Showing papers by "Sirindhorn International Institute of Technology published in 2018"

Human Identification From Freestyle Walks Using Posture-Based Gait Feature

Abstract: With the increase of terrorist threats around the world, human identification research has become a sought after area of research. Unlike standard biometric recognition techniques, gait recognition is a non-intrusive technique. Both data collection and classification processes can be done without a subject’s cooperation. In this paper, we propose a new model-based gait recognition technique called postured-based gait recognition. It consists of two elements: posture-based features and posture-based classification. Posture-based features are composed of displacements of all joints between current and adjacent frames and center-of-body (CoB) relative coordinates of all joints, where the coordinates of each joint come from its relative position to four joints: hip-center, hip-left, hip-right, and spine joints, from the front forward. The CoB relative coordinate system is a critical part to handle the different observation angle issue. In posture-based classification, postured-based gait features of all frames are considered. The dominant subject becomes a classification result. The postured-based gait recognition technique outperforms the existing techniques in both fixed direction and freestyle walk scenarios, where turning around and changing directions are involved. This suggests that a set of postures and quick movements are sufficient to identify a person. The proposed technique also performs well under the gallery-size test and the cumulative match characteristic test, which implies that the postured-based gait recognition technique is not gallery-size sensitive and is a good potential tool for forensic and surveillance use.

A hidden chaotic attractor in the classical Lorenz system

Abstract: Chaotic attractors in the classical Lorenz system have long been known as self-excited attractors. This paper, for the first time, reveals a novel hidden chaotic attractor in the classical Lorenz system. Either a self-excited or a hidden chaotic attractor is now possible in the classical Lorenz system depending on values of both system parameters and initial conditions. A systematically exhaustive computer search is employed to directly search for the hidden chaotic attractor with elegant values of both system parameters and initial conditions. Time series of trajectories, Lyapunov exponents, and bifurcations of the hidden chaotic attractor are reported. Basins of attraction of individual equilibria are depicted to verify that the hidden chaotic attractor is found. Dynamic regions of attractors are illustrated to reveal seamless connections between self-excited and hidden chaotic attractors in the classical Lorenz system with wide ranges of parameters.

Smart parking using IoT technology

Abstract: One of the main problems in many big and crowded cities is finding parking spaces for vehicles. With IoT technology and mobile applications, in this paper, we propose a design and development of a real smart parking system that can provide more than just information about vacant spaces but also help user to locate the space where the vehicle can be parked in order to reduce traffics in the parking area. Moreover, we use computer vision to detect vehicle plate number in order to monitor the vehicles in the parking area for enhancing security and also to help user find his/her car when he/she forgets where the car is parked. In our system, we also design the payment process using mobile payment in order to reduce time and remove bottleneck of the payment process at the entry/exit gate of the parking area.

Dilation Characteristics of PET FRP–Confined Concrete

Abstract: The improved strength and ductility of fiber-reinforced polymer (FRP) confined concrete is a result of external confinement, which places an effective constraint on its lateral dilation. Fo...

Highly sensitive acetone sensors based on flame-spray-made La2O3-doped SnO2 nanoparticulate thick films

Abstract: In the present work, flame-spray-made La2O3-doped SnO2 nanoparticles with 0.1−2 wt% La contents were systematically studied for acetone detection. The particle and sensing film properties were characterized by X-ray diffraction, nitrogen adsorption, electron microscopy, energy dispersive X-ray spectroscopy and X-ray photoelectron spectroscopy. The sensing films were tested towards 0.1–400 ppm acetone at operating temperatures ranging from 150 to 400 °C in dry air. Gas-sensing results demonstrated that the SnO2 sensing film with the optimal La content of 0.5 wt% exhibited a very high response of 3626 toward 400 ppm acetone with a short response time of 2.8 s at the optimal operating temperature of 350 °C. Moreover, the sensors displayed high acetone selectivity against SO2, H2S, NO2, C6H6, C7H8, C8H10 and CH2O. Therefore, the La2O3-doped SnO2 sensors are promising for sensitive and selective detections of acetone at low concentrations.

A decade of plant-assisted microbial fuel cells: looking back and moving forward

Abstract: Plant microbial fuel cells (PMFCs) have emerged as a renewable source of energy that can produce concurrent bioelectricity and biomass continuously in a clean, sustainable and efficient manner. The...

