of the popular agricultural waste products, especially in Vietnam, mostly discarded when copra and coconut water are taken, causing environmental pollution and waste of potential biomass. Various research has been done to reuse this resource as advanced materials. In this study, the NaOH-urea-H2O2 combination was utilized to make cellulose aerogel from coir fibers for the first time. Cellulose aerogel was synthesized by the sol-gel method combined with the freeze-drying technique. The properties of cellulose aerogel were determined, such as density, porosity, surface morphology analysis by FTIR, SEM, and thermal stability evaluation by TGA analysis. They exhibit low density (0.0099 - 0.0158 g/cm3), high porosity (98.96 - 99.35 vol%), and the methyl blue adsorption experiment shows cellulose aerogel's ability to treat color in water is significant.

Old apartment buildings in Hanoi were built before 1954 and from around 1960 to 1994 with 1600 apartments. Due to the end of construction lifespan, components are sinking, cracking, and tilting, combined with households expanding and repairing themselves, so many buildings are seriously damaged and dangerous, needing to be renovated and rebuilt in the near future to ensure residents’ safety. This work cannot slow down any longer. From 1999 until now, despite great efforts, Hanoi has only renovated and rebuilt 19 apartments (about 1.2%). The main reason is due to conflicts between the interests of the household community and the real estate investors. Contributing to resolving the above contradiction and promoting the speed of redevelopment, this study proposes a community-led model to rebuild old apartment buildings in Hanoi, Vietnam.

Recently, there has been an increasing interest in targeting the safety of bus operations worldwide; however, little is known about the determinants of the bus crash severity in developing countries. By estimating an ordered logit model using the bus-involved collision data in Hanoi (Vietnam), spanning the period from 2015 to 2019, this study investigates various factors associated with the crash severity. The results reveal that the severity risk increases for (1) large buses, (2) raining conditions, (3) evening or night, (4) sparse traffic, (5) non-urban areas, (6) roads with at least three lanes, (7) curved roads, (8) two-way roads without a physical barrier, (9) head-on collision, and (10) pedestrian-related crashes. Aside from confirming the crucial roles of a wide range of factors, this research has examined the effects of two determinants (traffic density and crash area) that have not been considered for the cases of developing countries previously. Based on the findings on the impacts of factors, a series of policy recommendations regarding improving road conditions in non-urban areas, promoting walking infrastructure, reminders of high-risk situations for drivers, safety notes when improving bus service quality, and recording bus-related crashes are proposed.

A highly sensitive magnetic measurement system was successfully developed using a modified commercial giant magnetometer resistance (GMR) sensor. The device was placed in a highly uniform magnetic field that was generated by two Helmholtz coil pairs which emit magnetic fields in perpendicular directions to magnetize the SPMNPs and bias the GMR sensor to a linear operating range. The system was used to quantitatively determine the concentrations of superparamagnetic nanoparticles in continuous flow. The characteristics of the proposed system were investigated using three types of superparamagnetic nanoparticles: CoFe2O4, CoFe2O4@Fe3O4, and Fe3O4 with different average particle sizes and magnetic saturation. Coupled with the lock-in measurements, the limit of detection (LOD) for the Fe3O4 nanoparticles was 15.5 μg/mL. The limits of detection for CoFe2O4 and CoFe2O4@Fe3O4 were 74 μg/mL and 96.5 μg/mL, respectively. The results show that Fe3O4 is suitable for this system for the separation and quantification of biomarkers in diagnostics. [ABSTRACT FROM AUTHOR]

This research proposes a new Big File Cloud (BFC) with its architecture and algorithms to solve difficult problems of cloud-based storage using the advantages of key-value stores. There are many problems when designing an efficient storage engine for cloud-based storage systems with strict requirements such as big-file processing, lightweight meta-data, low latency, parallel I/O, deduplication, distributed, high scalability. Keyvalue stores have many advantages and outperform traditional relational database in storing data for heavy load systems. This paper contributes a low-complicated, fixed-size meta-data design, which supports fast and highly-concurrent, distributed file I/O, several algorithms for resumable upload, download and simple data deduplication method for static data. This research applies the advantages of ZDB - an in-house key-value store which was optimized with auto-increment integer keys for solving big-file storage problems efficiently. The results can be used for building scalable distributed data cloud storage that support big-files with sizes up to several terabytes.

In a Wireless Rechargeable Sensor Network (WRSN), a mobile charger (MC) moves and supplies energy for sensor nodes to maintain the network operation. Hence, optimizing the charging schedule of MC is essential to maximize the network lifetime inWRSNs. The existing works only target the local optimization of network lifetime limited to MC’s subsequent charging round. The network lifetime has been normally reflected in a different metric that is not directly related to the final charging round period. To the best of our knowledge, this work is the first to address the global maximization of network lifetime in WRSNs, which optimizes not only the subsequent charging round but all charging rounds over the entire network lifetime. Another uniqueness is the joint consideration of both the charging path and charging time optimization problems. As a solution, we propose a genetic algorithm (GA)-based global optimization scheme that considers all the possible charging rounds. The GA has a novel mutation operation that mutates gene sizes for representing charging schedules with a varying number of charging rounds. The experiment results show that our algorithm can extend the network lifetime by 35.1 times on average and 38.6 times in the best case compared to existing ones. [ABSTRACT FROM AUTHOR]

Text in images is one of the most important cues for understanding a scene. In this paper, we propose a novel approach based on interest points to localize text in natural scene images. The main ideas of this approach are as follows: first we used interest point detection techniques, which extract the corner points of characters and center points of edge connected components, to select candidate regions. Second, these candidate regions were verified by using tensor voting, which is capable of extracting perceptual structures from noisy data. Finally, area, orientation, and aspect ratio were used to filter out non-text regions. The proposed method was tested on the ICDAR 2003 dataset and images of wine labels. The experiment results show the validity of this approach. [ABSTRACT FROM AUTHOR]

Locating text data in images automatically has been a challenging task. In this approach, we build a three stage system for text detection purpose. This system utilizes tensor voting and Completed Local Binary Pattern (CLBP) to classify text and non-text regions. While tensor voting generates the text line information, which is very useful for localizing candidate text regions, the Nearest Neighbor classifier trained on discriminative features obtained by the CLBP-based operator is used to refine the results. The whole algorithm is implemented in MATLAB and applied to all images of ICDAR 2011 Robust Reading Competition data set. Experiments show the promising performance of this method. [ABSTRACT FROM AUTHOR]