Tracking and Analysis of Pedestrian’s Behavior in Public Places

Crowd management becomes a global concern due to increased population in urban areas. Better management of pedestrians leads to improved use of public places. Behavior of pedestrian’s is a major factor of crowd management in public places. There are multiple applications available in this area but the challenge is open due to complexity of crowd and depends on the environment. In this paper, we have proposed a new method for pedestrian’s behavior detection. Kalman filter has been used to detect pedestrian’s using movement based approach. Next, we have performed occlusion detection and removal using region shrinking method to isolate occluded humans. Human verification is performed on each human silhouette and wavelet analysis and particle gradient motion are extracted for each silhouettes. Gray Wolf Optimizer (GWO) has been utilized to optimize feature set and then behavior classification has been performed using the Extreme Gradient (XG) Boost classifier. Performance has been evaluated using pedestrian’s data from avenue and UBI-Fight datasets, where both have different environment. The mean achieved accuracies are 91.3% and 85.14% over the Avenue and UBI-Fight datasets, respectively. These results are more accurate as compared to other existing methods.     

Collaborative user modeling with user-generated tags for social recommender systems

With the popularity of social media services, the sheer amount of content is increasing exponentially on the Social Web that leads to attract considerable attention to recommender systems. Recommender systems provide users with recommendations of items suited to their needs. To provide proper recommendations to users, recommender systems require an accurate user model that can reflect a user’s characteristics, preferences and needs. In this study, by leveraging user-generated tags as preference indicators, we propose a new collaborative approach to user modeling that can be exploited to recommender systems. Our approach first discovers relevant and irrelevant topics for users, and then enriches an individual user model with collaboration from other similar users. In order to evaluate the performance of our model, we compare experimental results with a user model based on collaborative filtering approaches and a vector space model. The experimental results have shown the proposed model provides a better representation in user interests and achieves better recommendation results in terms of accuracy and ranking.     

Suggestion Mining from Opinionated Text of Big Social Media Data

Social media data are rapidly increasing and constitute a source of user opinions and tips on a wide range of products and services. The increasing availability of such big data on biased reviews and blogs creates challenges for customers and businesses in reviewing all content in their decision-making process. To overcome this challenge, extracting suggestions from opinionated text is a possible solution. In this study, the characteristics of suggestions are analyzed and a suggestion mining extraction process is presented for classifying suggestive sentences from online customers’ reviews. A classification using a word-embedding approach is used via the XGBoost classifier. The two datasets used in this experiment relate to online hotel reviews and Microsoft Windows App Studio discussion reviews. F1, precision, recall, and accuracy scores are calculated. The results demonstrated that the XGBoost classifier outperforms—with an accuracy of more than 80%. Moreover, the results revealed that suggestion keywords and phrases are the predominant features for suggestion extraction. Thus, this study contributes to knowledge and practice by comparing feature extraction classifiers and identifying XGBoost as a better suggestion mining process for identifying online reviews.     

Weapons Detection for Security and Video Surveillance Using CNN and YOLO-V5s

In recent years, the number of Gun-related incidents has crossed over 250,000 per year and over 85% of the existing 1 billion firearms are in civilian hands, manual monitoring has not proven effective in detecting firearms. which is why an automated weapon detection system is needed. Various automated convolutional neural networks (CNN) weapon detection systems have been proposed in the past to generate good results. However, These techniques have high computation overhead and are slow to provide real-time detection which is essential for the weapon detection system. These models have a high rate of false negatives because they often fail to detect the guns due to the low quality and visibility issues of surveillance videos. This research work aims to minimize the rate of false negatives and false positives in weapon detection while keeping the speed of detection as a key parameter. The proposed framework is based on You Only Look Once (YOLO) and Area of Interest (AOI). Initially, the models take pre-processed frames where the background is removed by the use of the Gaussian blur algorithm. The proposed architecture will be assessed through various performance parameters such as False Negative, False Positive, precision, recall rate, and F1 score. The results of this research work make it clear that due to YOLO-v5s high recall rate and speed of detection are achieved. Speed reached 0.010 s per frame compared to the 0.17 s of the Faster R-CNN. It is promising to be used in the field of security and weapon detection.     

Application of high‐resolution NVD differencing schemes to the FTn finite volume method for radiative heat transfer

In this work, the STEP scheme and several schemes based on the normalized variable diagram (NVD), such as MINMOD, GAMMA, CLAM, NOTABLE, MUSCL, CUBISTA, SMART, WACEB, and VANOS schemes, are evaluated for solving the radiative transfer equation. Two‐dimensional and three‐dimensional rectangular enclosures containing transparent, emitting–absorbing, emitting–absorbing–scattering, or nonhomogeneous participating media are investigated using the modified FTn finite volume method. Although the NVD schemes are much more accurate than the STEP scheme, but they have more time‐consuming and require more iterations. Moreover, most of them often necessitate underrelaxation to ensure convergence. Results show that the MINMOD and GAMMA schemes are still much less accurate than other NVD schemes, but they converge the fastest of the NVD schemes, and do not require underrelaxation. Although the VANOS, WACEB, and SMART schemes give more accurate solutions, they are not competitive with other NVD schemes. However, the CLAM, NOTABLE, and CUBISTA schemes are relatively fast and accurate.     