H2S sensing characteristics of Ni-doped CaCu3Ti4O12 films synthesized by a sol-gel method

Abstract: The problem of detecting the toxic hydrogen sulfide (H2S) gas at part per million levels in air by means of simple solid state gas sensors is relevant to environmental and gas processing industries. In this work, high-performance H2S gas sensors are developed based on nickel-doped calcium copper titanate (Ni-doped CCTO) thin films synthesized by a sol-gel method. From gas-sensing measurements, the response of Ni-doped CCTO sensing films increased substantially with increasing Ni doping level from 1.5 to 7.3 wt%, revealing a catalytic effect of Ni on the surface reactions with adsorbed H2S molecules. In particular, 7.3 wt% Ni-doped CCTO sensors offered a high response of 120 for 10 ppm of H2S at the optimal operating temperature of 250 °C, an order of magnitude higher than that of undoped one. In addition, the response time dropped significantly from ∼80 s to ∼4 s while the recovery time slightly improved as the Ni doping content increased from 0 to 7.3 wt%. Moreover, the Ni-doped CCTO sensors exhibited good reproducibility and high H2S selectivity against ethanol, H2, NO2 and NH3. Therefore, the Ni-doped CCTO sensors are highly promising for sensitive and selective detections of H2S.

Entropy generation in a mixed convection Poiseulle flow of molybdenum disulphide Jeffrey nanofluid

Abstract: Entropy analysis in a mixed convection Poiseulle flow of a Molybdenum Disulphide Jeffrey Nanofluid (MDJN) is presented. Mixed convection is caused due to buoyancy force and external pressure gradient. The problem is formulated in terms of a boundary value problem for a system of partial differential equations. An analytical solution for the velocity and the temperature is obtained using the perturbation technique. Entropy generation has been derived as a function of the velocity and temperature gradients. The solutions are displayed graphically and the relevant importance of the input parameters is discussed. A Jeffrey nanofluid (JN) has been compared with a second grade nanofluid (SGN) and Newtonian nanofluid (NN). It is found that the entropy generation decreases when the temperature increases whereas increasing the Brickman number increases entropy generation.

Bio-electrochemical system for recovery of silver coupled with power generation and wastewater treatment from silver(I) diammine complex

Abstract: In this study, a bio-electrochemical system was used to investigate the recovery of silver from diammine complex [Ag(NH3)2]+, coupled with power generation and wastewater treatment, measured in terms of chemical oxygen demand (COD) removal. Acetate was oxidized by anodic microorganisms to produce electrons, which were transported to the cathode for the reduction of [Ag(NH3)2]+ complex. After 48 h of operation at different initial [Ag(NH3)2]+ concentrations (1000–3000 mg Ag/L), a Ag removal of over 96%, along with a COD removal of 60–76.80%, and a columbic efficiency of 8.55–14.69%, were achieved. Maximum power density slightly increased by 4.66%, from 3618 to 3795 mW/m3, when 10 mM NaNO3 was added to the catholyte containing 2000 mg Ag/L. The diffusion of [Ag(NH3)2]+ ions through the membrane affected the performance, as the inorganic fouling at the membrane surface increased the internal resistance. This transport was confirmed by the accumulation of silver in anodic biomass (0.853 mg Ag/g biomass) after 58 days. Scanning electron microscope (SEM), energy dispersive X-ray (EDX), and X-ray diffraction (XRD) analysis demonstrated Ag deposits with a dendritic structure formed at the cathode surface. A deeper understanding on bio-electrochemical technology used for Ag recovery and electricity production from [Ag(NH3)2]+ complex was achieved in this study.

Taxi Demand Prediction using Ensemble Model Based on RNNs and XGBOOST

Abstract: Taxis play an important role in urban transportation. Understanding the taxi demand in the future gives an opportunity to organize the taxi fleet better. It also reduces the waiting time of passengers and cruising time of taxi drivers. Even, there are some works proposed to predict the demand of taxi but there are few studies that consider the function of areas such as hospital area, department store area, residential area, and tourist attraction. One predictive model may not fit with all types of area. We use a point of interest (POI) to match taxi demand with a place to study the taxi demand in the area with a different function. In this paper, we investigate the best predictive models that can forecast demand of taxi hourly with 7 types of area function. The models that were selected for the experiment are long short term memory (LSTM), gated recurrent unit (GRU) and extreme gradient boosting (XGBOOST). Then, we proposed the ensemble model that can forecast the taxi demand well with all types of area function using the information from those machine learning models. We build the models based on a real-world dataset generated by over 5,000 taxis in Bangkok, Thailand for 4 months. The result shows that the proposed ensemble model can outperform other models in overall.