Evaluation of the FTn Finite Volume Method for Transient Radiative Transfer in Anisotropically Scattering Medium

In this paper, we formulated, applied, and tested the FTn Finite Volume Method (FTn FVM) for transient radiative transfer in three-dimensional absorbing, emitting, and anisotropically scattering medium. Both the STEP and the Curved-Line Advection Method (CLAM) are introduced for spatial discretization of the transient radiative transfer equation. The results show that FTn FVM reduces largely the ray effects and it is more accurate than the standard FVM. Also, using both STEP and CLAM schemes, FTn FVM has smaller convergence time than the standard FVM for all cases. On the contrary, the STEP scheme is faster than the CLAM scheme but it has less accuracy. Then, the effects of optical thickness, scattering albedo, and anisotropy factor on incident radiation and radiative flux are presented and discussed.     

Transient performance of direct steam generation solar power plants

Parabolic trough power plants are currently the most commercial systems for electricity generation. In this study, a transient numerical simulation of a solar power plant was developed by using direct steam generation (DSG) technology. In this system, condensate water from a Rankine cycle is pumped directly to solar parabolic trough collectors. The pressurized water is heated and evaporated before being superheated inside the solar collectors and directed back to the steam turbines, where the Rankine cycle is a reheated‐regenerative cycle. The plant performance with saturated steam production is compared with the performance of a superheated plant. A mathematical model of each system component is presented, with the solar power cycle modeled by the TRNSYS‐17 simulation program. Annual transient performance, including plant power and efficiency, is presented for both plants. As expected, the power of the superheated plant outperforms the saturated plant by approximately 45%, whereas the efficiency decreases by approximately 10%. Furthermore, the power of such plants is considerably improved under the weather of Makkah, 22.4°N, and it is approximately 40 MW for both the spring and autumn seasons. The annual generated energy is approximately 8062 MWh. The levelized electricity cost (LEC) was estimated for both the DSG and the corresponding synthetic oil plants. The DSG plant has an approximately 3% higher LEC than a synthetic oil plant with heat storage and an approximately 11.2% lower LEC than an oil plant if the plant has no storage. Copyright © 2016 John Wiley & Sons, Ltd.     

In vitro blood and fibroblast responses to BisGMA–TEGDMA/bioactive glass composite implants

This in vitro study was designed to evaluate both blood and human gingival fibroblast responses to bisphenol A–glycidyl methacrylate–triethyleneglycol dimethacrylate (BisGMA–TEGDMA)/bioactive glass (BAG) composite, aimed to be used as composite implant abutment surface modifier. Three different types of substrates were investigated: (a) plain polymer (BisGMA 50 wt%–TEGDMA 50 wt%), (b) BAG–composite (50 wt% polymer + 50 wt% fraction of BAG–particles, <50 μm), and (c) plain BAG plates (100 wt% BAG). The blood response, including the blood–clotting ability and platelet adhesion morphology were evaluated. Human gingival fibroblasts were plated and cultured on the experimental substrates for up to 10 days, then the cell proliferation rate was assessed using AlamarBlue assay?. The BAG–composite and plain BAG substrates had a shorter clotting time than plain polymer substrates. Platelet activation and aggregation were most extensive, qualitatively, on BAG–composite. Analysis of the normalized cell proliferation rate on the different surfaces showed some variations throughout the experiment, however, by day 10 the BAG–composite substrate showed the highest (P < 0.001) cell proliferation rate. In conclusion, the presence of exposed BAG–particles enhances fibroblast and blood responses on composite surfaces in vitro.     

Particle screening phenomena in an oblique multi-level tumbling reservoir: a numerical study using discrete element simulation

Due to their wide usage in industrial and technological processes, granular materials have captured great interest in recent research. The related studies are often based on numerical simulations and it is challenging to investigate computational phenomena of granular systems. Particle screening is an essential technology of particle separation in many industrial fields. This paper presents a numerical model for studying the particle screening process using the discrete element method that considers the motion of each particle individually. Dynamical quantities like particle positions, velocities and orientations are tracked at each time step of the simulation. The particular problem of interest is the separation of round shape particles of different sizes using a rotating tumbling vertical cylinder while the particulate material is continuously fed into its interior. This rotating cylinder can be designed as a uniform or stepped multi level obliqued vertical vessel and is considered as a big reservoir for the mixture of particulate material. The finer particles usually fall through the sieve openings while the oversized particles are rebounded and ejected through outlets located around the machine body. Particle–particle and particle–boundary collisions will appear under the tumbling motion of the rotating structure. A penalty method, which employs spring-damper models, will be applied to calculate the normal and frictional forces. As a result of collisions, the particles will dissipate kinetic energy due to the normal and frictional contact losses. The particle distribution, sifting rate of the separated particles and the efficiency of the segregation process have been studied. It is recognized that the screening phenomenon is very sensitive to the machines geometrical parameters, i.e. plate inclinations, shaft eccentricities and aperture sizes in the sieving plates at different levels of the structure. The rotational speed of the machine and the feeding rate of the particles flow have also a great influence on the transportation and segregation rates of the particles. In an attempt to better understand the mechanism of the particle transport between the different layers of the sifting system, different computational studies for achieving optimal operation have been performed.     

Residence Time Distributions in Staged Spouted Beds

A novel design of a multi‐staged, continuous flow spouted bed, suitable for counter‐current, cross‐current, and fed‐batch gas‐solid contacting is developed. Single and three‐stage conter‐flow units are studied, the residence time distributions (RTD's) of the solids therein are measured and two compartmental models are developed to fit this family of RTD's. The first (two‐parameter) model is satisfactory for the single stage units, and three‐stage units at high gas flow rates. The second (three‐parameter) model describes each stage as two CSTVs in parallel with a PFV in series. It provides better agreement with experimental data in the three‐stage case.