Developing Smart Farm and Traceability System for Agricultural Products using IoT Technology

Abstract: In the past, most farmers have taken care of their products by relying on basic observation and general knowledge. The quality of agricultural products depends on the farmer’s skill and experience. Therefore, transferring knowledge of planting techniques from one generation to another is not an easy task. In recent years, the smart farm concept has been introduced. However, most current smart farming focuses on monitoring, without utilising the observed data in other ways. Therefore, the aim of this research is to propose a traceability system, summarising and presenting observed data from the smart farm. The Internet of Things (IoT) has been introduced in this research, using several sensors to detect the environmental data in the smart farm. The entire data from the results was computed and presented using a traceability system. Customers are provided with more information support, especially concerning the quality of the planting process, before buying an agricultural product by scanning a quick response (QR) code through a mobile application. This gives the customer greater confidence in the product.

Improving Cell Detachment from Temperature-ResponsivePoly(N‑isopropylacrylamide-co-acrylamide)-Grafted Culture Surfaces by Spin Coating

Abstract: Poly(N-isopropylacrylamide-co-acrylamide) (PNIAM-co-AM)-grafted surfaces have been reported to promote cell adhesion and detachment by a hydrophobic-to-hydrophilic transition triggered by temperatu...

Factors affecting construction performance: exploratory factor analysis

Abstract: The present work attempts to develop a multidimensional performance evaluation framework for a construction company by considering all relevant measures of performance. Based on the previous studies, this study hypothesizes nine key factors, with a total of 57 associated items. The hypothesized factors, with their associated items, are then used to develop questionnaire survey to gather data. The exploratory factor analysis (EFA) was applied to the collected data which gave rise 10 factors with 57 items affecting construction performance. The findings further reveal that the items constituting ten key performance factors (KPIs) namely; 1) Time, 2) Cost, 3) Quality, 4) Safety & Health, 5) Internal Stakeholder, 6) External Stakeholder, 7) Client Satisfaction, 8) Financial Performance, 9) Environment, and 10) Information, Technology & Innovation. The analysis helps to develop multi-dimensional performance evaluation framework for an effective measurement of the construction performance. The 10 key performance factors can be broadly categorized into economic aspect, social aspect, environmental aspect, and technology aspects. It is important to understand a multi-dimension performance evaluation framework by including all key factors affecting the construction performance of a company, so that the management level can effectively plan to implement an effective performance development plan to match with the mission and vision of the company.

Fluidized bed co-combustion of rice husk pellets and moisturized rice husk: The effects of co-combustion methods on gaseous emissions

Abstract: This work explores the potential of three co-combustion methods for reducing NOx in a fluidized-bed combustor. Pelletized rice husk (base fuel) was co-fired with moisturized rice husk (secondary fuel) in this reactor using silica sand as the bed material. Four groups of experiments for (1) conventional combustion of rice husk pellets, (2) co-firing pre-mixed fuels, (3) co-firing using fuel staging with bottom air injection, and (4) co-firing using a reburning method combining fuel staging and air staging, were performed at a 200 kW heat input to the reactor. In the test series, the energy fraction of the secondary fuel in the total fuel supply (EF2) was within 0–0.25, with excess air (EA) varying from 20% to 80% at given EF2. During the reburning tests, the secondary-to-total air ratio (SA/TA) ranged from 0.1 to 0.4, at each EA. The findings revealed that the effects of EF2, EA, and SA/TA on the combustion and emission characteristics of the reactor were substantial. An optimization analysis was performed to determine the optimal EF2, EA, and SA/TA, leading to minimal emission costs of the applied co-firing techniques. Under optimal operating conditions, the combustor ensures high (∼99%) combustion efficiency with minimum emission costs and reduced NO emission: by about 13% when co-firing pre-mixed fuels, by 37% for the fuel-staged co-combustion, and by 53% when using reburning, as compared to 167–176 cm3 m−1 from burning the base fuel alone. However, some increase in the CO and CxHy emissions was observed when using the proposed co-firing techniques.